First

published in 2018 by Gloucester Publishers Limited, London

Copyright © 2018 Andrew Martin

The right of Andrew Martin to be identified as the author of this work has been
asserted in accordance with the Copyrights, Designs and Patents Act 1988.

All rights reserved. No part of this publication may be reproduced, stored in a

retrieval system or transmitted in any form or by any means, electronic,
electrostatic, magnetic tape, photocopying, recording or otherwise,
without prior permission of the publisher.

British Library Cataloguing-in-Publication Data

A catalogue record for this book is available from the British Library.

ISBN: 978 1 78194 351 9

Distributed in North America by National Book Network,

15200 NBN Way, Blue Ridge Summit, PA 17214. Ph: 717.794.3800.

Distributed in Europe by Central Books Ltd.,

50 Freshwater Road, Chadwell Heath, London , RM8 1RX. Ph +44 (0)20 8525 8800.

All other sales enquiries should be directed to Everyman Chess.

email: info@everymanchess.com; website: www.everymanchess.com

Everyman is the registered trade mark of Random House Inc. and is used in this work under licence from Random
House Inc.

Everyman Chess Series

Chief advisor and commissioning editor: Byron Jacobs

Typeset and edited by First Rank Publishing, Brighton.

Cover design by Horatio Monteverde.
Printed by TJ International Limited, Padstow, Cornwall.
About the Author
Andrew Martin is a FIDE Senior Trainer and International Master. He is the current Head of the newly-formed ECF
Academy, which provides elite training for strong, young players. He teaches in twelve schools, is an experienced chess
writer and has produced numerous chess DVDs.

Also by the Author:

Starting Out: The Sicilian Dragon
First Steps: the Queen’s Gambit
Contents
About the Author
Introduction

1 Strategic Ideas
2 The First Main Line: 3 Nc3/3 Nd2
3 The Advance Variation
4 The Panov-Botvinnik Attack
5 2 c4
6 The Exchange Variation
7 The Two Knights Variation
8 The King’s Indian Attack
9 Odds and Ends

Index of Complete Games

Introduction
The Caro-Kann Defence is one of Black’s most solid and reliable responses to 1 e4. It is introduced simply after the
following moves:
1 e4 c6 2 d4 d5
Black secures a firm foothold in the centre and plans modest development, with the idea
of gradually neutralizing White’s first move initiative. The Caro attracts players of all
levels and is a surprisingly diverse opening, with quite a few different approaches
available for Black, dependent on what White does. In this book I will introduce the
opening and we will take a tour of all the major variations. After that, it is my hope that
you will be ready to make a considered choice about how you want to play the Caro or
indeed play against it!

Andrew Martin, Bramley,

January 2018
Chapter One
Strategic Ideas
1 e4 c6 2 d4 d5

Let us break down the opening into strategic ideas. As White, I am considering:

1) A way to use my advantage in space.

2) Effective development.
3) How to set up an attack.

As Black I am concerned with:

1) Neutralizing White’s advantage of the first move.

2) Light-squared control.
3) Exploiting the varied nature of Caro-Kann pawn structures to fit my style.

Let’s take a look at each of these concepts in turn.

Space Advantage
In this first game White, a strong grandmaster, obtains an advantage in space out of the opening. Eventually he is
able to use this edge to put Black into an uncomfortable position. Black may be able to hold the game but makes a
mistake under pressure.

Game 1
N.De Firmian-M.Morris
Chicago Open 1995

1 e4 c6 2 d4 d5 3 Nc3 dxe4 4 Nxe4 Bf5

 This is the Capablanca Variation, a line that can be recommended to club players. One
of the great benefits of the Caro-Kann is that the way is usually left open for the bishop on
c8 to come out to an active square. Black’s general development in this line is easy to
understand, but he sometimes finds himself with less space, which means he has to defend
accurately.
5 Ng3 Bg6 6 h4 h6 7 Nf3 Nd7
An important defensive move, dissuading Ne5.
8 h5
White gains space on the kingside and sets up a possible advantage in the endgame,
where the pawn on h5 freezes the black kingside structure. Literally thousands of games
have gone this way and it is still an open question as to whether the pawn on h5 is a
strength or a weakness in the white camp. Naturally, Black hopes to attack this pawn as
the game goes on.
8 … Bh7 9 Bd3 Bxd3 10 Qxd3 Ngf6 11 Bf4 Qa5+ 12 Bd2 Qc7 13 0-0-0 e6 14 Ne4 0-
0-0 15 g3 Nxe4 16 Qxe4 Bd6 17 c4
Play has proceeded along the main line and, with 17 c4, De Firmian increases his
advantage in space. The queen is nicely centralized on e4 and, all in all, this is a position
where most strong players would prefer to be White. However, Black is solid, he has
developed comfortably enough, and he should be able to put up a good defence. He starts
by developing counterplay in the centre.
17 … Nf6! 18 Qc2 c5!
An important move, breaking up the centre. De Firmian is allowed to establish a passed pawn, but the pawn will be
isolated and will be a target for Morris to aim at.
19 d5 exd5 20 cxd5 Kb8?!
A bit too meek. 20 … Rhe8 21 Bc3 Ng4! is a good way to approach this position. The
idea is that if White plays 22 Bxg7?? (22 Rhe1 is rather better, and is about equal after 22
… Rxe1 23 Nxe1 f6) he gets his bishop trapped after 22 … f6.
21 Bc3!
Increasing the power of the bishop.
 21 … Rhe8?!
I really do not like this move at all, allowing White to mess up the black pawns. It may
be possible to defend the position to come, but it will not be easy and, most importantly, it
will certainly not be enjoyable.

TIP: If you have to defend, make sure there will be something about your game
that gives you hope. There is nothing worse in chess than struggling to draw in
a prospectless position.

Instead, Black should fall into defensive mode for the time being, with something like
21 … Ne8! (keeping his pawns together) 22 Rhe1 f6 23 Nh4 Qf7 24 Ng6 Rg8. This is not
so bad, and Black continues to eye the pawn on d5.
22 Bxf6
Of course.
22 … gxf6 23 Rhe1 Rxe1 24 Rxe1 Qd7 25 Qe4 f5 26 Qh4 Bf8 27 Qf4+ Ka8 28 Rd1
Bg7 29 Qc4 Qd6
Maybe Morris wants to put his queen on f6 and to threaten mate. De Firmian stops the
plan dead in its tracks.
30 Rd2! Re8 31 Re2 Rd8
I suppose Black could trade and fall back with 31 … Rxe2 32 Qxe2 Qd7, but it will be
a long, hard struggle to draw, a task impossible to relish.
 32 Nh4 b5
Black frees his back rank with this move and activates the queenside pawn majority. As
stated, the overall position might well be okay for Morris, but it is certainly not nice to
play.
33 Qb3

33 … Qd7?
The losing mistake, most likely a product of dissatisfaction. 33 … Bf6 34 Nxf5 Qxd5
35 Qxd5+ Rxd5 36 g4 Bg5+ 37 Kc2 Kb7 was the correct way to proceed, when Black
may hold a draw.
34 Qf3! Bf6
So that if 34 … Qxd5? then 35 Re8 is crushing.
35 d6+ Kb8
35 … Qb7 36 Qxb7+ Kxb7 37 Nxf5 should win for White.
36 Nxf5
White has won a crucial pawn and the knight on f5 dominates the position. It’s all over.
36 … Re8 37 g4
37 Rxe8+ Qxe8 38 Qe3 was more accurate.
37 … c4 38 Kb1 c3?
 38 … Re6 was better.
39 bxc3 Rxe2 40 Qxe2 Qc6 41 Qd3 1-0

OBSERVATION: A lack of space is sometimes a problem for Black in the Caro. As long as Black understands the
danger, he should be able to fight against it.

Effective Development
The first move happens to mean something and White often gets a lead in development playing against the Caro-Kann. It
takes Black a while to set up his solid formation. However, to use the small advantage of the first move to good effect is
a matter of consummate skill. In the coming game, grandmaster Gawain Jones gets a grip with White that Susan Lalic is
unable to shake off.

Game 2
G.Jones-S.Lalic
British Championship, Sheffield 2011

1 e4 c6 2 d4 d5 3 e5
Our first sight of the Advance Variation, a line which has become very popular in the
last decade and infused with original ideas.
3 … c5!?
Botvinnik’s favourite, but that was 60 years ago. Among grandmasters Keith Arkell
regularly plays 3 … c5, but the natural 3 … Bf5 is preferred more often. After 3 … c5,
Black plans easy queenside development with … Nc6 and … Bf5 or … Bg4, but White
can interrupt this plan.


4 dxc5!
If it wasn’t for 4 dxc5, I believe that 3 … c5 would become the main line.
4 … Nc6
4 … e6 is also possible and we will examine that later.
5 Nf3 Bg4
Black is already thought to have a good position in some theoretical works, but Jones
shows that assessment to be superficial.
6 c3 e6 7 b4
This is the problem. White establishes a queenside pawn majority and sets a little trap.
7 … Bxf3
Taking the e-pawn immediately with 7 … Nxe5? is a serious mistake, as after 8 Qa4+!
Black is already in a horrible mess; for example, 8 … Nd7 9 Ne5 Ngf6 10 Bb5 or 8 …
Nc6 9 b5 Nb8 10 Qxg4 and White wins a piece.
8 Qxf3 Nxe5 9 Bb5+

Momentum is gained by this simple check and, over the next few moves, Jones
develops a strong initiative.
9 … Nc6 10 Nd2 Nf6 11 Nb3 Be7 12 Nd4 Qd7 13 Qe2 0-0
Plans for the middlegame are set. White must try to advance the queenside pawn
majority and somehow use the bishop pair to increase pressure. Black has to prepare …
e6-e5, without which her position will remain permanently passive. I think this an
uncomfortable situation for Lalic.
14 0-0 a5!
The only real try for counterplay.
15 Bb2 Rfe8 16 a3 Bf8
16 … Bd8 17 Nf3 Bc7 was an alternative possibility, although it must have been a relief to get the black queen out of
the pin.
17 Nf3 Qc7 18 c4!
Jones has to try this to maintain his advantage.
18 … axb4 19 axb4 dxc4 20 Bxf6
After some thought. The straightforward 20 Qxc4 is met by 20 … Nd5!, so White
figures he has to weaken the black kingside in order to make progress.
20 … gxf6

Throughout the game, my computer engines have been explaining to me that Black is
okay. That may be so if you are rated 3200, but to a normal player Black faces obvious
problems:
1) The kingside structure is weakened.
2) White retains a queenside pawn majority.
3) The opposite-coloured bishops help the player with the initiative; in this case,
Gawain Jones.
21 Rac1
21 Qxc4 could provoke some tactics after 21 … Rxa1 (of course, Black can settle for
21 … f5) 22 Rxa1 Ne5!? 23 Nxe5 Qxe5 24 Rd1 Rb8, although it is doubtful whether the
position is totally satisfactory for Black after 25 Qg4+ Bg7 26 Rd7.
21 … Bg7 22 Qxc4 Red8 23 Bxc6
White trades in one of his advantages for what he hopes is a favourable pawn structure.
23 … bxc6 24 Rfd1 f5
24 … Rd5 is a sturdy-looking move, but there is no doubt that after 25 Rxd5! exd5 26
Qd3 Qf4 27 Rb1 White keeps the advantage.
25 g3 Ra4 26 Qh4 Rb8 27 Rc4
Well played by both sides over the last few moves but, with the black king not entirely
safe, this remains a tough position for Lalic to handle.
27 … Ra2 28 Rd6 Qa7?
Black has to play 28 … h6! and brave the attack after 29 g4!. Instead, under pressure,
Lalic commits a decisive mistake.
29 Rd8+ Rxd8 30 Qxd8+ Bf8 31 Ne5 Qb7 32 Qe8

Once the black queen left the centre the position deteriorated rapidly. Now it’s dead
lost.
32 … Ra8 33 Qxc6 Qa7 34 Kg2 Rd8 35 b5 Rd5 36 b6 Qa2 37 b7 Rd2 38 Qf3 1-0
White kept putting pressure on throughout the game and, although Susan put up a great
defence, a mistake eventually occurred. This is what White wants when he plays against
the Caro and we learn from this single game that Black must choose variations with care.

Setting up an Attack
Setting up an attack against the solid Caro is not just a dream, and White is sometimes
able to achieve this. Observe the following game, where the great Bent Larsen gets
pulverized, although he does take some liberties.

Game 3
A.Beliavsky-B.Larsen
Tilburg 1981

1 e4 c6 2 d4 d5 3 Nc3 dxe4 4 Nxe4 Bf5

As recommended in Game 1, this is a very good choice for players of all levels. The
lines with 4 … Bf5 are recognized as among the most solid of the entire Caro-Kann
complex, although not as Larsen interprets them here.
5 Ng3 Bg6 6 h4 h6 7 Nf3 Nd7 8 h5 Bh7 9 Bd3 Bxd3 10 Qxd3 Ngf6 11 Bf4
White has more space and is ahead in development. It stands to reason that Black must
be careful, or an attack can develop very quickly indeed.
11 … e6
11 … Qa5+! is a very useful move here and has to be preferred to 11 … e6. Black’s
ideas are:
1) If White plays c2-c3, castling long will be less attractive.
2) If White plays Qd2, the queens come off and there is less chance that Black will be
attacked.
3) If White plays Bd2, he is going backwards. The main line continues 12 Bd2 Qc7! 13
0-0-0 e6 14 Ne4 0-0-0 and Black is in a somewhat safer position than he achieves in the
game.
12 0-0-0 Be7 13 Ne5 a5!? 14 Rhe1 a4?!

This is really asking for it. Now with the black king still in the middle and a
considerable lead in development for White, Beliavsky scents blood. 14 … 0-0 should
have been preferred.
15 Ng6!!
Combinational ideas rarely happen at random. White spots the way to launch a direct
attack and, from now on, there is not really any advice that can be offered to Larsen.
15 … Nd5
If he captures, Black ends up almost paralysed after 15 … fxg6 16 Qxg6+ Kf8 17 Rxe6
Qe8 18 Nf5 Qxg6 19 hxg6 and now there is no move. For example, 19 … Bb4
(alternatively, 19 … Re8 20 Nxe7 Rxe7 21 Bd6 or 19 … Nd5 20 c4! Nxf4 21 Rxe7 Nxg6
22 Rxd7 Rb8 23 Re1 Rh7 24 Re6 Nf4 25 Ree7) 20 a3 Ba5 21 Bd6+ Kg8 22 Ne7+ Kf8 23
Nxc6+, winning.

16 Nf5!
White can also take the rook: 16 Nxh8 Nxf4 17 Qf3 Bg5 18 Kb1 Nf6 19 Nf5 with a
clear advantage.
16 … Bf8 17 Bd6
Not 17 Nxh8, when 17 … Nxf4 18 Qe4 Qg5 gives Black a chance to recuperate.
17 … Rg8 18 c4
Every move is powerful, keeps the attack going, and allows no respite.
18 … Nb4
18 … N5b6 19 d5 Nc5 20 Qg3 is winning, as is 18 … N5f6 19 Be7 Bxe7 20 Ngxe7.
19 Qh3
19 Qf3 is also decisive, as the following lines demonstrate:
1) 19 … Qg5+ 20 Kb1.
2) 19 … Nxa2+ 20 Kb1 Nb4 21 d5.
3) 19 … Ra5 20 Nxf8! Qg5+ (or 20 … Nxf8 21 Bxb4) 21 Ne3 Nxa2+ 22 Kb1 Nxf8 23
Kxa2 and wins.
19 … fxg6
Or 19 … Qa5 20 Nxf8 Nxf8 21 Re5 (21 a3 Na6 22 Bf4 Nc7 23 Qg4 wins) 21 … Nxa2+
22 Kb1 Qb6 23 Bc5 Qa6 24 Nd6+ Kd7 25 d5!, again breaking through.
20 Rxe6+ Kf7 21 hxg6+!
 A nice final touch.
21 … Kxe6 22 Re1+ Ne5 23 Bxe5 1-0
A brilliant game by Beliavsky, although he was given considerable help.

TIP: We have seen in this brief introduction that a sense of danger is important if you wish to play the Caro-Kann.
Black must understand the possibilities available to White and respect them.

Neutralizing White’s Advantage

So far we have seen that, when White crashes through, it is largely due to ill-considered choices or simply bad play by
Black. There should be a way to be able to gradually neutralize the advantage of the first move with care in all of the
main lines.

Game 4
S.Maze-T.Gharamian
French Championship, Nimes 2014

1 e4 c6 2 d4 d5 3 Nc3 dxe4 4 Nxe4 Nd7

This is the Smyslov Variation. Black argues that the knight on e4 is the strongest piece
on the board right now and wants to exchange it off on the next move by playing 5 …
Ngf6. Had Black played 4 … Nf6, he would inherit doubled pawns after 5 Nxf6+.
The line with 4 … Nd7 is very solid and has managed to resist most of White’s direct
attacking tries. White can play 5 Ng5, 5 Bd3 or 5 Nf3 here, but Maze settles for a quieter
life.
5 Bc4 Ngf6! 6 Nxf6+
More or less giving up his advantage of the first move. Black can develop comfortably
now.
 6 … Nxf6 7 Nf3 Bf5
The light-squared bishop comes out to an active square.
8 0-0 e6 9 a4
White can try 9 c3, but there is nothing wrong with Black’s position at all after 9 …
Be7 10 Bd3 Bg6 (10 … Bxd3 11 Qxd3 0-0 is also fine) 11 Bf4 0-0 with equality.
9 … a5 10 h3 Be7 11 Re1 0-0 12 c3 Rc8
A rock-solid alternative is 12 … h6! 13 Bf4 Nd5 14 Bg3 Bd6 15 Ne5 Qc7 as in N.De
Firmian-L.Yudasin, Philadelphia 1994.
13 Qb3 Qc7!

With 13 … Qc7, Black prevents Bf4 and sets up the idea of the freeing pawn break …
c6-c5. White has absolutely nothing in a position like this.
14 Ne5
Maybe intending Bf4.
14 … Nd5
So Black stops it again.
15 Bd2 Rfd8
The upcoming … c6-c5 is going to activate all of Black’s pieces.
16 Rad1 Bf6 17 Ng4 Be7 18 Ne5 Bf6 19 Bc1?!
Taking the draw was more sensible. There is nothing in the position to suggest that
White is better. Maybe Maze decided to let Black play … c6-c5 and hope to take his
chances after that.
 19 … c5! 20 f4?!
Perhaps the original idea was 20 g4 which sets a little trap, in that 20 … Bg6? 21 Nxg6
hxg6 22 dxc5 Nf4 23 Bf1 is good for White. Problems start after 20 … cxd4! 21 Nxf7
Kxf7 22 Bxd5 (22 gxf5 Qxc4) 22 … exd5 23 gxf5 Qd7 24 Re6 dxc3 25 bxc3 Kg8, when it
is the white king that is more exposed.
The retreat with 20 Nf3 is simply met by 20 … cxd4, attacking the bishop on c4, and
after 21 Bxd5 Rxd5 22 Nxd4 Bxd4 23 Rxd4 Rxd4 24 cxd4 Qd7 White faces a cheerless
ending.
20 … Nb6! 21 d5 Bxe5
21 … Nxc4! is more precise, as can be seen after 22 Qxc4 (22 Nxc4 exd5 23 Nb6 Bc2!)
22 … Bc2 23 Rd2 exd5 24 Rxd5 Rxd5 25 Qxd5 Bxa4.
22 Rxe5 Bc2!

A nice tactic.
23 Qxc2 Nxc4 24 Rh5 g6 25 Rg5 f6
Decoying the rook from defending the pawn on d5.
26 Rg4 Rxd5 27 f5 Rxd1+ 28 Qxd1 Ne5
All that is left is for Maze is to play out a poor position material behind. Recognizing
this, he goes for it, but never really gets any compensation for the exchange.
29 fxg6 Nxg4 30 Qxg4 hxg6 31 Qxg6+
31 Qxe6+ Kg7 32 Bf4 Qd8 defends comfortably.
31 … Qg7 32 Qh5 Qg3 33 Qe2 Kf7 34 Qh5+ Kg8 35 Qd1
 After 35 Qe2 perhaps Black would have found 35 … Rc6 this time round.
35 … Rf8 36 Bh6 Rf7 37 Qe2 Re7 38 Be3 Qe5 39 Qd3 Kg7 40 Bf2 b6 41 Qd8 Qc7
White is the exchange down. Winning takes some time because Black is having to cope
with an exposed king, but Gharamian gets there in the end.
42 Qd1 Qf4 43 Qd8 Kf7 44 h4
Or 44 Qh8 Qf5.
44 … Qc7 45 Qd3 Qf4 46 Qd8 e5
Time for the passed pawn to roll.
47 Qd5+ Kg7 48 Qd6 Rf7 49 Qe6 e4 50 c4 Qe5 51 Qh3
Or 51 Qxb6 e3 52 Be1 Rd7.
51 … f5 52 Be1 Kh7 53 h5
After 53 Bc3 seems right, but then 53 … Qf4 wins.
53 … f4 54 Bc3 Qf5 55 Qh4 Rd7 56 Kf2 f3 57 g3 e3+ 0-1
A game where Black neutralized White’s pressure very cleanly, setting the scene for the
ambitious 19 Bc1. Maze ended up trying to make something out of nothing and failed.

TIP: If you are content to play solidly and patiently in the opening, players at lower levels become impatient and gift
the game to you with silly mistakes.

Light-Square Control
Light-square control and consequent pressure is one of the most important themes of the Caro-Kann. White must be
careful not to fall into the positional grip that Black is trying to set up. The coming game is one of the most famous Caro
masterpieces and you are advised to play through it several times.

Game 5
A.Nimzowitsch-J.R.Capablanca
New York 1927

1 e4 c6 2 d4 d5 3 e5 Bf5 4 Bd3 Bxd3 5 Qxd3 e6 6 Nc3 Qb6 7 Nge2 c5!




This is the way to put pressure on the white centre. Black can also play 7 … Qa6,
which is very safe.
8 dxc5 Bxc5 9 0-0 Ne7 10 Na4 Qc6 11 Nxc5 Qxc5 12 Be3 Qc7 13 f4 Nf5
I am not sure I like this position for White at all, and the main reason is the subsequent
inflexibility of his game. All he can really do is to try to control the d4-square and hope
that this is enough to hold the position, but it is a cheerless task. Nimzowitsch fails to
understand that his best course is to do nothing and do it well.
14 c3 Nc6 15 Rad1 g6 16 g4?
White has just got to sit there with 16 Bf2.
16 … Nxe3 17 Qxe3 h5!
Puncturing the pawn chain. Pushing on with g4-g5 gives Black total light-square
control, but Nimzowitsch cannot see a decent alternative.
 18 g5
18 h3 hxg4 19 hxg4 Qb6! 20 Qxb6 axb6 21 Nc1 g5! is the key idea for Black,
destroying the white centre. Meanwhile, 18 gxh5 Rxh5 19 Nd4 goes for more exchanges,
but after 19 … Qb6! 20 Rd2 0-0-0 Black doubles on the h-file and the white pawn
structure is ghastly.
18 … 0-0 19 Nd4 Qb6
Much manoeuvring now follows, with White running on the spot and Black making
slow but sure improvements to his position. It’s not completely lost for Nimzo, just
extremely unpleasant.
20 Rf2 Rfc8 21 a3 Rc7 22 Rd3 Na5 23 Re2 Re8 24 Kg2 Nc6 25 Red2 Rec8 26 Re2
Ne7 27 Red2 Rc4 28 Qh3 Kg7 29 Rf2 a5 30 Re2 Nf5! 31 Nxf5+ gxf5 32 Qf3
32 Qxh5 Rh8 33 Qf3 Rh4 is an important variation, when play continues 34 Rf2 Rcxf4
35 Qxf4 Rxf4 36 Rxf4 Qxb2+ and all the white pawns drop off.
32 … Kg6
The king is very useful as a defender.
33 Red2 Re4 34 Rd4 Rc4 35 Qf2?!
35 Rxc4 Rxc4 36 Qf2 was a better defence. Black still holds the advantage after 36 …
Qxf2+ 37 Rxf2 d4 but is not yet winning.
35 … Qb5 36 Kg3 Rcxd4 37 cxd4
If 37 Rxd4 then 37 … Re2 and the game is over.
37 … Qc4 38 Kg2 b5
 Throughout the game Capablanca is in total command of virtually every light square on
the board.
39 Kg1 b4 40 axb4 axb4 41 Kg2 Qc1
It is getting very difficult to find moves for Nimzo now.
42 Kg3 Qh1 43 Rd3 Re1 44 Rf3 Rd1 45 b3 Rc1

Beautifully played. Zugzwang!

46 Re3
Or 46 Qe2 Rc3! 47 Rxc3 bxc3 48 Qc2 h4+ 49 Kxh4 Qf3!.
46 … Rf1 0-1
If you are going to block the position in the Caro-Kann, you have to guard against
rigidity. Basically, after 3 e5, White is obliged to play sharply and advised to avoid too
many exchanges. Otherwise, as here, White can be completely encircled.

The next game is a slight affair, coming as it does from the 2016 World Blitz
Championship. It nevertheless offers a further glimpse of the type of light-square control
that Black can establish in the Caro-Kann right out of the opening.

Game 6
Lu Shanglei-S.Mamedyarov
World Blitz Championship, Doha 2016

1 e4 c6 2 Nc3 d5 3 Nf3




This is the Two Knights Variation. Black is now happy to cede the two bishops in order
to develop freely and quickly.
3 … Bg4 4 h3 Bxf3
4 … Bh5 5 exd5 cxd5 6 g4 Bg6 7 Bb5+ Nc6 8 Ne5 is another story altogether – a line
that we will cover later on.
5 Qxf3 e6
5 … Nf6 at once is also possible.
6 d4 dxe4
6 … Nf6! is slightly more accurate. Black would not be worried by 7 e5 Nfd7 as the
blocked central position favours the knights.
7 Qxe4 Nf6 8 Qd3 Na6 9 a3 Qa5

Black’s pawns on c6 and e6 form what is known as the small centre. They not only
control important central squares, but also act to restrain White from attacking directly.
White has two bishops, but Black has no problem pieces and can take aim at the d4-pawn.

OBSERVATION: Small centre positions have to be handled carefully by the defender. They tend towards passivity.

10 Rb1
Preventing … Bb4 or … Nb4.
10 … Nc7 11 b4 Qf5 12 Qxf5 exf5 13 Bd3 Ne6 14 Ne2
I think he should have played 14 Bxf5 Nxd4 15 Bd3 Bd6 16 0-0 with a nice, open
position for the two bishops.
14 … g6 15 c4 0-0-0
Black is okay. He has active play and his light-square pawns put a temporary wall in
front of the bishop on d3. Meanwhile, as expected, there is some pressure against d4.
16 Be3 Bg7 17 Kd2?!
Lu should have taken the decision to break up Black’s positional grip with 17 d5! cxd5
18 Bxa7. Of course, this is a blitz game against one of the best players in the world and the
heat is on to make a quick move.
17 … Ne4+ 18 Bxe4 fxe4 19 Kc3 f5 20 Rbd1 g5
20 … Rd7 21 Rd2 Rhd8 22 Rhd1 g5 may have intensified the pressure even more.
 21 f4 exf3 22 gxf3 Rhe8

Simple and straightforward. White has unprotected pieces on e3 and e2 and so Black
sets up tactical threats.
23 Bc1 b6 24 Kb3 f4 25 h4 g4!
The decisive strike comes on the light squares.
26 fxg4 f3 27 Nf4 Nxd4+ 28 Ka2 Re4 29 Rhf1 Be5 30 Nd3 Bg3 31 Bg5 Rd7 32 c5 Re2+ 33 Kb1 f2 34 cxb6 axb6 35
Nf4 Re1 36 Nh5 0-1

Varying Pawn Structures

One might get the impression after two moves that Black is concentrating on the light
squares alone, but nothing could be further from the truth! Both White and Black can
fashion different pawn structures, from which fascinating middlegames can arise.

Game 7
D.Eggleston-N.Short
British Championship, Sheffield 2011

1 e4 c6 2 d4 d5 3 Nd2 dxe4 4 Nxe4 Nf6 5 Nxf6+ gxf6


This is almost certainly the sharpest line in the second player’s armoury. Throwing
caution to the winds, Black splits up his pawn structure, seeking aggressive, active play.
Nimzowitsch was the first to recognize the potential of the black set-up some hundred
years ago and to this day it remains a good way to play for the win. Of course, it is risky.
In our introductory game we see Nigel Short at work in the first round of the 2011 British
Championship. Short is always happy to play sharp, unorthodox systems in the search for
victory.
6 c3 h5!?
Black doesn’t yet know the right square for the bishop on c8 and so makes an attempt
to turn the potentially weak pawn on h7 into a strength. We are already experiencing the
unusual nature of this system. A pretty good alternative is 6 … Bf5.
7 Bc4
White has plenty of freedom to choose a development plan and 7 Bf4 is a perfectly
good alternative to 7 Bc4. An earlier Short game continued 7 … Nd7 8 Nf3 h4 9 h3 Nb6
10 c4 Bf5 11 Qb3 Bh6 12 Bxh6 Rxh6 13 Bd3 Bxd3 14 Qxd3 Qc7 (S.Grover-N.Short,
Gauteng 2011) and now Grover should have settled for 15 0-0 0-0-0 16 Rfe1 with a slight
advantage.
7 … Nd7 8 Ne2 Nb6 9 Bb3 Bf5 10 Nf4
10 Bf4 h4 11 Qd2 Nd5 shows just how quickly the advantage can pass in a position like
this when White has to cede the bishop pair on unfavourable terms. Meanwhile, 10 Ng3
Bg6 11 Qf3 Qd7 12 h3 0-0-0 13 a4 leads to a similarly unbalanced position.
10 … Qd7 11 h3 h4 12 Qf3 0-0-0 13 Be3 e5
 It looks as if this thrust solves any problems, but there might be some difficulties with
the pawn structure.
14 Nh5 e4 15 Qe2 Qe7 16 0-0-0 Nd5 17 Nf4 Nxe3 18 fxe3 Kb8 19 Rhf1

Which is the stronger minor piece: the bishop on f8 or the knight on f4? White looks a
bit better here, with the easy plan of doubling rooks on the f-file, although Black will
always have plenty of counterplay with the plan of … Rg8-g3.
19 … Bc8 20 Kb1 f5 21 a3
21 Qc4! was a definite alternative, but the positions arising are hardly clear. Play might
continue 21 … Rd7 (also possible is 21 … Rh7 22 d5 c5 23 Rd2 Rg7) 22 d5! c5 23 d6
Rxd6 24 Rxd6 Qxd6 25 Qxf7 Bh6 26 Rd1 Qf8 27 Qxf8 Rxf8 28 Ng6 Rf6. As played,
White loses the initiative.
21 … Bh6 22 Qd2 Rhg8 23 Rf2 Be6
23 … Rg3 24 Re1 Be6 was also good.
24 c4 c5 25 Nxe6
Eggleston bites, but his position is not that easy anymore. After 25 d5 Bd7 26 Rdf1
Bg7, the dark-squared bishop is very powerful.
25 … fxe6 26 d5 Rg3! 27 Re2 f4
The initiative has now passed fully to Short and the best that Eggleston can do is to go
for an opposite-coloured bishop position a pawn down.
28 d6 Qf8 29 d7
29 Ba4 Rxe3 30 d7 Rd3 is good for Black.
 29 … Rxe3 30 Bc2
Or 30 Rxe3 fxe3 31 Qe2 Qf2.
30 … Rxe2 31 Qxe2 e3 32 Qh5 Bg7 33 Be4 Qf6 34 Qe2 Kc7 35 Qc2 Qe5
The alternative 35 … Rxd7 36 Rxd7+ Kxd7 37 Bxb7 (37 Bf3 Kc8) 37 … Qxb2+! 38
Qxb2 Bxb2 39 Kc2 Bxa3 also looks winning and probably more quickly.
36 Bf3 Rxd7 37 Rxd7+ Kxd7 38 Qa4+ Ke7 39 Qc2 b6

White can keep the game going for a long time from here but with very little hope of
salvaging a draw. Black has the attack and a huge pawn on e3, which makes counterplay
for Eggleston very hard to find.
40 Qe2 Qd4 41 Kc1 Kd6 42 Qc2 Kc7 43 a4 a6 44 Kb1 b5!
To win, Black must develop an attack on the white king.
45 axb5 axb5 46 cxb5 Qb4 47 Be2 Kb6 48 Kc1 Bd4 49 Kb1 c4 50 Kc1 Ka5 51 Kb1
Kxb5 52 b3 Qe1+ 53 Bd1 cxb3 54 Qxb3+ Kc5 55 Qa3+ Kd5 56 Qb3+ Ke5 57 Qb8+
Kf5 58 Qf8+ Bf6 59 Qc5+ Kg6 60 Qc2+ Kh6 61 Qe2 Qb4+ 62 Ka2 Qa5+ 63 Kb3
Qd5+ 64 Kc2 Qa2+ 65 Kd3 Qa6+ 66 Kc2 Qa4+ 67 Kd3 Qd4+ 68 Kc2 f3! 69 Qxf3
69 Qb5 Qc3+ 70 Kb1 e2 71 Bxe2 fxg2 72 Qh5+ Kg7 73 Qg4+ Kf8 wins.
69 … Qc3+ 70 Kb1 Qb2 mate
At last!

The 5 … gxf6 variation has to be studied, and the risks properly evaluated, before any
club player can venture into this labyrinth.

Experimental play in chess should always be applauded. The innovator lays himself
open to ridicule, because new ideas rarely succeed immediately. Mike Surtees, an
eccentric player from England, is always looking for something original. Occasionally he
comes out with a real gem.

Game 8
M.Surtees-J.Houska
British Championship, Liverpool 2008

1 e4 c6 2 Ne2 d5 3 e5 c5 4 b4!?

Already the game has entered the twilight zone, a long way from the normal Caro-Kann
variations. White’s idea has a sound positional basis. By getting rid of the pawn on c5 he
can take over the centre.
4 … cxb4
Off the top of my head, I wonder whether 4 … b6 is a good move?
5 d4 Nc6 6 a3 e6
An earlier Surtees game went 6 … bxa3 7 c3 e6 8 Nf4 Nge7 9 Bxa3 Ng6 10 Nh5 Bd7
11 Bd3 Bxa3 12 Nxa3 0-0 13 0-0 f6 14 exf6 gxf6 15 Qb1! f5 16 Qxb7 (16 Re1!, keeping
up the pressure, may well have been better) 16 … Rb8 17 Qa6 Rb6 18 Qa4 Qg5 19 Qd1 f4
20 Be2 Nh4 21 Bg4 was M.Surtees-K.Arkell, Bradford (rapid) 2002, when 21 … f3! is
very good for Black.
7 axb4 Bxb4+ 8 c3 Ba5?!
 The computer engines show this move as good for Black, but I do not like it at all as it
loses touch with the kingside dark squares. Thus 8 … Be7 9 Ng3 Bd7 10 h4 f5 11 Bd3 a6
12 Nd2 Nh6 13 Nf3 Nf7 (P.Richmond-G.Lee, British League 1999) or even 8 … Bf8 was
better.
9 Ng3 Nge7 10 Nh5!
This is in many ways the winning move and, at this point, I recall that Houska started to
look very uncomfortable. You know if you lose such a game it is probably going to be
published, as White’s opening play is remarkably original.

10 … 0-0 11 Bg5 Qc7 12 Rxa5!! Qxa5 13 Nxg7!

The shock effect continues with a breathtaking double sacrifice, exposing the black
king.
13 … f5
13 … Kxg7 14 Bf6+ Kg8 15 Qg4+! (not 15 Qh5 Rd8 16 Bd3 Kf8 17 Qxh7 Ke8, when
Black escapes) 15 … Ng6 16 Qg5! is the main point. You will not find a defence here for
Black; for example, 16 … Nxd4 (or 16 … Qb6 17 Bd3 Ncxe5 18 dxe5) 17 Bd3! Qa2 18
cxd4 is winning.
14 Nh5 Ng6
14 … Qc7 seems to be a better defence.
15 Nf6+ Kh8
Taking the knight hardly solves Houska’s problems but may at least have enabled some
counterplay. After 15 … Rxf6 16 Bxf6! b5!? 17 Qc1 Rb8 18 h4 b4, the position is not at
all clear.
16 Nxh7!

Surtees is a man on fire.
16 … Qc7
16 … Kxh7 17 Qh5+ Kg7 18 Qh6+ Kf7 19 Qh7+ Ke8 20 Qxg6+ Rf7 21 Be2 Ne7 22
Bxe7 Kxe7 23 Bh5 is just lost for Black.
17 Nxf8 Nxf8 18 Bf6+ Kg8 19 g4
Ripping open the last defences of the black king.
19 … f4 20 Bd3 b5 21 g5 b4 22 Qh5 Ne7 23 g6 1-0
It is very rare to see Houska beaten in this way, as she is such a solid, reliable player.
Hats off to Surtees.

Game 9
J.Ehlvest-J.Timman
Tal Memorial, Riga 1995

1 e4 c6 2 c4 d5 3 exd5 Nf6 4 Nc3 cxd5 5 d4

 Via a slightly unusual move order, this is our first sight of the Panov-Botvinnik Attack.
There will be no closed positions today, as White moves to open up the game without
delay. The Panov is a dangerous attacking weapon.
5 … g6!?
Another one of those moves that shows the versatility of the Caro. We are experiencing
quite a few different pawn structures in this introductory chapter. 5 … g6 is an aggressive
response. For the time being, Black offers a pawn in the interests of speedy development
and positional pressure.
6 Qb3! Bg7
6 … dxc4 7 Bxc4 is poor for Black, so he must give up the pawn on d5.
7 cxd5 0-0 8 Be2 Nbd7 9 Bf3 Nb6 10 Bf4 Bf5
For the pawn Timman has active play but no clear way to regain the material. He would
like to pile up on d5 with … Qd7 and … Rfd8, but the bishop on f5 is a target for White.
In general, White gets better results from this type of position.
11 Rd1 Qd7 12 h3!
 Threatening g2-g4, interrupting the black plan.
12 … h5 13 Nge2
It is perhaps a little too early for 13 Be5, after which Black has counterplay with 13 …
Rfd8! 14 Bxf6 Bxf6 15 Nge2 h4 16 0-0 Rac8 17 Rfe1 Nc4! as in B.Gulko-V.Tukmakov,
Vilnius 1978. Black’s pieces are on good squares and he has the bishop pair. This game
fragment shows us that it is not at all easy to play the white position correctly.
Sophisticated strategy is required and this is why many strong masters continue to use 5
… g6.
13 … Rfd8 14 d6!
An excellent move. Ehlvest gives back the pawn at the right moment, opens up his
bishop on f3, and makes the rook on d8 look misplaced.
14 … exd6 15 0-0 Rac8 16 Ng3!
Very accurate. If instead 16 Rfe1 then 16 … Re8 17 Ng3 Rxe1+ 18 Rxe1 Bd3 allows
Black to keep his light-squared bishop.
16 … Rc4
If 16 … d5 then simply 17 Nxf5 Qxf5 18 Be5 gives White a big edge.
17 Bg5
Ehlvest could also have tried 17 Nxf5 Qxf5 18 Bc1 d5 19 Rfe1 when, once again,
Black is struggling to find active play.
17 … Rdc8 18 Be2
It all looks fine for Timman until we contemplate 18 Be2. The rook on c4 hasn’t really
got anywhere to go …
 18 … Be6
… as we can verify after 18 … R4c7 19 Bxf6 Bxf6 20 Nxf5 Qxf5 21 Nb5!.
19 d5!

Another powerful move. The situation is getting desperate, so Timman tries to mix
things up by sacrificing on h3.
19 … Bxh3
After 19 … Bf5 White simply chops everything off with 20 Bxc4 Nxc4 21 Nxf5 Qxf5
22 Bxf6 Bxf6 23 Qxb7, achieving a completely winning position.
20 Bxc4 Rxc4 21 Bxf6 Bxf6 22 Nce4!
This is an impressive performance by Ehlvest, who ruthlessly stamps out any
counterplay from Timman.
22 … Be5 23 gxh3 Qxh3
23 … h4 24 Ng5! wins, as after 24 … hxg3 25 fxg3 the attack on f7 is fatal; e.g. 25 …
f6 26 Ne6.
24 f4 Bg7 25 Ng5 Qg4 26 Qf3 Bd4+ 27 Rxd4 Rxd4 28 Qe3! 1-0
Accurate to the end. The twin threats of Qe8+ and Qxd4 decide.

The last four games have clearly demonstrated the variety and interest of Caro-Kann
positions, and now it is time now to look at the main lines in a bit more detail.
Chapter Two
The First Main Line: 3 Nc3/3 Nd2
1 e4 c6 2 d4 d5 3 Nc3

3 Nc3 is the most straightforward move for White, bringing out a piece and defending
e4. In order not to concede the centre Black normally captures on e4.
3 … dxe4 4 Nxe4
 The path splits now into:
1) 4 … Bf5, Capablanca’s line which is simple and good.
2) 4 … Nd7, Smyslov’s line, which is solid and respectable, although more difficult to
play than 4 … Bf5.
3) 4 … Nf6, which is complex, allowing doubled pawns after 5 Nxf6+.

The Capablanca Variation

1 e4 c6 2 d4 d5 3 Nc3 dxe4 4 Nxe4 Bf5
4 … Bf5 introduces the Capablanca Variation, possibly the most popular line for Black in
the entire Caro-Kann. Black aims for simple, harmonious, solid development and the
results over time suggest he is able to achieve this. The bishop attacks the well-placed
knight on e4, drives it from the centre, and makes a small gain. White has tried many ways
to make progress, and we will start with the main line.
The first point to be made is that the black position is bomb-proof. There is no way to
take Black apart as his development is strategically impeccable. Let’s examine a recent
game, where Black neutralizes the White initiative completely.

Game 10
L.Lenic-M.Ragger
Austrian Team Championship 2017

1 e4 c6 2 d4 d5 3 Nc3 dxe4 4 Nxe4 Bf5 5 Ng3

 5 … Bg6
Both players withdraw a piece from the centre, but I certainly feel that the bishop on g6
is better placed than the knight on g3. Nevertheless, White has a bit more space at this
early stage.
6 h4
Trying to gain further space and setting up a favourable pawn structure for a variety of
endgames. However, the endgame is a long way off.
6 … h6 7 Nf3 Nd7
It is unwise for Black to allow Nf3-e5. 7 … Nd7 defends against the idea.
8 h5 Bh7 9 Bd3 Bxd3 10 Qxd3 e6




Note the pawn structure on the kingside and, in particular, the ‘freezing’ effect of the
pawn on h5, keeping the pawn on g7 backward. A lot of games are decided in this line
around this one technical point.
Is Black permanently passive on the kingside or can he attack the pawn on h5 as the
game goes on? For the time being let’s just note 10 … e6 and say that Black has
alternatives in the form of 10 … Qc7 and 10 … Ngf6 and that the jury is still out as to
which is the best. Really, the next important decision that Black has to make is to where
the king is going.
11 Bd2 Ngf6 12 0-0-0 Be7 13 Kb1 0-0
A big moment in the game, as Black often plays the queen to c7 and then castles long.
Has Lenic just been given the opportunity to attack the black king? We will see.
14 Ne4
Basically, an admission that the knight on g3 is poorly placed. The pawn advance g4-g5
also now comes into consideration.
14 … Nxe4 15 Qxe4 Nf6 16 Qe2 Qd5!
An excellent centralization of the queen, hitting the pawn on h5 and threatening to take
all the fire out of any attack with the idea of … Qd5-e4. I’d like to say this is all original
stuff, but it’s not. Black’s move has been seen many times.


17 Ne5
After 17 Be3 Qb5 (provoking c2-c4) 18 c4 Qf5+ 19 Ka1, White went on to win in this
recent high-powered game: 19 … Ng4 20 Bc1 Rad8 21 Nh4 Bxh4 (21 … Qxh5 22 f3 Nf6
23 g4 gives White too much momentum) 22 Rxh4 Nf6 23 g4 Qe4 24 Be3 c5 25 f3 Qh7?
(the queen remains out of play here for the rest of the game; 25 … Qc6 was better and
then White presses on with 26 g5 cxd4 27 Bf4, which is rather unclear) 26 dxc5 (the pawn
on c5 is highly useful, as it supports Bd6 in the not too distant future) 26 … Rd7 27 Rhh1
Rfd8 28 Rxd7 Rxd7 29 Rd1 Kf8 30 Bf4 Ke8 31 Bd6 g5 32 a3 Ng8 33 Ka2 Ne7 34 Qe5
Nc6 35 Qf6 Rd8 36 Rd2 a5 37 b3 Rd7 38 Kb2 Ne7 39 b4 axb4 40 axb4 Nc8 41 Kc3 b6 42
c6 1-0, A.Fier-V.Akopian, Dubai 2017.
17 Be3 is clearly a stiff test of this variation.
17 … Qe4 18 Qxe4
18 Be3 Nd5 seems an effective way to play. As long as Black ensures his pieces are not
in the way of advancing white pawns, he should be okay.
18 … Nxe4 19 Be3 Rfd8
It’s already looking rather equal.
20 g4 c5 21 f3 cxd4 22 Bxd4 Bc5 23 Bxc5 Nxc5 24 c3 Kf8 25 Kc2
Incredibly, this is the first new move of the game! However, I don’t think it is going to
win any awards. Black has previously lost this position, although it is hard to believe that
he should. Is the white queenside pawn majority really that dangerous? I.Saric-R.Kreisl,
Zadar 2010, saw 25 b4 Na4 26 Kc2 Ke7 27 Kb3 Nb6 28 a4 Rac8 29 a5 Nd7 30 Nc4 a6 31
Rhe1 Rc6 32 Ne3 Ne5? (a serious error, walking straight into a trick; 32 … Rcc8! 33 Nf5+
Kf8 34 Nd6 Rb8 35 Rd2 Nf6 36 Red1 Rd7 appears to hold the position) 33 Nf5+ exf5 34
Rxe5+ Re6 35 Rxd8 Rxe5 36 Rb8 fxg4 37 Rxb7+ Kd8 38 fxg4 Rg5 39 Rxf7 Rxg4 40 b5
axb5 41 a6 Ra4 42 a7 Kc8 43 Rxg7 1-0.
25 … Ke7 26 b4 Nd7 27 Rxd7+ Rxd7 28 Nxd7 Kxd7 29 Rd1+ Kc7 30 c4 b6 ½-½
With one rook each these endings are quite drawn.
A game that shows it is tough to break Black down after 4 … Bf5.

TIP: In the 4 … Bf5 line, it’s very important to study the endgames that can arise. In particular, the fate of the h5-
pawn is often enough to decide the game.

Since, in this long main line, there are various alternatives for Black at move ten, we
will take them one by one. Each line has its own small, individual differences.

Game 11
G.Quparadze-D.Khamrakulov
Ordu 2017

1 e4 c6 2 d4 d5 3 Nd2 dxe4 4 Nxe4 Bf5 5 Ng3 Bg6 6 h4 h6 7 Nf3 Nd7 8 h5 Bh7 9 Bd3
Bxd3 10 Qxd3 e6

Black keeps some options open with 10 … e6. It was previously thought to be unwise
to allow White to play Bf4, hence 10 … Qc7, but Black seems to be able to deal with 11
Bf4 easily enough.
11 Bf4 Qa5+!
 White usually castles queenside in this variation, so if he blocks with 12 c3, the whole
process is slowed down. The check on a5 is good, interrupting the flow of White’s
development.
12 Bd2 Bb4
Surely, White has to play c2-c3 now or pieces will come off with an easy game for
Black.
13 c3 Be7 14 c4 Qb6 15 0-0-0 Ngf6
Black has achieved what he wanted when he played 10 … e6. He can still castle on
either side and the white king has a bit of fresh air around it.
16 Rhe1

Developing normally and maybe thinking of Nf5 sometime soon, but Black is very
solid.
16 Kb1 is also reasonable, counting on a small advantage in space to give an edge, but
with care Black can equalize, as in the game E.Dervishi-P.Eljanov, Baku Olympiad 2016:
16 … 0-0 17 Ne4 Rad8 18 Bc3 Qa6 19 Nxf6+ Nxf6 20 Ne5 Nd7! 21 Qe2 Nxe5 22 dxe5
b5 (a typical strike in this type of position; for the first time it is Black who can now think
of an advantage, since his bishop is markedly better than the overgrown pawn on c3) 23
cxb5 cxb5 24 f4 Qb7 25 g4 b4 26 Bd2 Rd5 27 Bc1 Rfd8

OBSERVATION: There are dangers to White in the Caro ending, where he finds that all his pawn pushing comes
with a price. As the pieces come off, the black pawn structure is often more compact and flexible.

28 Rxd5 Qxd5 29 Re1 a5 30 Qc2 Bh4 31 Re4 Be7 32 Rc4 Bf8 33 b3 Qf3 34 f5 exf5 35
Qxf5 Qd5 36 g5 hxg5 37 Bxg5 Re8 38 Bf4 a4 (Black is clearly better now, as White is
still labouring with a bad bishop and has numerous pawn weakness to defend) 39 Qe4 Qd7
40 bxa4 (40 Rd4 Qe6 41 bxa4 b3 42 a3 Bxa3 43 Rd3 is a better defensive try, according to
Houdini 5; the cold-blooded computer may enjoy allowing … b4-b3, but I can see why
Dervishi does not) 40 … Qxa4 41 Qd5 Qa7 42 Bd2 Qg1+ 43 Kb2 Qh2 44 h6 gxh6 45 Kc2
Qh5 46 Bxb4 Bxb4 47 Rxb4 Rxe5 48 Qd8+ Kh7 (both kings are open, but Eljanov has an
extra pawn; Dervishi succumbs to the pressure) 49 Qd4?? (49 Qd3+) 49 … Qe2+ 50 Kb1
Qe1+ 51 Kb2 Re2+ 52 Kb3 Qb1+ 53 Kc4 Qxa2+ 54 Kb5 Qe6 55 Qd3+ f5 56 Qd4 Re5+
0-1. A textbook example of how Black may win in this variation. Care in the opening,
patience and technical know-how brought the full point home.
16 … Rd8 17 Qc2 0-0 18 Ne4
White sees no future after 18 Nf5 Bb4 and just goes through with the traditional idea of
exchanging off his poorly placed knight on g3, which in turn makes way for the g-pawn to
advance.
18 … Qa6 19 Kb1 b5
Striking at the light squares. Once the pawn on c4 is gone, Black is in charge of d5, a
key central outpost.
20 Nxf6+ Nxf6 21 Rc1 Qb7 22 Be3
22 cxb5 cxb5 23 Qc7 Rd7 24 Qxb7 Rxb7 is the type of exchanging sequence that
White may find tempting, but which is actually prospectless. The pawn on d4 is a target
and Black can place a knight or a rook on d5 in comfort.
22 … Rb8 23 c5 Rbd8 24 Ne5 Nd5 25 Bd2 Bf6
Compare the bishops and we arrive at another key strategic idea of this line: Black is
always trying to prove that his dark-squared bishop is superior.
26 g4 Ne7 27 Re4 Rd5
Many would have been tempted by 27 … Bxe5 28 dxe5 Rd5, but maybe Khamrakulov
felt that White would have an attack after 29 g5. However, 29 … Qd7! 30 Be3 Nf5 31
gxh6 gxh6 32 Rg1+ Kh8 leaves Black well placed.
28 f4 Rfd8 29 Ba5 Bxe5!





A sharp attempt to win. It is hard to say what prompts a choice like this. Possibly White
was under time pressure and so Black decided to unbalance the game. He could have
settled for 29 … Rb8 30 Qd2 Bxe5, but that is nowhere near as exciting.
30 Bxd8
White bites.
30 … Bxd4 31 Bxe7 Qxe7 32 Qe2 a5
In return for the exchange, Black has strongly placed pieces and pressure against the
white pawns. A very practical choice by Khamrakulov.
33 a4?!
Let’s say White stands still with 33 Qg2. Can Black make progress? He can try 33 …
Qd7, but maybe this is better for White than the game. It is always hard to do nothing.
33 … bxa4 34 Rc4 Bxc5 35 Qc2 Qh4! 36 Qxa4
If 36 Rxc5 then 36 … Qh1+ wins for Black after 37 Ka2 Rxc5 38 Qxc5 Qxe4 39 Qxa5
Qc4+.
36 … Qxg4
The over-extended pawns start to disappear.
 37 Ka2 Bb4 38 Qxc6 Qxh5 39 Qe8+ Kh7 40 Rc7 Rd2 41 Qc6?
Or 41 Qxf7 Qd5+ 42 Rec4 Bc3!.
41 … Qf3 0-1
The killing … Qa3 is coming.

From the previous game, I think we can deduce the following:

1) Endgames can often be good for Black in this variation.
2) White can often be left with a poor dark-squared bishop.
3) If White forms a centre with pawns on c4 and d4, Black must aim to break up this
formation with … b7-b5, trying to gain control of d5.
4) If Black castles short, it is not straightforward for White to launch an attack, since the
knight on g3 gets in the way. Let us see how Sergey Karjakin tries to solve these issues.

Game 12
S.Karjakin-S.Mamedyarov
FIDE Grand Prix, Zug 2013

1 e4 c6 2 d4 d5 3 Nc3 dxe4 4 Nxe4 Bf5 5 Ng3 Bg6 6 h4 h6 7 Nf3 Nd7 8 h5 Bh7 9 Bd3
Bxd3 10 Qxd3 e6 11 Bd2
Karjakin rejects Bf4.
11 … Ngf6 12 0-0-0 Be7 13 Kb1 Qb6!?

I am not at all sure about this move, since it is not clear that the queen is that great on
b6. Importantly, she will have to move again for Black to be able to play … b7-b5.
 Two other plans spring to mind:
1) Black can play 13 … Qc7 and castle long.
2) Black can castle immediately with 13 … 0-0 and play for … c6-c5.
The choice between these ideas is clearly a question of taste.
14 Rhe1 0-0 15 Nf5!
I like this move very much. It solves the problem of the knight on g3 and lines up a
sacrifice on h6, which is what Karjakin had prepared.
15 … Bb4
15 … exf5 16 Rxe7 is clearly unpleasant for Black and so his move is forced.

16 Nxh6+!
Very direct and perhaps a bit surprising, as White is not in any immediate position to
finish Black off.
16 … gxh6 17 c3! Ba3 18 Bc1 Rfe8
To bring the bishop back to f8, so that the king can be defended.
19 g4!
This is the real point of the sacrifice made several moves ago and the idea that elevates
the whole plan with 16 Nxh6+. White wants to open the g-file and get the rooks into the
game.
19 … Nh7 20 c4 Bf8 21 g5 hxg5 22 Nxg5 Nxg5 23 Bxg5 Kh8 24 Rg1
 Karjakin has wasted no time at all and has built up a massive attack. Note that Black’s
queen on b6 is doing precisely nothing.

24 … Qc7
24 … Qa5, trying to get the queen back into play, can be met by 25 c5! Bg7 26 Qf3 f5
27 Qf4, when Black is defenceless; e.g. 27 … Rg8 (27 … Rf8 28 Be7! Rg8 29 Qd6! Nf8
30 h6 Bxh6 31 Bf6+ Bg7 32 Qh2+ Nh7 33 Rh1) 28 h6 Bf6 29 Bxf6+ Nxf6 30 Qe5 and
White is just winning.
25 Qf3 f6 26 Bf4
26 Bxf6+? would allow Black to survive after 26 … Nxf6 27 Qxf6+ Bg7 28 Qg5 Qf7.
26 … Bd6
26 … e5 27 Qg3 is the end of the road.
27 Qg3 Nc5
Mamedyarov cannot face 27 … Rg8 28 Bxd6 Rxg3 29 Bxc7 Rxg1 30 Rxg1 Kh7,
although Black can play on for a while here.
28 Bxd6 Qh7+ 29 Ka1 Ne4 30 Qh4 Rg8 31 f3
Throughout this superb game, Karjakin plays very precise chess.
31 … Ng5
31 … Nxd6 loses to 32 Qxf6+ Rg7 33 h6.
32 Qf4 Qf5 33 Qxf5 exf5 34 Be7 Nxf3 35 Rgf1
35 Bxf6+ Kh7 36 Rxg8 Rxg8 (or 36 … Kxg8 37 d5) 37 d5 was not bad either.
35 … Rg3 36 d5 cxd5 37 cxd5 Kg7 38 Rd3 Kf7 39 Bd6 Nh2 40 Bxg3 1-0
 The knight is trapped on f1; i.e. 40 … Nxf1 41 Bf4.
By delaying castling for only one move, Mamedyarov got himself into all sorts of
trouble, although the way Karjakin played the attack was quite brilliant.

The Caro is one of those defences that, over time, has proved resistant to all of White’s
aggressive ideas. I think it’s because the pawn structure that Black sets up is
fundamentally very sound.

Game 13
Hou Yifan-R.Rapport
FIDE Grand Prix, Sharjah 2017

1 e4 c6 2 d4 d5 3 Nc3 dxe4 4 Nxe4 Bf5 5 Ng3 Bg6 6 h4 h6 7 Nf3 Nd7 8 h5 Bh7 9 Bd3
Bxd3 10 Qxd3 e6 11 Bd2 a5
Various ideas have been tried by Black to spice this line up and 11 … a5 has not been
scoring badly in recent times. Larsen liked to push his flank pawns all the way, but the
modern treatment is to prepare … Bb4.
12 0-0-0 Bb4
Trying to force a concession on the queenside. The notion is that if White moves the c-
pawn forward he (or she) is weakening his (or her) king position. Political correctness is a
must in any modern chess book. Hou Yifan agrees.
13 Ne4 Ngf6

14 c3
 14 Bxb4, aiming for Nd6+, can be met in two successful ways:
1) 14 … axb4! 15 Nd6+ Ke7 16 Nxb7 Qb6 17 Nc5 Nxc5 18 dxc5 Qxc5 19 Qb3 Rhd8
and the very weak pawn on h5 gives Black an undisputed advantage.
2) 14 … Nxe4 is also good; e.g. 15 Be1 (15 Qxe4? axb4 16 Kb1 Qa5) 15 … Nef6 16
Kb1 0-0 is equal.
14 … Nxe4 15 Qxe4 Nf6 16 Qe2 Be7 17 g4
Perhaps White should prevent … Qd5 with the traditional 17 c4, but Black is
comfortable after 17 … a4 18 Kb1 a3 19 b3 Qc7.
17 … Qd5!
Inconvenient, as the queen attacks a2 and threatens … Qe4.
18 c4 Qe4 19 Qxe4 Nxe4 20 Be3 f5!

Rapport is known as an original and creative player and we see this with his choice of
20 … f5. With the game already heading for a draw, Black finds a way to spice it up.
21 gxf5 0-0! 22 Nh4 exf5
White gets a little more active than she does in the game if Black trades on h4; e.g. 22
… Bxh4 23 Rxh4 exf5 24 d5 cxd5 25 Rxd5 Rac8 26 b3 a4 27 Kc2 axb3+ 28 axb3 and
White has an edge, with more active pieces and a better king.
23 Ng6 Rfe8 24 Nxe7+
 The trump card can be played with 24 d5 but Black is safe enough after 24 … Bg5.
Therefore Hou chops the bishop off.
24 … Rxe7 25 Rh4 Nf6 26 Rf4 Nxh5 27 Rxf5 Ng3 28 Rf3 Ne4 29 d5 cxd5 30 Rxd5
This way around, Black doesn’t have a vulnerable pawn on f5 to worry about.
30 … a4 31 Kc2 Rc8 32 b3 axb3+ 33 axb3 Ra8 ½-½
I think we can safely say that 10 … e6 is a tough nut to crack.

Game 14
B.Ivanov-I.Vladimirov
Ruse 2017

1 e4 c6 2 d4 d5 3 Nc3 dxe4 4 Nxe4 Bf5 5 Ng3 Bg6 6 Nf3 Nd7 7 h4 h6 8 h5 Bh7 9 Bd3
Bxd3 10 Qxd3 Qc7
10 … Qc7 is an older try, still popular today and still very reliable. Black prevents Bf4
for the time being and prepares to castle long. I dip down now to a game between lower-
rated players, where White seems to know a lot of theory, but it doesn’t do him much
good.
11 Bd2 Ngf6 12 0-0-0 0-0-0 13 Qe2 e6
The basic, solid set up. Black is thinking about … Bd6 and maybe … c6-c5.

14 Ne5
Attacking the f7-square.
14 … Nb6
 14 … Nxe5 15 dxe5 Nd5 (or 15 … Nd7 16 f4) 16 f4 gives White a slight edge, as his
pawn on e5 cramps Black slightly.
15 Ba5
15 Rh4!? is an interesting move in this position and it’s an idea that Black should
become accustomed to if he wants to place his queen on c7. White activates the rook,
supports Bf4 in certain cases, and intends Rg4 if possible. However, here Black can
simply play 15 … c5! and the rook looks somewhat out of place.
15 … Rd5 16 Bxb6 axb6 17 c4 Rd8 18 f4
18 Ne4 Nxe4 19 Qxe4 Bd6 20 f4 Rhe8 21 Rhe1 ½-½, S.Tiviakov-F.Amonatov, Havana
2008, doesn’t tell us much, apart from how solid the black position is.
18 … c5!

The classic break in this type of position, equalizing immediately.

19 dxc5
19 d5 exd5 20 cxd5 is based on the tactic 20 … Nxd5? (or 20 … Rxd5 21 Rxd5 Nxd5
22 Qc4) 21 Qe4 Nf6 22 Qf5+ Kb8 23 Rxd8+ Qxd8 24 Nxf7, winning; so Black should
play 20 … Kb8! 21 Nf5 Ne8, followed by … Nd6.
19 … Bxc5 20 Ne4 Bd4 21 Rd2
White’s theory runs out and he makes an immediate mistake (21 Kb1 is better). How
many times have I seen that among lower-rated players?

TIP: When you run out of theory at the board, take extra time and care to make a good move. Assess the situation.

 21 … Bxe5 22 fxe5 Nxe4 23 Qxe4 Rxd2 24 Kxd2 Rd8+ 25 Kc3 Qc5
I quite like 25 … b5 now.
26 Rc1 Qf2
White has completely lost the initiative and his wide open king makes his position
extremely uncomfortable.
27 Rc2 Qg3+ 28 Kb4 Qg5

29 c5?
Overambitious. He should try 29 g4! Qe7+ 30 Kb3 Qc5 31 Qf4 and put the onus on
Black to improve his position.
29 … Qe7! 30 Kb5?! Rd5
This would have happened whatever White had played.
31 Kxb6
Let nobody say that he isn’t brave.
31 … Qc7+ 32 Kb5 Rxe5 33 Qf3 Rf5 34 Qe4??
34 Qe2 is more tenacious, but Black is still well on top; e.g. 34 … Qc6+ 35 Kb4 Rf4+
36 Kc3 (or 36 Rc4 Rxc4+ 37 Qxc4 Qxg2, and not 36 Kb3 Qa4+ 37 Kc3 Qb4+ 38 Kd3
Qd4 mate) 36 … Qa4 37 a3 Rg4, winning.
34 … Rf4 0-1
The point at which the well-timed jab … c6-c5 was played is very instructive.

 I mentioned the idea of Rh4 in the last game. It is not really very dangerous. Black can
angle for the exchange of pieces and, with it, easy equality.

Game 15
F.Gheorghiu-V.Hort
Alekhine Memorial, Moscow 1971

1 e4 c6 2 d4 d5 3 Nc3 dxe4 4 Nxe4 Bf5 5 Ng3 Bg6 6 h4 h6 7 Nf3 Nd7 8 h5 Bh7 9 Bd3
Bxd3 10 Qxd3 Qc7 11 Rh4
Preparing Bf4.
11 … e6 12 Bf4 Bd6
Black counters the idea simply enough.
13 Bxd6 Qxd6 14 Ne4 Qe7 15 0-0-0 Ngf6 16 Nxf6+ gxf6!

A good move, originally played by Petrosian. Black avoids any endgames where his
pawn gets pinned down on g7 and he can develop counterplay by placing a rook on g8.
17 Re4
Older top-class games are always worth playing through.

TIP: Do not neglect the classics.

This game is S.Gligoric-T.Petrosian, Bled/Zagreb/Belgrade 1959: 17 Qd2 Nb6 18 Qa5
Qd6 19 Rd3 Qd5 20 Qa3 Nc4 21 Qb4 Nd6 22 b3 a5 23 Qd2 a4 (if anyone is better here, it
is Black; it’s easy to see good squares for the black rooks, and White must loosen his
position further to give his king some air) 24 c4 axb3! 25 axb3 (or 25 cxd5 bxa2 26 Qxa2
Rxa2 27 dxc6 bxc6 28 Rd2 Ra3! 29 Rc2 Rg8) 25 … Qa5 26 Qxa5 Rxa5 27 Nd2 Ke7 28
Nb1 Rg8 29 g3 f5 (29 … Rf5! seems pretty strong; e.g. 30 f4 Ne4 or 30 c5 Nb5 or 30 Rd2
Rgg5 is much better for Black) 30 Nc3 Ra1+ 31 Kd2 Rg4 32 Rxg4 fxg4 33 Ke3 b5 34
cxb5 cxb5 35 Kf4 Rf1 36 Nd1 f5 (Black’s advantage grows; it’s incredible that he lost this
position) 37 d5 Re1? (this was the moment for 37 … Ne4! which is very strong indeed,
since38 d6+ Kd7 39 Ke5 Re1! wins, or if 38 dxe6 Kxe6 39 Ke3 then 39 … b4 40 Ke2
Rxd1! or 40 Kd4 Nxf2 41 Nxf2 Rxf2 42 Kc4 f4 43 gxf4 Rxf4+ 44 Kc5 Kf5, again
winning) 38 Ne3 Re2 39 dxe6 Rxf2+? (compounding his mistake; 39 … Kxe6! was
correct) 40 Ke5 Re2 41 Rc3 Ne8 42 Kxf5 Rf2+ 43 Kg6 Rf6+ 44 Kh7 Rf3 45 Kxh6 and
White is now winning.
17 … 0-0-0 18 Ree1 Qd6 19 c4 Qf4+
The f4-square can often be a focal point for Black in this main line. Always be on the
lookout to see if you can use this square to gain counterplay.
20 Qe3 Qxe3+ 21 Rxe3 Rhg8 22 g3 Rg4 ½-½
Not that interesting, except to show that the teeth of 11 Rh4 can be drawn.

As is often the case in this modern era, the theory of respectable openings has extended
to move 20 and beyond. The main line of the 4 … Bf5 Caro is one of those openings.
Luckily for the common man, most of Black’s moves are very natural.

Game 16
H.Hamdouchi-C.Bauer
French Championship, Nancy 2013

1 e4 c6 2 d4 d5 3 Nd2 dxe4 4 Nxe4 Bf5 5 Ng3 Bg6 6 h4 h6 7 Nf3 Nd7 8 h5 Bh7 9 Bd3
Bxd3 10 Qxd3 Ngf6 11 Bd2 e6 12 0-0-0 Qc7 13 Kb1 0-0-0 14 c4 Bd6 15 Ne4 Nxe4 16
Qxe4 Nf6 17 Qe2 c5 18 Bc3

18 … cxd4
I slightly prefer 18 … Rhe8 here, not wishing to encourage a white knight into the
centre. Black waits, constructively, with a simple rook move. Play can continue as in
R.Ponomariov-E.Postny, European Cup, Kallithea 2008: 19 Ne5 (Black doesn’t fear a
sequence like 19 dxc5 Bxc5 20 Bxf6 gxf6 21 Nd2 as 21 … f5 keeps the knight from
coming to e4) 19 … cxd4 20 Bxd4 Bxe5 21 Bxe5 Qc6 22 g4 Rxd1+ 23 Rxd1 Rd8 24 Re1
(White can try for a marginally better ending with 24 Rxd8+ Kxd8 25 Bxf6+ gxf6 26 f3,
but after 26 … Ke7 the draw is pretty much guaranteed; White holds a three to two
queenside pawn majority for sure, but with only queens left, it really is a Herculean task to
make anything of the minute advantage) 24 … Ne8 25 f4 Qd7 26 Bc3 Qd3+ 27 Qxd3
Rxd3 28 f5 Kd7 29 fxe6+ fxe6 30 Rf1 Ke7 31 Kc2 (31 Bb4+ Kd8 32 Rf7 Rd7 33 Rxd7+
Kxd7 34 Kc2 seems to favour White, but Black has counterplay after 34 … e5 35 Kd3
Ke6 thanks to his passed e-pawn) 31 … Rg3 32 Bb4+ Kd8 33 Rf7 Rxg4 34 Kb3 Kc8 35
Bc5 Rg3+ 36 Ka4 Rg2 37 Kb3 Rg3+ 38 Ka4 Rg2 39 b4 Rxa2+ 40 Kb3 Rg2 41 Re7 Nf6
42 Bxa7 Nxh5 43 b5 Nf6 44 Bd4 Rg4 45 Bb6 h5 46 Rc7+ Kb8 47 Rf7 Kc8 48 Rf8+ with
a draw in 83 moves somewhere in the distance.
19 Nxd4 a6 20 g4 Bf4 21 Nb3 Rxd1+ 22 Rxd1 Rd8 23 Rxd8+ Qxd8 24 f3 Ne8 25
Nc1 Bxc1 26 Kxc1 Qc7 27 b3
 The scene is set for White to initiate a long grind, or so it seems. White again has a
three to two majority on the queenside and his bishop is better than the knight.
Furthermore, Black’s pawn majority in the centre appears restrained. Black can certainly
defend these positions, but he has to realize that it can slowly become bad, unless he takes
action now.
27 … Qc6
27 … b5! was indicated.
28 Qe4 Qxe4 29 fxe4 f6
29 … Kd7 30 Kd2 Kd6 is a little less committal.
30 e5 Kd7 31 Kd2 Ke7 32 Ke3 f5?
A poor move. I doubt White can win after 32 … fxe5 33 Bxe5 Nf6 34 Kf4 Kf7 35 Bc3
Nd7 36 Ke4 Nf6+ 37 Kf3 Nd7.
33 gxf5 exf5 34 Kf4 Ke6 35 Ba5
Suddenly, Black has no good moves.
35 … g5+ 1-0
As the grandmaster probably knows the en passant rule, Bauer resigns. The alternative
35 … b5 fails to 36 c5 Kd5 37 b4 Ke6 38 c6.

This is no place to conduct a forensic examination of the line with 4 … Bf5. It’s
basically a sound variation, which is pretty good for club players to take up. There is a fair
bit to know, but most of the strategic and tactical ideas are very natural and easy to play.
You won’t be facing grandmaster preparation in these matches. At higher levels, Black
must know his stuff, that much is clear. Our final game in this main line features one of his
most promising tries.

Game 17
F.Bentivegna-D.Rombaldoni
Torino 2010

1 d4 d5 2 Nc3 c6 3 e4 dxe4 4 Nxe4 Bf5 5 Ng3 Bg6 6 h4 h6 7 Nf3 Nd7 8 h5 Bh7 9 Bd3
Bxd3 10 Qxd3 e6 11 Bd2 Qc7 12 0-0-0 Ngf6 13 Kb1 c5!

To me, 13 … c5 looks good. Black keeps options open with his king and could well
play … c5-c4 if allowed, in order to gain the d5-square for his knight.
14 c4
If 14 Qe2, then 14 … c4! is possible, with a very unclear and sharp position on the
board after 15 Nf5!? b5.
14 … cxd4 15 Nxd4 a6
Black should definitely stop Nb5.
16 Bc3 Nc5
The graveyards are full of players who leave the king in the centre; e.g. 16 … Ne5? 17
Nxe6! (crushing!) 17 … fxe6 18 Bxe5 Qxe5 19 Qg6+ Ke7 20 Rhe1 1-0, B.Ramazzotti-
J.Caliguire, correspondence 1986.
17 Qc2 Be7 18 Nf3 Ncd7
There is nothing wrong with 18 … 0-0! here. White has yet to improve the position of
the knight on g3 and thus the kingside attack with g4-g5 is stalled. 19 Bxf6 Bxf6 20 Ne4
Nxe4 21 Qxe4 does not help, as after 21 … Rac8 22 b3 b5 23 Rc1 Qc5 Black has the
upper hand.
19 Rhe1
19 Qe2 Rc8 20 Rd4 0-0 21 Rhd1 Rfd8 is perfectly satisfactory for Black.
19 … Qxc4
I prefer 19 … 0-0 and keeping calm after 20 Nf5 with 20 … Bd8 21 Nd6 Be7, which is
equal. 21 … b5!? is more complicated after 22 cxb5 axb5 23 Nxb5 Qc4 24 Na3 Qg4.
20 Nf5 Rc8?
This should have been punished. 20 … Bf8 had to be played, when 21 N5d4 0-0-0 is
rather unclear.

21 Rd4
21 N3d4!! is very strong indeed, probably winning. Black cannot evacuate his king
from the centre in time as the following variations show: 21 … Bf8 (or 21 … exf5 22
Nxf5 0-0 23 Nxe7+) 22 Ne3 Qc5 23 Nxe6! fxe6 24 Qg6+ Ke7 (or 24 … Kd8 25 Bxf6+
gxf6 26 Qxf6+ Kc7 27 Qxh8) 25 Nf5+ Kd8 26 Bxf6+ gxf6 27 Nd4! Rc6 28 Nxc6+ bxc6
29 Qxf6+, winning.
21 … Qb5 22 Nxg7+ Kf8 23 a4?!
Why deflect the black queen back to a square where she defends e6? Instead, both 23
Rxd7 and 23 Nxe6+ would have maintained the White attack. For example:
1) 23 Rxd7 Qxd7 24 Ne5 Qd5 25 Nxe6+ fxe6 26 Ng6+ Ke8 27 Nxh8 with a clear
advantage.
 2) 23 Nxe6+ fxe6 24 Rxd7 Rxc3 (24 … Nxd7 25 Bg7+! Kxg7 26 Qg6+ Kf8 27 Nd4
Rc6 28 Nxb5 wins) 25 Qxc3 Qxd7 26 Ne5 and again White is much better.
23 … Qb6 24 Nxe6+ fxe6 25 Rf4 Rc5 26 Nd4 e5
The black queen turns out to be usefully placed.

27 a5?
White forgets about … Rxc3. He had to try 27 Qf5, which is anything but clear after 27
… Rd5! (or 27 … Rxc3 28 Qxd7 Qxd4 29 Qxd4 exd4 30 bxc3 dxc3 31 Rc4) 28 Ne6+ Kf7
29 Rg4 Rh7 30 Rc4.
27 … Rxc3! 28 Qxc3
28 Qf5 Qd6 29 bxc3 exd4 30 Rxd4 Qc5 wins.
28 … exd4 29 Qc8+ Qd8 30 Qxb7 Nc5 31 Qf3 Kf7
Black finally gets to breathe a little easier. White has shot his bolt and the game comes
to a swift conclusion.
32 Re5 d3 33 Rd5
If 33 Rxf6+ Bxf6 34 Rxc5 then 34 … Re8.
33 … Qxa5 34 b4 Rb8 0-1
Strong nerves appear necessary if one wishes to play 4 … Bf5.

There can be no doubt that White can always get a position with slightly more space
and slightly more active pieces in the 4 … Bf5 variation, if that is what he wishes.
However, the slight advantage often evaporates into thin air.
Game 18
B.Spassky-A.Karpov
Candidates Semi-final (Game 6), Leningrad 1974

1 e4 c6 2 d4 d5 3 Nc3 dxe4 4 Nxe4 Bf5 5 Ng3 Bg6 6 Nf3 Nd7

7 Bd3
Solid chess. Black has no need to capture here, so Karpov continues developing. 7 Bc4
is a little more aggressive and – who knows – there may be a sacrifice on e6 in the offing.
A possible line is 7 … e6 8 0-0 Ngf6 and now:
1) In the game M.Tal-T.Petrosian, Moscow 1961, Tal tried 9 Ng5 but Petrosian was on
the ball: 9 … h6! 10 Nh3 (10 Nxe6? just doesn’t work after 10 … fxe6 11 Re1 Bf7 12
Bxe6 Bxe6 13 Rxe6+ Kf7) 10 … Bd6 11 Nf4 Bxf4 12 Bxf4 Nd5 with equality.
2) If White is given time for Qe2 or Re1 a sacrifice on e6 may well be possible. For
instance, 9 Qe2 Be7 10 Re1 0-0 11 Bb3 Nd5! 12 c4 Nb4 13 a3 Nd3 14 Rd1 Nxc1 15
Raxc1 is an absolutely typical position for this line, where White has a lead in
development and all his pieces in play, but Black is so solid that White can’t do anything
with it. After the key prophylactic move 15 … Re8!, Black can start to think about
activity; e.g. 16 Bc2 (or 16 Ne5 Nxe5 17 dxe5 Qb6) 16 … a5 is equal.
7 … e6
7 … Bxd3 8 Qxd3 develops the white queen unnecessarily.
8 0-0 Ngf6 9 c4 Bd6
 Careful players might choose 9 … Be7 here, which is also perfectly respectable. Play
can continue 10 Re1 0-0 11 Bxg6 hxg6 12 Bf4 Re8 13 Qe2 c5 14 Rad1 cxd4 15 Nxd4 Bb4
16 Bd2 Bxd2 17 Qxd2 a6 18 b4 Qc7 19 Rc1 Rad8 20 Qc3 Nb6 with equality.

TIP: White has a queenside pawn majority, his only trump. It is worth remembering that the more pieces that are
exchanged, the more dangerous this majority becomes. Above all, Black needs to keep the queens on the board and to
keep active. This is a key tip that will help you navigate through many similar positions.

10 b3 0-0 11 Bb2 Qc7
11 … c5 also appears to be fine, as seen in an earlier game from the same match: 12
Bxg6 hxg6 13 Re1 Qc7 14 dxc5 Bxc5 15 Qc2 Rfd8 16 Ne4 Nxe4 17 Qxe4 ½-½,
B.Spassky-A.Karpov, Candidates Semi-final (Game 2), Leningrad 1974.

12 Bxg6
12 Qc2 Rfe8 13 Rfe1 Bf4 14 Ne4 Nxe4 15 Bxe4 Nf6 16 Bxg6 hxg6 17 Rad1 Rad8 ½-
½ was the swift conclusion of R.Byrne-V.Hort, Havana Olympiad 1966.
12 … hxg6 13 Qe2 Rfe8 14 Ne4 Nxe4 15 Qxe4 Be7
Making sure that Ng5 is never possible. Karpov plays it safe, happy to neutralize with
the black pieces and go for it when playing White. That was the way to win matches in
those days.
15 … e5!? is a bit sharper, but I suppose Karpov did not want to encourage the bishop
on b2 to get into the game. Nevertheless, Black is fine after 16 Rae1 (or 16 c5 Be7 17
Nxe5 Nxe5 18 dxe5 Bxc5) 16 … Nf6 17 Qh4 e4 18 Ng5 Bf4 with equality.
 16 Rad1 Rad8 17 Rfe1 Qa5 18 a3 Qf5 19 Qe2 g5 20 h3
20 d5 is thematic once White has played c2-c4. However, it is not dangerous here: 20
… exd5 21 cxd5 (or 21 Nd4 Qh7!, eyeing h2, and if 22 cxd5?! Nf6 23 dxc6?? then 23 …
Bd6! wins) 21 … g4 22 Nd4 Qxd5 23 Qxg4 Nf6.
20 … g4 21 hxg4 Qxg4 22 d5
22 Rd3 is a nice centralization and would probably have led to a draw. After his quiet
opening play it looks like Spassky is waking up, but all the coming activity seems to work
in Black’s favour.
22 … cxd5 23 cxd5 e5

24 d6
The advanced pawn will become the source of White’s problems in the endgame to
come. He should just play 24 Rc1.
24 … Bf6 25 Nd2?
Botvinnik recommended 25 Qb5! e4 26 Nh2 Qe6 27 Bxf6 Nxf6 28 Qxb7 Rd7 29 Qc6
Red8 with equal play.
25 … Qxe2 26 Rxe2 Rc8 27 Ne4 Bd8 28 g4 f6 29 Kg2 Kf7 30 Rc1 Bb6 31 Rec2
Rxc2 32 Rxc2 Ke6
Steady improvement by Karpov has resulted in quite a difficult endgame for Spassky to
defend. The pawn on d6 requires constant attention.
33 a4 a5 34 Ba3 Rb8!
 An excellent move, played with a later … b7-b5 in mind.
35 Rc4 Bd4 36 f4 g6 37 Ng3 exf4
He could try 37 … b5 straight away with the possible continuation 38 f5+ gxf5 39
gxf5+ Kd5 40 Rc7 bxa4 41 Rxd7 axb3 42 Nh5 e4 43 Nf4+ Ke5. True to his style, Karpov
is not willing to give the opponent even the slightest tactical chance.
38 Rxd4 fxg3 39 Kxg3 Rc8 40 Rd3 g5 41 Bb2 b6 42 Bd4 Rc6 43 Bc3 Rc5
Capturing on d6 right away is possible. He probably did not do it because of the
following line: 43 … Rxd6 44 Rxd6+ Kxd6 45 Kg2 Kd5 46 Kf3 Ne5+ 47 Kg3, when a
curious situation has arisen where Black still has to prove a win.
44 Kg2 Rc8 45 Kg3 Ne5 46 Bxe5 fxe5 47 b4 e4 48 Rd4 Ke5 49 Rd1
49 d7 Rd8 50 Rd1 axb4 51 Kf2 Kf4 52 Rd6 b3 looks completely lost for White.
49 … axb4 50 Rb1 Rc3+ 51 Kf2 Rd3 52 d7 Rxd7 53 Rxb4 Rd6 54 Ke3 Rd3+ 55
Ke2 Ra3 0-1
Nice technique at the end.

There are various systems that can be employed to try to take Black out of his comfort
zone. The first one we will examine involves the unexpected advance of the f-pawn.

Game 19
P.Duboue-G.Battaglini
Nimes 2014
1 e4 c6 2 d4 d5 3 Nc3 dxe4 4 Nxe4 Bf5 5 Ng3 Bg6 6 f4!?

Not so bad. White gains space and clamps down on the centre. He can play his knight
into e5 with assurance. Those are good points, but the black position cannot be taken by
storm and there are resultant dark-squared weaknesses for White to guard. Long ago,
Nimzowitsch tried 6 f4 against Capablanca. Then the move was novel but today good
responses are known. It may be that 6 h4 first is a better way to play this line; e.g. 6 h4 h6
7 f4! e6 8 h5 Bh7 9 Nf3 Nd7.
6 … e6
Black can strike out independently with 6 … h5!? 7 f5 (naturally 7 Nf3 is better) 7 …
Bxf5 8 Nxf5 Qa5+. This line is not possible after 6 h4.
7 Nf3
If one only considers the result of the following game, then 7 h4 seems okay. However,
we will see that Karpov goes astray in a good position. This game is S.Mamedyarov-
A.Karpov, World Blitz Championship, Moscow 2009: 7 h4!? h5 8 Nf3 Nd7 9 Bc4 Nh6! (a
very nice move, clamping down on the light squares) 10 Qe2 Nf5 11 Nxf5 Bxf5 12 c3 Be7
13 Be3 (the bishop on e3 is less than impressive and perhaps this is the main drawback of
the whole f2-f4 idea) 13 … Bg4 14 0-0-0 Nf6 15 Rde1 g6 16 Bd2 Qd7 (with more time,
Karpov might have chosen 16 … 0-0 17 Bd3 c5, when Black is a bit better) 17 Qf2 Bxf3
18 Qxf3 a5 19 a4 0-0 20 Qg3 Kg7 21 Re5 Ng4 22 Be2!? (an imaginative exchange
sacrifice, no doubt prompted by time considerations) 22 … Nxe5 23 fxe5 Rh8 (23 … c5!
looks very strong: 24 Bxh5 cxd4 25 Bd1 dxc3 26 Bxc3 b5) 24 Rf1 Raf8 25 Bd3 Qe8 26
c4 Qd8 27 Rf4 b5 (the machine recommends 27 … Rh7!) 28 axb5 cxb5 29 cxb5 Rh6??
(29 … a4) 30 Kb1? (missing 30 Rxf7+! Kxf7 31 Bxh6 Rg8 32 Qf4+ Ke8 33 Be4 or 30 …
Rxf7 31 Bxh6+ Kh8 32 Qxg6 and wins) 30 … Rhh8 31 Be4 a4 32 Bc3 Kh6 33 Qe3 g5 34
Rf6+ 1-0 (presumably on time).
7 … Nd7 8 h4 h6 9 h5 Bh7 10 Bd3 Bxd3 11 Qxd3

11 … Bd6
Black can delay developing his bishop on f8 and prepare … c6-c5. This is another good
approach; for example, 11 … Ngf6! 12 Bd2 Qc7 13 0-0-0 c5 14 dxc5 (14 Ne4 c4 15 Qe2
Nxe4 16 Qxe4 Bd6 17 Kb1 Nf6 18 Qe2 0-0 19 Ne5 Nd5 is equal) 14 … Bxc5 15 Qe2 0-0
16 Ne5 Rac8 17 Kb1 Bd6 18 Bc3 Nb6 19 Bd4 Nbd5 (if I could see that f2-f4 helped the
attack on the kingside in any way I would like this line more; as it is, the pawn on f4 just
looks like a weakness to me) 20 Ne4 Nxe4 21 Qxe4 f5 22 Qf3 Qxc2+ 23 Ka1 Qa4 24 Kb1
Qc2+ 25 Ka1 Nb4 26 Bc3 Bxe5 27 fxe5 Nxa2 28 Rd2 Qa4 0-1 was S.Mamedyarov-
V.Topalov, Nice (blindfold rapid) 2008.
12 Ne4 Ndf6 13 Nxd6+
13 Bd2 Nxe4 14 Qxe4 Nf6 15 Qd3 Rc8! is good.
13 … Qxd6 14 Ne5 Ne7 15 Be3 Nf5
Black can castle on either side and is restraining any pawn storm.
16 0-0-0
If he hits out with 16 g4 then the simple 16 … Nxe3 17 Qxe3 Qb4+! 18 Qc3 Qxc3+ 19
bxc3 is, again, a little better for Black.
16 … Qd5!
Forking a2 and g2. White takes emergency measures.
 17 g4 Nxe3 18 Qxe3 Qxa2 19 c4
White hopes to strand the black queen out of play.
19 … Nd7!
But there is no chance of that. The c4-pawn is simply too vulnerable.
20 Qc3
20 Nxd7 Qxc4+ 21 Kb1 Kxd7 22 f5 exf5 23 Qf4 Rae8 24 Qxf5+ Kc7 demonstrates that
White cannot blast through.
20 … Rd8 21 f5 Nxe5 22 dxe5 0-0!

This is all the better for having been delayed.
23 f6 b5! 24 fxg7 Kxg7 25 Qf3
White cannot improve with 25 Kc2, when the queen is extricated immediately: 25 …
Qa4+ 26 Qb3 (others don’t help: 26 Kc1 bxc4 or 26 b3 Qa2+ 27 Qb2 Qa6 28 cxb5 cxb5
29 Kb1 Qc6) 26 … Qxb3+ 27 Kxb3 bxc4+ 28 Kxc4 Rd5.
25 … Qxc4+ 26 Kb1 Rd5! 27 g5
27 Qf6+ Kh7 28 g5 leads to 28 … Qe4+ 29 Ka2 Qa4+ 30 Kb1 Rxd1+, winning.
27 … hxg5 28 h6+ Kh7 29 Rdg1 Rd2 30 Rc1 Qd3+ 31 Qxd3+ Rxd3 32 Rh5
After 32 Rxc6 Rd5! Black consolidates; e.g. 33 Ra6 Rxe5 34 Rxa7 g4.
32 … Kg6 33 Rh2 Re3 0-1
 A nice combination of attack and defence by Battaglini. In summary, Black should be
able to deal with lines involving an early f2-f4 comfortably.

Game 20
Y.Zherebukh-H.Nakamura
US Championship, St. Louis 2017

1 e4 c6 2 d4 d5 3 Nc3 dxe4 4 Nxe4 Bf5 5 Ng3 Bg6 6 Nh3

6 … Nf6
After 6 … e6 7 Nf4 Bd6 8 h4 Qc7 9 Nxg6 hxg6, 10 Ne4! is an old suggestion of
Kasparov on which White pins his hopes. He simply wants to nab the bishop pair and then
go slowly to work. Here 10 … Bf4 is a reasonable response; for example, 11 Bxf4 Qxf4
12 Ng5 Nf6 13 g3 Qd6 14 c3 Qd5 15 Rg1 Nbd7, which we will look at in the next game.
Instead, 10 … Nf6?! rather falls in with White’s plan, and after 11 Nxd6+ Qxd6 12 c3 Qd5
13 Be2! Qe4 (13 … Qxg2 14 Bf3) 14 Bg5 Nbd7 15 Qd3 Qxd3 16 Bxd3, White was
clearly better in E.Tomashevsky-D.Navara, Huaian (rapid) 2016.
The problem is not development, nor king safety, nor immediate counterplay. The
problem is potential. Black is not really going anywhere with his position and meanwhile
White can play for slow improvement and a gradual squeeze. This is the type of thing you
don’t really want against somebody strong, where the chances of getting ground into the
dust are very high. Having said all this, our featured game was drawn in 65 moves, so
perhaps Black’s game is not quite as bad as I’m making out.
7 Nf4
 White’s idea is a quick h2-h4, to embarrass the bishop. Nakamura seems unconcerned
and prepares … e7-e5.
7 … Nbd7! 8 Bc4
8 h4 is now met by 8 … e5!, which is a very suitable response. There is no time to play
h4-h5; for example, 9 dxe5 Qa5+! 10 c3 Qxe5+ 11 Be2 (or 11 Qe2 Bc5 12 h5 Bc2) 11 …
0-0-0 12 h5 Nc5.
8 … Qc7 9 0-0 e6 10 c3 Bd6
Black has obtained his usual solid Caro position. Zherebukh chops off the bishop on g6,
but a surprise awaits him.
11 Qf3
11 Nxg6 hxg6 12 Qf3 c5 is equal.
11 … c5!

Freeing his position. The seemingly well-placed white pieces have nothing to bite on.
12 dxc5 Bxc5 13 Bb3 0-0-0 14 Nxg6 hxg6 15 Bf4 e5 16 Be3 e4 17 Qe2?
One mechanical move too many. White is probably okay after the superior 17 Qf4 Bd6
18 Qg5 Ne5 19 Rfd1 but, if anything, I prefer Black thanks to his more active pieces.
17 … Bxe3
17 … Rxh2! launches the same combination; e.g. 18 Kxh2 Bxe3 19 fxe3 Rh8+ 20 Kg1
Qxg3.
18 Qxe3
 18 fxe3 is met by 18 … Rxh2.
18 … Rxh2!
Oh dear! White has overlooked something big.
19 Rfd1
Or 19 Kxh2 Ng4+.
19 … Rdh8 20 Qxa7 e3! 21 Qxe3
21 Qa8+ Nb8 22 Rd4 is slightly more tenacious, but even there 22 … R2h4! 23 Rxh4
exf2+ 24 Kxf2 Rxh4 is crushing. In the game, presumably demoralized, White gives up.
21 … Ng4 0-1

Game 21
J.Arizmendi Martinez-J.Houska
Llucmajor 2017

1 e4 c6 2 d4 d5 3 Nd2 dxe4 4 Nxe4 Bf5 5 Ng3 Bg6 6 Nh3

It is worth pointing out that 6 N1e2 can lead to the same position after 6 … e6 7 Nf4. In
fact, 6 N1e2 is more common than 6 Nh3.
6 … e6 7 Nf4 Bd6 8 h4 Qc7 9 Nxg6 hxg6 10 Ne4 Bf4
I prefer what Nakamura did in the last game, but this is solid enough.
11 Bxf4 Qxf4 12 Ng5


12 … Nf6
There may be more than one way to play here for Black, as the game C.Wagener-
R.Dive, Baku Olympiad 2016, shows: 12 … Nd7 13 g3 Qc7 14 Bg2 Nh6 15 0-0 Nf6 16
c3 Nf5 17 Qf3 0-0 (I suppose White is claiming a miniscule edge as the bishop on g2
dissuades … c6-c5, but this really is very little indeed; meanwhile, Black is looking for a
pawn break to liven his game up) 18 Rfe1 a5 19 Rad1 Rab8 20 Qe2 b5 21 Nf3 c5 (I am
not sure about this; 21 … b4 seems a better way to go; e.g. 22 Ne5 bxc3 23 bxc3 Nd5) 22
dxc5 (22 Qe5! is better) 22 … Qxc5 23 Qe5 Qb6 24 Nd4 Nxd4?! (24 … Rfc8) 25 Rxd4
Rfd8 26 Red1 b4 27 c4 Rxd4 28 Qxd4 Qxd4 29 Rxd4 Rc8 30 b3 Kf8 31 Rd6.

A classic endgame position has arisen where White should be close to winning. The
pawn on a5 is weak and White has very active pieces, plus a supported passed pawn on c4.
Nevertheless, Black can get his king into the game quickly and this gives him some
counterplay.
Play continued 31 … Rc5 32 Ra6 Ne8 33 Bc6 Nc7 34 Rb6 Ke7 35 Kf1 Kd6 36 Ba4+
Ke5 37 Ke2 Kd4 38 Rd6+? (I would not have been so keen to let Black’s king in; 38 Kd2!
has to be right and then 38 … Ke4 39 Bc6+ Kf5 40 Ke3 Kg4 41 Be4 and Black can still
expect to suffer) 38 … Kc3 39 Rd7 Re5+ 40 Kf1 Na6 41 Bb5 (White was obviously afraid
of 41 Rxf7 Nc5 42 Be8 Nd3 43 Rxg7 Re1+ 44 Kg2 Re2 45 Bxg6 Rxf2+ 46 Kh3 Rxa2, but
47 h5! should win; this is one of those positions where human instinct kicks in, you take
the wrong decision and the game gets away) 41 … Nc5 42 Ra7 Nxb3 (a good practical
try) 43 axb3 Kxb3 44 f4 (44 Rxf7! looks strong, and if 44 … a4? 45 Rf3+) 44 … Rc5 45
Ke2 Kc3 46 Rd7 (46 Rxa5 b3 47 Ra3 Kb2 48 Ra8 Kc3 is equal) 46 … b3 47 Rd3+ Kb4
48 Kd2 g5 (48 … e5! is better) 49 hxg5 Rc8 50 Kc1? (50 Rd7 is good; White’s failure to
make full use of his rook condemns him to a long and fruitless evening at the office) 50 …
Rh8 51 Rd2 Kc3 52 Bc6 a4 53 Bxa4 Ra8 54 Re2 Rxa4 55 Re3+ Kb4 56 g6 Ra1+ 57 Kb2
Ra2+ 58 Kb1 fxg6 59 Rxe6 Rg2 60 c5 Rxg3 61 c6 Ka3 62 Re1 Rc3 63 Re7 Rxc6 64 Rxg7
Rf6 65 Ra7+ Kb4 66 Rg7 Kc3 67 Rc7+ Kd4 68 Kb2 Ke3 69 Kxb3 Kxf4 70 Kc4 g5 71
Kd5 g4 72 Rc4+ Kf3 73 Ke5 Rf8 74 Rc3+ Kg2 75 Ke4 g3 76 Rc2+ Kg1 77 Ke3 g2 78
Rc7 Rf1 79 Rg7 Ra1 80 Ke2 Ra2+ 81 Ke1 Ra5 82 Ke2 Ra2+ 83 Ke1 Ra8 84 Ke2 Re8+
85 Kf3 Kf1 0-1.
13 g3 Qd6 14 c3 Qd5 15 Rg1 Nbd7 16 Bg2 Qa5
Another option is 16 … Qf5, with equality in J.Duda-D.Navara, European Rapid
Championship, Wroclaw 2014.
17 Qb3 0-0 18 0-0-0
18 Qxb7 Rab8 19 Qxc6 Rfc8 is not the way to go for White at all.
18 … Rac8 19 Kb1 b5 20 Bh3 c5

Black has been seeking counterplay and has now found it.
21 Rge1 cxd4 22 cxd4
22 Rxd4 Nc5 23 Qc2 Qc7 gives level chances.
22 … Qa4?
 She is rightly afraid of 22 … Rfe8 23 Nxf7! Kxf7 24 Bxe6+ Kf8 25 g4. With that line
in mind the machine is showing me 22 … Kh8!, which is not really a human-like move
but effective in this case. Black keeps the queens on, avoid any sacrifices on e6 or f7, and
pursues queenside counterplay; e.g. 23 Re3 (or 23 Rc1 Nb6) 23 … Rc6 24 Bg2 Rc7 25 d5
exd5 26 Bxd5 Nc5.
23 Qxa4 bxa4 24 d5! e5
Or 24 … Nxd5 25 Rxd5 exd5 26 Bxd7 with a clear advantage.
25 Nf3 Rc5 26 Bxd7 Nxd7 27 Nxe5 Nxe5 28 Rxe5
A long and cheerless endgame awaits Houska.
28 … a3
28 … Rd8 29 Re7 a5 30 d6 keeps White on top.
29 Re7 axb2 30 d6 Rd8 31 Rxa7 Rc1+ 32 Rxc1 bxc1Q+ 33 Kxc1 Rxd6 34 Rc7 Ra6 35 Rc8+ Kh7 36 Rc2 Kh6
37 Kb2 Ra8 38 Rc4 Rb8+ 39 Kc3 Rb1 40 a4 g5 41 hxg5+ Kxg5 42 a5 Ra1 43 Kb4 Ra2 44 Rf4 f5 45 Kb5 Rb2+ 46
Kc5 Rc2+ 47 Kb6 Rb2+ 48 Kc6 Ra2 49 Kb5 Rb2+ 50 Rb4 Rxf2 51 a6 Ra2 52 Kb6 Kf6 53 a7 g5 54 Rb5 1-0
Lines with 6 Nh3 and 6 N1e2 have come back into fashion, so Black has to be
prepared. Perhaps this stems from a desire to play a more original game than the main line
with 6 h4 allows. By following the Nakamura recipe, Black should be fine.

Now we move to 6 Bc4, a very direct idea.

Game 22
G.Kasparov-P.Bertacco
Simultaneous display, Asiag 1991

1 e4 c6 2 d4 d5 3 Nd2 dxe4 4 Nxe4 Bf5 5 Ng3 Bg6 6 Bc4


In a simultaneous display the master will look to gain the initiative right from the word
go. You never know quite who you are playing and so the choice of opening line is
important. Kasparov picks 6 Bc4 as his lethal patzer-bashing weapon, utilizing old and
very dangerous analysis to good effect.
6 … e6 7 N1e2 h6?!
Making room to retreat his bishop, as White could be threatening the usual Nf4 and
then h2-h4. There are also sacrifices on e6 that Black has to watch out for. However, 7 …
h6 looks like a panic reaction to me and unnecessary at this moment in time.
Instead, 7 … Nf6 8 0-0 Bd6 was the more sensible reaction of another of Kasparov’s
‘victims’, although this did not stop Garry for a moment. G.Kasparov-C. Van de Vlierd,
Cologne (simul) 1988, continued 9 f4! 0-0 10 f5 exf5 11 Nxf5 Bxf5 12 Rxf5 Nbd7 13 Qd3
Qc7 14 Ng3 Bxg3 (keeping calm with 14 … Rfe8 was also possible) 15 hxg3 Nb6
(leaving the knight on d7 for defence was maybe more secure, but Van de Vlierd only goes
wrong a little later) 16 Bb3 Nbd5 17 Bg5 Rae8 18 c4 Ne7 (18 … Nb4! would have given
the great Kasparov some problems; e.g. 19 Qd2 Ne4! 20 Qxb4 Qd7!! 21 Raf1 g6! and
three precise computer moves in a row have established a Black advantage; obviously it is
hard for the amateur to find these ideas against a master) 19 Rf3 Ng4 20 Raf1 Qd7 21 Bd2
Rd8 22 Bb4 Nf6?? (22 … Rde8!) 23 d5? (missing 23 Bxe7! Qxe7 24 Bc2 and wins) 23 …
cxd5 24 cxd5 Rfe8 25 Rxf6 gxf6 26 Rxf6 (26 d6!) 26 … Nxd5 27 Rh6 Nf6 28 Rg6+ Kh8
29 Bc3 hxg6 30 Bxf6+ Kh7 31 Qc4 Qf5 32 Bxd8 (here 32 Qxf7+ leads to mate in four, so
one must assume Kasparov was under some kind of pressure) 32 … Re1+ 33 Kh2 Qh5+
34 Qh4 1-0.
8 0-0 Bd6 9 Nf4 Bh7 10 Ngh5!

I seem to remember a discussion of this line during the old Botvinnik-Tal World
Championship matches and Ngh5 was one of the key attacking ideas. The knight hits g7
and stokes up the pressure. This is the type of idea that the amateur finds extremely
uncomfortable to play against. With limited time to react, what is he to do?
10 … Bf8 11 Re1 Ne7 12 Bxe6!
This wipes Black out immediately.
12 … fxe6 13 Nxe6 Qd7 14 Qf3 Rg8 15 Nxf8 Rxf8 1-0

The last, violent game suggested that 6 Bc4 is not without sting, so Black must get his
defence right. If he does, a decent position awaits.

Game 23
O.Mihok-M.Annaberdiev
Budapest 2017

1 e4 c6 2 d4 d5 3 Nd2 dxe4 4 Nxe4 Bf5 5 Ng3 Bg6 6 Bc4 e6

This has to be right, erecting a brick wall in front of the bishop.
7 N1e2 Bd6!
 8 h4
8 Nf4 is met by 8 … Nf6 9 h4 Qc7! (Black secures his position, attacking the knight
before White can play h4-h5) 10 Nxg6 (10 Qf3 Bxc2 gives White no compensation) 10 …
hxg6.
8 … h6 9 Nf4 Bxf4!
9 … Bh7 10 Nfh5 is unnecessarily complicated.
10 Bxf4 Nf6 11 Qe2 0-0
White has the bishop pair but that’s about it. He would like to use his dark-squared
bishop to attack the black king, but there seems little prospect of that. Meanwhile, Black
will put his knight on d5 and chances are equal.
12 h5
12 0-0-0 Nbd7 13 Bd3 Bxd3 14 Rxd3 Qa5 15 Kb1 Rfe8 16 Re1 ½-½ was S.Rublevsky-
A.Riazantsev, St. Petersburg (rapid) 2015. For White to make progress in a position of this
type, he has to move the knight from g3. This frees up the g-pawn to advance or prepares
Rg3. However, this costs time and Black can comfortably deal with the idea; e.g. 16 …
Rad8 17 Ne4 Nxe4 18 Qxe4 Qh5 with equality.
12 … Bh7 13 0-0-0 Nd5!
 An effective centralization.
14 Bd2 a5 15 Kb1 Nb4
Attacking c2. White can easily forget that Black can also attack in the Caro-Kann
Defence.
16 Ne4
After 16 Bb3 Black sacrificed a pawn in the grandmaster game, R.Robson-A.Ipatov,
Moscow 2012: 16 … a4 17 Bxb4 axb3 18 axb3 Re8 19 Ne4 Na6 20 Be1 Qb6 21 f3 Nc7
22 Qc4 Nd5 23 Rd3 Ra6! (putting White on the back foot) 24 Qc5 Qc7 25 Bg3 b6 (25 …
Rea8! 26 c4 Qa5 27 Qxa5 Rxa5 looks better still) 26 Bxc7 bxc5 27 Nxc5 Ra7 28 Bh2
Rea8 29 Na4 Bxd3 30 cxd3 Rb7 31 Nc5 Rba7 32 Na4 Nb6 33 Rc1 Nxa4 34 bxa4 Rxa4 35
Rxc6 Rxd4 36 Kc2 Rad8 37 Rc3 Rh4 38 Bc7 Rd5 39 g4 Rh3 40 b4 Rxf3 41 Bb6 Kf8 42
Bc5+ Ke8 43 Ra3 Kd7 44 Kd2 e5 45 Ke2 Rh3 0-1.
16 … Nd7 17 a3 b5

TIP: With the kings on opposite sides of the board, speed of attack is very important.
Black throws caution to the winds with 17 … b5 and obtains a strong initiative.

18 axb4 bxc4 19 bxa5
 19 … Rxa5!
Even the exchange is given up to accelerate the attack.
20 Bxa5 Qxa5 21 Rh3 Rb8
Setting up the threat of … Bxe4 to be followed by … c4-c3. White tries to run for it
now, but he cannot get away.
22 Kc1 Qb4 23 Rc3
23 b3 Ra8! terminates.
23 … Qxb2+ 24 Kd2 Qb4 25 Qe3 Bxe4 26 Qxe4 Nf6 27 Qf3 Nd5 28 Ra1 c5 29 Ra7
Or 29 dxc5 Nxc3 30 Qxc3 Rd8+.
29 … Rf8 30 dxc5 Nxc3 0-1
The same … Rd8+ is coming up. A fine attack!

4 … Bf5 Summary
1 e4 c6 2 d4 d5 3 Nc3 dxe4 4 Nxe4 Bf5
We could say much more about the Capablanca Variation and present literally hundreds
of recent games, such is the popularity of 4 … Bf5 in the present day. But I think I’ll leave
it there. The whole line is rock solid and suits counterpunchers and careful defenders. If
that is you, then 4 … Bf5 is a dependable choice.

The Smyslov Variation
 1 e4 c6 2 d4 d5 3 Nc3 dxe4 4 Nxe4 Nd7

I am not sure that this variation has a widely-accepted name, but I remember seeing it
called the Smyslov Variation after the famous former world champion a while back, and
so I will stick with that. 4 … Nd7 is very solid, preparing 5 … Ngf6, intending to
exchange off the powerful knight on e4. Black’s idea is logical and sound. You generally
find that White has to attack in this line or he gets nothing. If not, Black’s plan runs on
oiled wheels and he gets to a comfortable position easily. This line, too, suits good
defenders and counterpunchers. You must also be happy to play some dry positions, as if
White wants to make a draw by exchanges and simple development, he will probably be
able to do so.
We start with a game (or games) by the master himself. Smyslov was renowned for his
harmonious play and the way that the black set-up after 4 … Nd7 gradually unfolds
seemed to suit him down to the ground.

Game 24
I.Bilek-V.Smyslov
Tel Aviv Olympiad 1964

1 e4 c6 2 d4 d5 3 Nc3 dxe4 4 Nxe4 Nd7 5 Nf3

5 Bc4 is another main line, after which Simagin introduced an early aggressive twist in
V.Simagin-V.Smyslov, Alekhine Memorial, Moscow 1963: 5 … Ngf6 6 Ng5 e6 7 Ne2!?
(7 Qe2 is more common; perhaps now White intends Nf4 and a sacrifice on e6? Smyslov
loses no time repelling the dangerous knight on e5) 7 … h6! 8 Nf3 Bd6 9 0-0 Qc7 10 Nc3
(White has achieved nothing; indeed, his bishop on c4 helps Black gain counterplay) 10
… b5! 11 Bd3 b4 12 Ne4 Nxe4 13 Bxe4 Nf6 14 Bd3 0-0 15 Qe2 Bb7 16 Bd2 c5! (this
very thematic break frees up the bishop on b7 and gives Black excellent play)

TIP: In the Smyslov Variation, Black must carefully consider how he intends to

deploy the bishop on c8. In some case the bishop may come to f5 or g4, but more
often than not it deploys to b7, after which … c6-c5 becomes a necessity.

17 dxc5 Qxc5 18 h3 e5 19 Be3 Qa5 20 Bc4 Rac8 21 Rfd1 Bb8 22 Nd2 Qc7 23 Bb3
Qc6 24 f3 e4! (as both black bishops are pointing towards the white king, Smyslov opens
up the diagonals to create an attack) 25 fxe4 Nxe4 26 Nf1 Nf6 27 Rd4 a5 28 Rad1 Ba6 29
Qf3 Qxf3 30 gxf3 Be2 31 Re1 Bxf3 32 Ba4 Be5 33 Rd2 Bxb2 34 Rf2 Bc6 35 Bxc6 Rxc6
36 Ng3 Re8 37 Nf5 Kh7 38 Ref1 Rxe3! 39 Nxe3 Bd4 40 Re1 Re6 0-1. A game noticeable
for the complete absence of any White initiative.
5 … Ngf6

6 Ng3
Bilek keeps the pieces on, which is logical, as he seems to possess more space, but
illogical, because the knight drops back to a poor square. I would go so far as to say that
after 5 Nf3 Ngf6, Black is already equal.
6 Nxf6+ is the sort of move that Black is hoping for. Exchanges free his game and
make it easy to deploy the pieces to solid squares. The game S.Gligoric-V.Smyslov,
Alekhine Memorial, Moscow 1963, continued 6 … Nxf6 7 Bc4 Bf5! (if you can develop
the bishop on c8 so easily, then why not?) 8 Ne5 e6 9 0-0 (9 g4 Be4 10 f3 Bd5 11 Be2 b5
looks a bit odd, but the position is quite good for Black; White cannot take advantage of
the bishop on d5 after 12 b3 b4 or 12 a4 a6) 9 … Be7 10 Bb3 a5 11 c4 0-0 12 Bf4 c5! 13
dxc5 Bxc5 14 Qe2 Qd4! (a strong and perhaps unexpectedly powerful centralization;
suddenly, Smyslov takes over the initiative) 15 Bg3 a4 16 Bc2 (16 Rad1 Qe4 17 Qxe4
Nxe4 18 Bc2 Nxg3 is one of the main points of the position; White loses a piece) 16 …
Qxb2 17 Nd3 Bxd3 18 Qxd3 Rfd8 (at the highest level, Black expects to win such a
position and that is exactly what happens) 19 Qe2 Qb6 20 Be5 Bd4 21 Rab1 Qa5 (giving
back the pawn to maintain the initiative) 22 Bxf6 Bxf6 23 Rxb7 Qd2 (23 … Rd2 24 Qe1
Rad8 is very strong too) 24 Qe4 g6 25 g3 Qc3 26 Qf4 Rac8 27 Bd1 Rxc4 28 Qe3 Qd4 29
Qe2 Rc1 30 Bc2 Rxf1+ 31 Kxf1 Qd5 32 Be4 Qd2 33 Ra7 Qb4 34 Ra8 Rxa8 35 Bxa8 a3
36 Be4 Kg7 37 Kg2 Bd4 38 Bd3 Qc3 39 Bc4 Qa1 40 Bb3 Qc3 41 h4 h5 42 Qc2 Qe1 43
Bc4 Bxf2 44 Qxf2 Qe4+ 45 Qf3 Qxc4 46 Qxa3 Qc2+ 47 Kf3 e5 48 Ke3 Qg2 0-1.
6 … e6 7 Bd3 c5!
An excellent moment to play … c6-c5.
8 0-0 cxd4 9 Nxd4 Bc5 10 Nb3 Bb6 11 Nh5

Bilek comes to his senses and realizes that it is high time to exchange off his
unfortunate knight on g3 and this keeps the position equal. Smyslov is not bothered in the
slightest and proceeds calmly. At grandmaster level, to get such an easy position with
Black is a victory of sorts.
11 … Nxh5 12 Qxh5 Nf6 13 Qh4 Bd7 14 Nd4
At first sight, I was tempted by 14 Bg5, but after 14 Bg5 Bc6 (14 … h6 15 Rfd1 0-0 16
Bxf6 Qxf6 17 Bh7+ Kxh7 18 Qxf6 gxf6 19 Rxd7 Rac8 20 c3 Rc7 is equal) 15 Bc4 Qe7
16 Rad1 Rd8, Black is absolutely fine. However, this might have been better than the
game for Bilek, who seems to be enjoying moving his knights around.
14 … h6 15 Nf3 Bc7 16 Rd1 Qe7 17 Be3 Nd5! 18 Qxe7+ Nxe7 19 c4
It’s worthwhile taking a look at the pawn formation. White has a queenside pawn
majority and Black 4-3 on the kingside. Which is more important?
Generally speaking, the more pieces that are exchanged, the more the queenside
majority emerges as an important factor. Black should try to keep pieces on and advance
his own majority in the middlegame to try to gain the initiative. These key points should
be remembered, as this particular pawn structure is one of the most common in all light-
squared opening variations.
19 … a6 20 Bd4 f6
Smyslov prepares … e6-e5.
21 Nh4 Kf7 22 Be4 Bc6 23 Bxc6 Nxc6 24 Bc3 Rhd8 25 Nf3 e5!

There are still enough pieces on the board for Smyslov to take the initiative. The plan is
… Ke6, to be followed by a general advance of the kingside pawns. White must defend.
26 Ne1 b5!
Black has a minority attack to boot.
27 cxb5
27 b3 bxc4 28 bxc4 Rac8 just leaves White with a grotty pawn on c4.
 27 … axb5 28 Nc2?
It is still very much game on after either 28 Rxd8 Rxd8 29 Nc2 or 28 Rdc1, although
Black is better in both cases.
28 … Ke6?!
Although 28 … Ke6 is strong, I am not sure why he doesn’t play 28 … Rxa2! instead.
Maybe he just wanted to retain full control and didn’t like 29 Rdc1, when White may
scramble some tactics together. Or maybe he missed it!
29 Rdc1 Bb6 30 Kf1 Ra4 31 b3?
In view of what happens,31 g3 would have been more sensible, with Black maintaining
a sizeable edge after 31 … h5 32 b3 Ra7.
31 … Rf4! 32 Be1
White has to play 32 f3 e4 33 Ne1 and hope for the best. At a distance of 54 years, I
will make the charitable assumption that Bilek was in time trouble.
32 … Bxf2! 33 Bxf2 Rd2 34 Ne1 Rdxf2+ 0-1
In all, Smyslov played the 4 … Nd7 variation 24 times, scoring 15½-8½ – and 4 times
against it, scoring 2½-1½. It is fitting that the line bears his name.

I have always felt pretty comfortable playing against 4 … Nd7. It’s one of those lines
where Black always seems to be playing to equalize, rather than trying to confront White
head on. You don’t feel too troubled during the opening skirmishes and certainly not
threatened. The following game pretty much follows this pattern but, as we will see,
Ledger misses a few chances.

Game 25
A.Martin-A.Ledger
British Championship, Plymouth 1992

1 e4 c6 2 Nc3 d5 3 d4 dxe4 4 Nxe4 Nd7 5 Bc4

One of the main lines, where White tries to develop as actively as possible. 5 Ng5 is
also very direct and we will examine this later.
5 … Ngf6 6 Ng5 e6
6 … Nd5 may also be playable, but 6 … e6, trying to put a wall in front of the bishop
on c4, is much more common.
7 Qe2 Nb6
A necessary move, anticipating sacrifices on f7 or e6.
8 Bd3 h6
Taking on d4 looks as though it verges on the suicidal, but computers demonstrate that
it might not be too bad; e.g. 8 … Qxd4? 9 N1f3 Qd5 10 Ne5 Qxg2 11 Rf1 (the threat to f7
looks terrifying, but Black can cope) 11 … Be7! 12 Bd2 (12 Ngxf7 is met by 12 … 0-0)
12 … 0-0 13 0-0-0 Qd5 14 Kb1 and I’d prefer to play White.
 9 N5f3 c5 10 dxc5 Bxc5 11 Ne5 Qc7 12 Ngf3 Nbd7 13 0-0 0-0

Straightforward play by Black, who has his usual solid position and will contest the e5-
square for as long as it takes to reach equality. I was happy with a slight edge.
14 Bf4 Bd6 15 h3
To give the bishop a bit of air.
15 … Nxe5
15 … Nd5 16 Bg3 Nc5 17 Rad1 Nxd3 18 Qxd3 b6 is systematic, although I still prefer
White slightly after 19 Rfe1.
16 Bxe5 Bxe5 17 Nxe5 Rd8
There were other choices:
1) After 17 … b6 18 f4! is a good move; e.g. 18 … Bb7 19 Rae1 Qc5+ (or 19 … Rfd8
20 f5) 20 Kh2 and White has attacking chances with g4-g5 coming up.
2) 17 … Nd7! continues to downgrade White’s control of e5 and is the move I like best.
It’s hard to keep any edge whatsoever now. After 18 Rfe1 (or 18 f4!? Qb6+ 19 Kh2 Qxb2
20 Rad1 Nf6) 18 … Nxe5 19 Qxe5 Qxe5 20 Rxe5 Rd8 21 Be4 (or 21 Rd1 g6) 21 … Rb8,
the position is equal.
18 Rae1 Qe7 19 f4!
This is the idea. 19 f4 gains space and creates the preconditions for an attack on the
black king. The game quickly gets sharp.
19 … Bd7 20 Qf2 Be8 21 f5 b5
21 … exf5 22 Nc6 is not what the doctor ordered for Black.
 22 Nf3 Qb4 23 fxe6 Qxb2 24 e7
It is difficult for Black to deal with the advanced pawn.
24 … Rd7

25 Re5?!
I now see, with the help of my engine friends, that 25 Nh4! was the move; for example,
25 … Qd4 (or 25 … Qxa2 26 Qg3! Qa6 27 Nf5 g6 28 Nxh6+ Kg7 29 Qg5) 26 Qxd4 Rxd4
27 Nf5 Rd5 28 Nxh6+ and White is winning.
25 … Rb8 26 Qc5 Qxa2 27 Nd4
I honestly thought that he was pretty tied up and as far as I recall that was Andrew
Ledger’s assessment of the position too. But there are resources, even here.
27 … Rd5
27 … Qb2! 28 c3 b4 is the way to go, disrupting White’s cosy domination. In order not
to be worse now, White must try 29 Rxf6 gxf6 30 Re3, but this only leads to a draw: 30 …
Qc1+ 31 Kf2 Qd2+ 32 Kf3 (or 32 Kf1 Qxe3 33 Qf5 Qc1+ 34 Kf2 Qd2+ 35 Kf1, and not
35 Kf3? Qxd3+ 36 Qxd3 Rxe7 37 Nf5 Re5) 32 … Qd1+ 33 Kf2 (or 33 Be2 Qb1 34 Bd3
Qd1+) 33 … Qd2+ and draws. This is quite a difficult line to negotiate through and I’m
not sure I saw any of it at the time.
28 Qc7 Rxe5 29 Qxe5 Nd7 30 Qe4 g6

31 Nf5
I was obsessed with playing Nf5, but it may not be best. The machines tell me I should
have played 31 Bxb5! and who am I to argue? White is winning; e.g. 31 … a6 32 Bc6 Qb2
33 Qd5.
31 … Qe6
31 … Qa6 32 Nxh6+ Kg7 seems to put up resistance, but now 33 Qd4+! is a crusher as
the following variations demonstrate: 33 … Nf6 (33 … Kh7 34 Rxf7+! Bxf7 35 Nxf7 Qb6
36 Qxb6 axb6 37 Nd6 Nf6 38 Bxb5 is winning, as is 33 … Kxh6 34 Qh8+ Kg5 35 h4+
Kg4 36 Be2+ Kg3 37 Qd4 Qb6 38 Rf3 mate) 34 Nf5+! gxf5 35 Rxf5! Qb6 36 Rg5+ Kh8
37 Rc5!! Rc8 38 Qh4+ Kg7 39 Qg5+ Kh8 40 Qh6+ Kg8 41 Qf8 mate.
32 Qxe6 fxe6 33 Nd6 a6 34 Be4 Kg7 35 Bc6 b4 36 Rd1
I couldn’t see what he could do now.
36 … Kf6 37 Nxe8+ Rxe8 38 Rxd7 Rxe7 39 Rxe7 Kxe7 40 Kf2 Kd6 41 Be8 a5 42
Ke3 Kd5 43 Kd3 g5 44 Ba4 e5 45 Ke3 e4
45 … Kc4 46 Ke4 Kc3 47 Kxe5 Kb2 48 Kd4 Ka3 49 Bb3 a4 50 Kc4! wins.
46 g4 Ke5 47 Bb3 1-0
One of those games where I have really benefited from studying with the engines and
seeing how my original assessments of the game were flawed.

TIP: Use all the electronic resources available to you in order to speed up your
improvement.

Let’s have a look at one of the most recent games I can find featuring 5 Bc4. White tries
to spice things up with a relatively unusual eighth move.

Game 26
D.Andreikin-D.Bocharov
Russian Team Championship 2017

1 e4 c6 2 d4 d5 3 Nc3 dxe4 4 Nxe4 Nd7 5 Bc4 Ngf6 6 Ng5 e6 7 Qe2 Nb6

8 Bb3!?
8 Bd3 h6 9 N5f3 c5 10 dxc5 can also be met by 10 … Nbd7!?, which is a lot sharper
than 10 … Bxc5. Some Black ideas were demonstrated in A.Martin-E.Meduna, Bad
Wörishofen 1988: 11 b4 a5! (11 … g5!? is possible) 12 c3 Be7 13 Nd4 (13 a3!? Nd5 14
Bd2 axb4 15 cxb4 Bf6 16 Ra2 Nc3 17 Bxc3 Bxc3+ 18 Kf1 comes into consideration) 13
… 0-0 14 Ngf3 (14 Nh3? is poor after 14 … axb4 15 cxb4 Nxc5, but 14 Bb2 is playable;
e.g. 14 … axb4 15 cxb4 e5! 16 Nf5 b6 17 Bxe5 Nxe5 18 Qxe5 Bxf5 19 Qxf5 bxc5 20 Ne2
Ra3) 14 … e5 15 Nf5 (not 15 Nxe5? Nxe5 16 Qxe5 axb4 17 cxb4 Ng4) 15 … e4 16
Nxe7+ Qxe7 17 Bc4 axb4 18 cxb4 b6.

This is the culmination of Meduna’s plan, opening up the position with the white king
still in the middle. I thought for a long time here and found an unsettling piece sacrifice,
which is probably not sound:
19 Bf4!? Nh5 (19 … Re8 20 Bd6 Qd8 21 Nd2 is better for White; 19 … bxc5 20 Rd1!
Re8 21 Bd6 Qd8 22 Nh4 is possible) 20 Bd6 Qf6 21 Nd4 Qxd4? (Meduna goes for it but it
was better to capture on c5 first: 21 … bxc5! 22 bxc5 Qxd4! 23 0-0 Nxc5 24 Rad1 Nd3)
22 0-0 Ne5 (22 … Re8 23 Rfd1 Qc3 24 Rac1 Qxb4 25 Qxh5 wins, as does 22 … bxc5 23
Rfd1 Qc3 24 Rac1 Qxb4 25 Qxh5) 23 Qxh5 Nxc4 24 Rfd1 Nd2? (correct was 24 … Qf6!
25 Bxf8 e3 26 Qe2 Qxf2+ 27 Qxf2 exf2+ 28 Kxf2 Kxf8 29 cxb6 Nxb6 30 Rd6! Nc4) 25
Bxf8 e3 26 Qe2! (White avoids 26 fxe3 Qxe3+ 27 Kh1 Ne4 and is now winning) 26 …
exf2+ 27 Qxf2 Qxf2+ 28 Kxf2 Ne4+ 29 Ke3 Bb7 30 Be7 bxc5 31 Rd8+ Rxd8 32 Bxd8
cxb4 33 Kd4 Nc3 34 g3 Bc8 35 Ba5 Be6 36 Bxb4 Nxa2 37 Bd2 g5 38 Ke5 Kg7 39 Rf1
Kg6 40 Rf6+ Kg7 41 h4 Kh7 (or 41 … gxh4 42 gxh4 h5 43 Rxe6! fxe6 44 Kxe6) 42
Rxe6! fxe6 43 h5 Kg8 44 Kxe6 Kh8 45 Kf6 1-0. A flawed but exciting game.
8 … h6 9 N5f3 c5 10 Bf4
White has tried various plans to take Black out of his comfort zone. I doubt the
combination of bishops on b3 and f4 should worry him at all. As usual, it comes down to
how Black is going to develop his queenside pieces. Where and how will the bishop on c8
come into the game?
10 … Be7
Clearly, Black can also try 10 … Bd6! 11 Be5 0-0 12 0-0-0 Qe7 13 Bxd6 Qxd6 14 Ne5
Bd7 15 Ngf3 Rac8 16 Kb1 a6 and Black has fine chances, with the bishop on b3 a target
he can aim at.
 11 dxc5 Bxc5 12 Rd1 Qe7 13 Ne5 0-0 14 Ngf3 Nbd5 15 Bc1 b6!

White has two trumps in this type of position:
1) A strongly placed knight on e5
2) His queenside pawn majority.
Can he make anything of these advantages? I don’t see how. Black is about to complete
development by actively deploying his c8-bishop and after that he can bring his rooks to
the centre. Chances are equal.
16 a3
White would like to play 16 c4, but 16 … Bb4+! is inconvenient; e.g. 17 Kf1 (17 Nd2
Ba6 18 0-0 Qc7 is equal) 17 … Ba6 18 Nc6 Qc7! 19 Nxb4 Nxb4 20 a3 Nc6.
16 … a5
I quite like the sensible 16 … Qc7 17 0-0 Bb7 18 c4 Ne7 19 Bc2 Rfd8, when there is
absolutely nothing wrong the black position.
17 c4 Ba6 18 0-0 Qb7 19 Nd3 Ne7 20 Nxc5 bxc5
The split black pawns restrain White’s majority.
21 Ba4 Nf5 22 h3 Qb6 23 b3 Bb7 24 Bb2 Rfd8
 The first sign that the game is going to end in a draw. Black offers suitable exchanges
along the only open file, not fearing Bxf6.
25 Rxd8+ Rxd8

26 Rd1
Many club players would automatically take the opportunity to mess up the black
kingside, so let’s try and figure out why Andreikin doesn’t do this. After 26 Bxf6 gxf6
there are three considerations:
1) The first problem for White is the bishop on a4, which is completely out of play. It is
doubtful that this piece can help to attack the black king.
2) Where and how is Black going to be mated? His king can cover everything from g7.
3) Black has control of d4 and a very powerful bishop on b7. It may be White who
could come under attack because of these features of the position. So, in order to keep the
balance here, White must play the rather tedious 27 Rd1, when a draw not far off.
26 … Ne4 27 Qe1 f6 28 Nh2 e5
Carefully locking the bishop on b2 out of the game.
29 f3 Ng5 30 Bc3 Rxd1 31 Qxd1 e4! 32 Qd7 Nd4 33 Kf2
33 Bxd4 cxd4 34 Qe7 exf3 35 Bd7! aims for Qe8 and then Bf5+, but Black can deal
with this with 35 … f2+!! 36 Kxf2 g6 37 Qe8+ (37 Ng4 d3+ 38 Kg3 Qc7+ 39 Kh4 Qf4!
40 Qe8+ Kg7 41 Qe7+ etc) 37 … Kg7 38 Qe7+ Kg8 with a draw.
33 … exf3 34 gxf3 Nge6 35 Qe7 Bc6 36 Qd6 Qb7 37 Bxd4 Nxd4 38 Qxc5 Bxa4 39
Qxd4 Qxb3
 The various exchanges have left the position burned out. A draw is imminent.
40 Qd5+ Kh8 41 Qa8+ Kh7 42 Qe4+ Kg8 43 Qd5+ Kf8 44 Qc5+ Kf7 45 Qh5+ Kf8
46 Qc5+ Kf7 ½-½

We now come to the move 5 Ng5 which is most popular in 2017 and has been for some
time now. White is trying to find the best way to organize an attack and thinks he has
found it. 5 Ng5 announces outright aggression, with possible sacrifices on f7 in the offing.
Grandmasters are happy to defend the black cause, but one has to be very careful.

Game 27
M.Huerga Leache-D.Larino Nieto
Elgoibar 2010

1 e4 c6 2 d4 d5 3 Nc3 dxe4 4 Nxe4 Nd7 5 Ng5 e6?!

Already a bit unusual and rather passive. The main line is well established; Black
should play 5 … Ngf6, when the game S.Collins-N.Short, British League 2016, is
instructive: 6 Bd3 g6!? (Nigel Short rejects the usual 6 … e6 and takes a walk along an
offbeat path; 6 … g6 is playable, but we will see Black ending up in a passive situation) 7
N1f3 Bg7 8 0-0 0-0 9 c3 a5 (a position that resembles lines of the Alekhine or
Scandinavian; Black is bit worse due to his lack of space and his ponderous development)
10 a4 Nd5 11 Re1 b6 12 Qe2 Ra7 13 h4!? (the simple 13 Bd2 would have maintained an
easy edge; Collins abandons respect and decides to go for it) 13 … N7f6?! (a peculiarly
tame move from such an aggressive player; Black must respond sharply and immediately
with 13 … e5! 14 h5 exd4 15 hxg6 hxg6 16 Bxg6! N7f6! – maybe Short came to this
position in his analysis and did not like it, but Black is surviving after 17 Nxd4 fxg6 18
Nxc6 Qd7 19 Nxa7 Qxa7; note too that 14 dxe5 Nxe5 15 Nxe5 Bxe5 fails to 16 Qxe5
Re7) 14 Ne5 Rc7 15 h5 (White has been allowed everything he could possibly wish for,
and his attack proceeds uninterrupted) 15 … Nxh5? (15 … e6) 16 Qxh5! h6 (16 … gxh5
17 Bxh7+ Kh8 18 Nexf7+ wins) 17 Nexf7 Rxf7 18 Nxf7 1-0.
6 Bd3 Be7 7 N1f3 Ngf6 8 Qe2
At this point in all lines where White sets up with a knight on g5, Black must ask
himself, is he threatening Nxe6 or Nxf7 - ?
8 … Qa5+ 9 Bd2 Qb6

10 0-0-0
Personally, I cannot understand why he does not play Nxf7 immediately; e.g. 10 Nxf7!
Kxf7 11 Ng5+ Ke8 12 Qxe6 Nf8 13 Qf7+ Kd8 14 0-0-0 Qc7 15 Rde1 Bg4 16 Qxg7 Rg8
17 Qh6. This is just one variation, but the impression is formed that it is hellishly difficult
to unravel Black’s game.
10 … c5
Really asking for it.
11 Nxf7! Kxf7 12 Ng5+ Kg8 13 Nxe6 Nb8 14 Nxg7!
Maybe he missed this at the start. 14 Nxg7 wipes Black out immediately.
 14 … Bf8 15 Nh5 Kf7 16 Bc4+ Kg6 17 Qd3+ 1-0

Twenty years ago, Garry Kasparov lost a famous game to Deep Blue in the Ng5
variation, more or less by playing what looked to be a losing move very early on. Let us
see whether this is the full story.

Game 28
H.Stefansson-J.Granda Zuniga
Reykjavik 2015

1 e4 c6 2 d4 d5 3 Nd2 dxe4 4 Nxe4 Nd7 5 Ng5 e6

5 … h6 6 Ne6! Qa5+ is just about playable for Black but cannot really be
recommended; for instance, 7 Bd2 Qb6 8 Bd3 Ngf6 9 Nxf8 Nxf8 is the best Black can do.
If you like his game, by all means give this line a whirl. All we can say is that 5 … h6
does not lead to a forced loss.
6 Bd3 Ngf6 7 N1f3 h6!?

Granda Zuniga repeats Kasparov’s ‘mistake’. Since he analysed the 1997 debacle for
ChessBase, it will be intriguing to see the improvement.
8 Nxe6! fxe6
8 … Qe7? was played by Kasparov, after which Black is totally lost. Let us see the rest
of the horror show: 9 0-0 fxe6 10 Bg6+ Kd8 11 Bf4 b5 12 a4 Bb7 13 Re1 Nd5 14 Bg3
Kc8 15 axb5 cxb5 16 Qd3 Bc6 17 Bf5 exf5 18 Rxe7 Bxe7 19 c4 1-0, Deep Blue-
G.Kasparov, New York (Game 6) 1997. Granda has burned the midnight oil and
recognizes that Black can take the piece and live.

TIP: Opening ideas which have failed at the highest level are often written off

prematurely. Researching these ideas can pay dividends.

9 Bg6+ Ke7
White has excellent compensation, but Black plans … Qc7 followed by … Kd8, slowly
unravelling himself.
10 0-0 Qc7 11 Re1 Kd8

12 Ne5?
Confronted by a line he could not possibly have expected, Stefansson misplays the
attack. Perhaps he should just take on e6 as 12 Rxe6 seems to be the right way to continue.
After 12 … Bd6 13 Re1 Nf8 14 Bd3 Be6 15 c4 Kc8 White only has two pawns for the
piece, but Black has the devil of a job to untangle.
12 g3 is another possibility, intending Bf4. Black should meet this with 12 … Bd6 13
c4 Nf8 14 Bc2 Qe7 15 Bd2 Bd7, leading to another one of those hard-to-judge situations,
where White has obvious positional compensation for the piece but no clear way through.
It will be seen that both of these lines are better than the way Stefansson plays.
12 … Nxe5 13 dxe5+ Nd7 14 Bf7 Bc5! 15 Be3?
A second error, underestimating Black’s bold next move. For the second time, White
must take on e6; e.g. 15 Bxe6 Re8 16 Qg4 Nxe5 17 Bg5+! hxg5 18 Rad1+ Nd7 19 Qxg5+
Be7 20 Bf7, when Black is two pieces up but is in what looks like a straitjacket.
Extraordinarily, he can strike back with 20 … Qxh2+!! 21 Kxh2 Rh8+! 22 Qh5 Rxh5+ 23
Bxh5 Bf6 with equality. Two pieces for a rook and pawn. Who would have believed it?
15 … Qxe5 16 Qd2
Or 16 Bxe6 Kc7.
16 … Qd6 17 Qe2 Bxe3 18 Qxe3 e5 0-1
White has nothing for the piece really. As Granda is renowned for his grittiness and
tough defending, it is quite in order to resign. Can Black really get away with the move 7
… h6 - ?

Short games, such as the couple that we have just witnessed, force players to examine a
variation closely, lest they get caught out in the same way. Thus the 5 Ng5 line is heavily
analysed these days and the coming game is pretty much the latest word on the theory.

Game 29
J.Vykouk-P.Simacek
Ceske Budejovice 2017

1 e4 c6 2 d4 d5 3 Nd2 dxe4 4 Nxe4 Nd7 5 Ng5 Ngf6 6 Bd3 e6 7 N1f3 Bd6 8 Qe2 h6 9
Ne4
Sacrifices on e6 and f7 simply don’t work here.
9 … Nxe4 10 Qxe4 Qc7

11 Qg4
 White plays this move to stop Black from castling. 11 Bd2 is a decent alternative, but
by sticking to the easy plan of … b6, … Bb7 and 0-0-0, Black reaches equality, as in
I.Ortiz Suarez-M.Vachier-Lagrave, FIDE World Cup, Tromsø 2013: 11 … b6! 12 0-0-0
Bb7 13 Qe2 0-0-0 14 Ba6 Rhe8 15 Rhe1 Bxa6 16 Qxa6+ Kb8 17 Qe2 e5! (once Black
frees himself in this way, he has a perfectly satisfactory position; nothing more than that,
but for Vachier-Lagrave, the game must continue) 18 Kb1 e4 19 Nh4 Bf4 20 Bxf4 Qxf4
21 g3 Qg5 22 Ng2 Nf6 23 Ne3 Re6 24 Rd2 g6 25 c4 Red6 26 Red1 Qa5 27 c5 Re6 28
Nc4 Qb5 29 Rc1 Nd5?! (there was no rush to occupy d5; 29 … h5, slowly strengthening
the position, would have maintained a balance) 30 Ne5 Qxe2 31 Rxe2 f5 32 cxb6 Ne7 33
bxa7+ Kxa7 34 Rc4 Rd5 35 Re3 c5 36 Rb3 (White will have to work hard to lose this
position, but Ortiz manages it) 36 … cxd4 37 Rc7+ Ka8 38 Nd7 Rxd7 39 Rxd7 Nc6 40
Kc1 g5 41 Kd1 f4 42 f3 e3 43 Ke2?? (43 gxf4 gxf4 44 Ke2 was the right way to play) 43
… g4! (suddenly the game turns on its head) 44 Rf7 gxf3+ 45 Ke1 Ne5 46 Rxf4 Rc6 47
Kd1 e2+ 48 Kd2 Rc1 0-1.
11 … Kf8 12 0-0 b6
The same plan for Black, but with a slightly more awkward king position.
13 Re1 Bb7 14 b3 Nf6 15 Qh3 Rd8
Theoreticians are out as to whether 15 … Rd8 or 15 … Nd5 is the better move. Both
seem fine, with White having to work hard to prove any advantage whatsoever. For
example, 15 … Nd5 16 c4 Nf4 17 Bxf4 Bxf4 18 b4 Rd8 19 Rad1 h5 is equal.
16 c4 c5 17 dxc5 Bxc5 18 Bc2 b5!

19 Ne5
All known to here, which I cannot say I am pleased to report, with 19 Ne5 being the
new move. 19 Bb2 bxc4, as in J.Jackson-S.Kozhuharov, Sunny Beach 2015, seems a bit
better for Black already thanks to the raking bishops: 20 Rad1? (20 bxc4 Bxf2+ 21 Kxf2
Qb6+ 22 Bd4 Rxd4 23 Nxd4 Qxd4+ 24 Kf1 Qxc4+ 25 Bd3 Qf4+ 26 Kg1 Ng4 was
preferable) 20 … Rxd1 21 Rxd1 c3 22 Bxc3 Bxf2+ 23 Kxf2 Qxc3 24 Bb1 Qc5+ 25 Ke2
Ba6+ 26 Kd2 Ke7 27 Ne1 Rd8+ 28 Nd3 Ne4+ 29 Ke1 Qg1+ 30 Ke2 0-1.
19 … bxc4 20 bxc4 Kg8 21 Bf4 g5! 22 Bg3 Qe7 23 Bg6 Kg7!
So much for inconveniencing the black king. Suddenly, it is Black who is better
developed and putting on the pressure.
24 Bxf7

24 … g4?
But not like this, a move which lets the white queen out of jail. 24 … Bd4! was the way
to go, when White is struggling: 25 Bh5 (25 Rab1 Be4! or 25 Rad1 Bxe5 26 Bxe5 Rxd1
27 Rxd1 Kxf7) 25 … Bxa1 26 Rxa1 Rd2! 27 Ng6 Qf7! 28 Be5 Rhd8.
25 Qh4 Rd2 26 Bh5 Bd4 27 Rab1 Ba8 28 Nxg4 Ng8 29 Qxe7+ Nxe7 30 Rbd1
30 Bf4 Rxa2 31 Rbd1 also wins.
30 … Rxd1 31 Rxd1 Nf5 32 Rxd4
Or 32 Be5+ Bxe5 33 Rd7+!, winning.
32 … Nxd4 33 Be5+ Kh7 34 Nf6+ Kg7 35 Ne8+ 1-0
It is clear that the main line is okay for Black, even with his king on f8 – but, as usual
these days, you have to know it backwards.

5 Bd3 can also be played, a move that has similar designs to 5 Ng5.

Game 30
I.Selivanov-V.Burmakin
Chigorin Memorial, St. Petersburg 2016

1 e4 c6 2 d4 d5 3 Nc3 dxe4 4 Nxe4 Nd7 5 Bd3

We must note that after 5 Ng5 Ndf6, White has 6 Bc4!.
5 … Ndf6!?
Play often transposes back to the 5 Ng5 main line after 5 … Ngf6 6 Ng5 e6 7 N1f3.
6 Ng5 Bg4

7 N5f3
7 N1f3 seems a bit more natural, but Black is able to complete his development
satisfactorily, as in Yu Yangyi-Ding Liren, Chinese Championship, Xinghua 2014: 7 …
Bh5 8 c3 e6 9 Qb3 Qc7 10 Ne5 Bd6 11 0-0 Ne7 12 Re1 0-0 13 Bf4 Rad8 (once a position
like this is reached, it remains to look for a pawn break to free up the pieces) 14 Bg3 b6 15
a4 c5! 16 Rac1 Qc8 17 Bc4 Ned5 18 Bb5 h6 19 Ngf3 Bxf3 20 Nxf3 Bxg3 21 hxg3 Ng4
22 Qc4 Ndf6 23 dxc5 Qxc5 24 Qxc5 bxc5 25 b4 cxb4 26 cxb4 Nd5 27 Rc4 h5 28 Bc6
Rd6 29 b5 Nb6 30 Rc2 Nxa4 31 Ra2 Nc5 32 Rxa7 Rfd8 33 Ng5 Rd1 34 Ra1 Rxa1 35
Rxa1 Ne5 36 f4 Nxc6 ½-½. Not that exciting to watch, but very good defence by Black.
7 … e6 8 Be3
8 Bg5 Qb6! 9 Qc1 Bd6 10 c3 Bh5! equalizes comfortably. Note the effectiveness of the
idea … Bg4-h5, covering f7. T.O’Donnell-G.Spraggett, Canadian Championship,
Hamilton 1994, continued 11 Nd2 Qc7 12 Ngf3 Nd5 13 0-0 h6 14 Bh4 Nf4 15 Qc2 Ne7
16 Rfe1 Bxf3 17 Nxf3 Nxd3 18 Qxd3 0-0 19 Bg3 Nf5 20 Bxd6 Qxd6 21 Rad1 Rfd8 22
Ne5 Rac8 23 h3 Qc7 24 Qe4 Nd6 25 Qf3 Qe7 26 Rd2 Qg5 27 Qd1 Qe7 28 Rd3 Ne8 29
Re2 Nf6 30 Rdd2 ½-½.
8 … Qb6
As soon as the bishop on c1 leaves home, … Qb6 becomes an option.
9 Qc1 Nd5 10 Ne5 Nxe3 11 fxe3 Nf6 12 Ngf3 Be7

This is simple fare. Black is under no pressure and can bring all his pieces out to good
squares. The success of 5 … Ndf6 goes a long way to explain why 5 Ng5 is preferred to 5
Bd3 in current practice.
13 0-0 Bxf3
13 … 0-0 is another good option. We will quickly see that Burmakin is looking to
unbalance the position in an effort to win and not just play to maintain equality.
14 Rxf3 Qc7
Objectively, 14 … 0-0 is strongest.
 15 Qe1 Bd6 16 Qg3 Bxe5 17 dxe5 Nd7
Black is taking some risks here.

18 Qxg7
White should play 18 Raf1! with the idea 18 … Nxe5? 19 Rxf7! Qxf7 20 Rxf7 Nxf7 21
Qxg7 Ke7 22 Bxh7 Raf8 23 h4! and Black is lost, as White will simply trundle the
kingside pawns up the board. Instead of 18 … Nxe5, Black can try:
1) 18 … Qxe5 is not much better after 19 Rxf7 Qxg3 20 hxg3 g6 (or 20 … g5 21 Rg7)
21 Rxh7 Rxh7 22 Bxg6+.
2) 18 … 0-0 19 Qh4 h6 20 Qe4 f5 21 exf6 Nxf6 22 Qg6! is terminal.
3) 18 … Rf8 seems best, but even there 19 Qxg7 Nxe5 20 Rh3 0-0-0 21 Rxh7 is very
good for White.
18 … Rf8 19 Bxh7
19 Raf1 Nxe5 20 Rh3 transposes to line 3 in the previous note.
19 … Nxe5 20 Rff1 0-0-0 21 Be4 Rg8
Black is starting to get on top. We understand that this is nowhere near as good for
White as the other lines, because he has not yet brought the rook on a1 into the game.
22 Qf6 Ng4 23 Qf4 f5 24 Bf3
After 24 Qxc7+ Kxc7 the e3-pawn drops.
24 … Qb6 25 Rfe1?
Under pressure, the weaker player crumbles. White should play 25 Bxg4 Rxg4 26 Qh6
Re4 27 Rad1, when he should save the game.
25 … Qxb2 26 Rab1 Qc3 27 Qa4 Ne5

28 Be2?
28 Qxa7 Nxf3+ 29 Kh1 was the last chance and could have forced a very long game
after 29 … Qxe1+ 30 Rxe1 Nxe1 31 Qa8+ Kc7 32 Qa5+ Kd7 (or 32 … b6) 33 Qxe1 Rg4.
28 … Qxe3+ 29 Kf1 Rd4 0-1

Smyslov Variation Summary

1 e4 c6 2 d4 d5 3 Nc3 dxe4 4 Nxe4 Nd7

4 … Nd7 can clearly be recommended. Some of the positions reached can be a little bit
dry, and the average player generally looks for a more exciting life. But if you have a
patient temperament, go for it.

The Variation with 4 … Nf6
1 e4 c6 2 d4 d5 3 Nc3 dxe4 4 Nxe4 Nf6



4 … Nf6 is a controversial move, because Black allows his pawns to be doubled after 5
Nxf6+. How he recaptures then produces two completely different types of game. He may
play 5 … gxf6, the Bronstein-Larsen variation, which is unbalancing and aggressive.
Alternatively, he may also play 5 … exf6, a much quieter move, where Black relies on
comfortable development to compensate him for the inferior pawn structure. We will look
at each in turn. It’s worth pointing out that any other move apart from 5 Nxf6+ leads to
nothing for White.

Game 31
A.Gipslis-Y.Seirawan
Saint John 1988

1 e4 c6 2 d4 d5 3 Nc3 dxe4 4 Nxe4 Nf6 5 Nxf6+ gxf6

 We reach a critical position already. What should White be thinking about here?
1) White has to be careful as Black is about to get pretty active. In particular, White has
to watch out for an attack against his king.
2) The advance g2-g3 could be useful.
3) Quick development, followed by c2-c4 and d4-d5 could disrupt Black’s plans.
4) White should keep calm. He has the superior pawn structure but must recognize that
it will take patience to prove it.
Let us also list what Black is considering:
1) He deliberately allows White to double his pawns but in return will get active play.
2) Black may now use the g-file to attack the white kingside.
3) Black has no real problem developing his pieces. His ideal set-up involves a bishop
on f5 or g4, a pawn on e6, queen on c7, bishop on d6 and castling long.
4) Black may use his isolated h-pawn as a battering ram against White’s kingside.
With all this in mind, let us view a thematic win by Black.
6 c3
Back in 1988, players were still unsure about the best way to tackle Black’s system. 6
c3 preserves a few options, including denying Black the possibility for 6 Nf3 Bg4. Gipslis
is encouraging Seirawan to develop his light-squared bishop to f5, after which he will hunt
it down.
6 … Bf5
Black obliges. Short played 6 … h5!? in Game 7.
 7 Ne2 Nd7 8 Ng3 Bg6 9 h4 h5
9 … h6?! hasn’t turned out too well, so the game move is thought to be forced.
However, it means Black has to sacrifice a pawn.
10 Be2 Qa5!? 11 b4 Qc7

12 Nxh5
Current thinking states that the pawn sacrifice might not be 100% sound, but we can
see that it is very interesting. White will have problems finding a secure haven for his
king, especially now that he has played both h2-h4 and b2-b4. Practically, I would not
mind playing Black at all here, especially if my opponent came unprepared.
12 … e5
Hitting hard at White’s shaky centre.
13 Be3
Preparation was not as good in those days as it is now and I wonder whether Gipslis
knew of a previous Seirawan game in this line. J.Peters-Y.Seirawan, US Championship,
Berkeley 1984, continued 13 Ng3 0-0-0 14 h5 Bh7 15 Qb3 Nb6 16 b5 c5! (Yasser tries to
block the queenside, but Peters ploughs on) 17 dxc5 Bxc5 18 a4 Rhg8 19 a5 Nd5 20 b6
axb6 21 a6 bxa6 22 Bxa6+ Kb8 23 Bc4 (it looks as though White has a dangerous attack,
but he is way behind in development and this will cost him in the end) 23 … Nf4 (23 …
Nf4 is good, but 23 … Nxc3! is even better; e.g. 24 Qxc3 Bd4 25 Ra8+ Kxa8 26 Bd5+
Kb8 27 Qxc7+ Kxc7 28 Bxf7 Rg7 29 Bb3 Bd3, or the exceptionally attractive variation 24
Ba3 Bxa3 25 0-0 Bc5 26 Qxc3 Bd3! 27 Rfc1 Bxf2+ 28 Kxf2 Qc5+ 29 Kf3 e4+ 30 Nxe4
Qf5+ 31 Ke3 Qxe4+ 32 Kd2 Qe2 mate) 24 Kf1 Qb7 25 Rh2 b5 26 Be2 Nxe2 27 Nxe2
Be4 28 f3 Bd3 29 Ke1 Bc4 30 Qc2 Rd3 31 h6 Qd5 32 h7 Rd8 33 Bg5 Bb3 34 Qxd3 Qxd3
35 h8R Rxh8 36 Rxh8+ Kb7 37 Bd2 Bc4 38 Ng3 e4 39 Rh5 Bd6 0-1.
13 … 0-0-0 14 0-0 Be7

15 f4?!
I am really not sure about this as opening the kingside seems to be playing into
Seirawan’s hands. Perhaps White should just press on with 15 a4, although Black is fine
after 15 … Nb6 16 a5 (or 16 b5 Kb8 17 bxc6 Qxc6 18 a5 Nd5) 16 … Nd5 17 Bd2 exd4
with excellent play in both cases.
15 … Nb6! 16 Rc1 Nd5
The manoeuvre of the knight to d5 has been enhanced by f2-f4. Black is now clearly
better.
17 Bf2 exf4
White is going to be nailed to g2.
18 b5?
Gipslis collapses in a position that he does not like. 18 Bf3 appears forced, but Black is
still much better after 18 … f5! 19 Qe2 Ne3 20 Nxf4 (or 20 Bxe3 fxe3 21 Qxe3 Bxh5) 20
… Nxf1 21 Rxf1 Bxh4.
18 … Ba3 19 Ng7 Bxc1 20 Qxc1 Rdg8 21 Bg4+ 21 … Kb8 22 Nf5 Bxf5 23 Bxf5 f3
24 g4 Ne7 0-1
This is the sort of game that could be very inspiring if one wishes to take up 5 … gxf6.

 Modest play cannot challenge 5 … gxf6, but it is an ideal approach when you are the
much stronger player. In the next game, the grandmaster is happy to suck all the tactics out
of the position and let his opponent self-destruct.

Game 32
L.Bruzon Batista-J.Moreno Mateos
Calvia 2006

1 e4 c6 2 d4 d5 3 Nc3 dxe4 4 Nxe4 Nf6 5 Nxf6+ gxf6 6 Bc4

This was the main line for quite some time, until superseded by 6 c3, which keeps more
options open for the white minor pieces. The light-squared bishop is probably best placed
on g2.
6 … Bf5 7 Ne2 e6
A wall is erected in front of the bishop on c4. Instead, 7 … Nd7 8 0-0 Nb6 9 Bb3 Qd7
10 c4 Bg7 11 Be3 0-0 12 Qd2 Bg6 13 Rfd1 Nc8 14 Nf4 Nd6 was a little bit better for
White in P.Popovic-Y.Seirawan, Sarajevo 1987. The structural weaknesses seem
accentuated when the king is on g8 rather than c8, and now with 15 d5! the position opens
to White’s advantage.

8 Bf4!
Very good if you just want to play it safe. Black is quite keen to play … Qc7 and …
Bd6, and so this option is removed.
8 … Bd6 9 0-0 Qc7 10 Bxd6 Qxd6
 The position is approximately equal but, of course, Black hasn’t repaired his pawns yet.
Bruzon puts the thought of getting ground down into his opponent’s mind.
11 Bd3
11 f4! was rather interesting here. Black would be unwise to allow f4-f5, so 11 … Bg4!
is the best reply, but even here White is a tad better after 12 c3 Nd7 13 Qb3 0-0-0 14 Ng3
Rhg8 15 f5. I suspect Bruzon doesn’t do this, because he wants to play without the
slightest risk.
11 … Bg6 12 Qd2 Nd7 13 Rfd1 0-0-0 14 Qe3 Kb8 15 c4
I mentioned the idea of c2-c4 earlier and here it is in action. White is thinking of a
queenside pawn storm and there may be an occasion where d4-d5 is a possible idea too,
although obviously not yet.
15 … c5 16 Bxg6 hxg6 17 h3 cxd4 18 Rxd4

18 … Qe5?
The grandmaster has assessed the situation very shrewdly. At the first sign of pressure,
Black collapses and gets into a terrible mess on the open file. Instead, Black should play
18 … Qe7! 19 Rad1 Ne5 20 b3 Nc6, when the position is level.
19 Qd2 Qc7 20 Rd1 Kc8 21 Nc3
Suddenly, Black is under serious attack, as he cannot stop the knight from entering the
game.
21 … Ne5?
21 … a6 22 Ne4 Ne5 was more tenacious, but there can be no doubt about White’s
large advantage after 23 c5! Rxd4 24 Qxd4 Rd8 25 Nd6+ Kb8 26 b4.
 22 Nb5! Rxd4 23 Qxd4 Qb6 24 Nxa7+ Kc7 25 Nb5+ 1-0
A cautionary tale.

TIP: Give weaker players plenty of choices to make and deny them active plans.
Mistakes follow.

We now move into the modern era, where 5 … gxf6 is not scoring as well as it used to.
I get the impression that the line is good up to certain level, say 2500, after which it
becomes dubious. Of course, for club players this is not a concern.

Game 33
R.Hess-A.Stripunsky
US Championship, St. Louis 2013

1 e4 c6 2 d4 d5 3 Nc3 dxe4 4 Nxe4 Nf6 5 Nxf6+ gxf6 6 g3!?

Very logical, snuffing out any counterplay on the g-file.

6 … Nd7?!
To me, this just cedes the advantage to White without a fight. Other possibilities are:
1) Black can try 6 … Qd5! (as in E.Rozentalis-I.Kovalenko, Tromsø Olympiad 2014),
disrupting the flow of White’s development. Whether he is fully equal after 7 Nf3 Bf5
(other tries are 7 … Na6!? 8 c3 Bg4 9 Bg2 Qe6+!? 10 Be3 Bh6 11 Qe2 Bxe3 12 Qxe3
Qxe3+ 13 fxe3 0-0-0 14 0-0 and 7 … Bg4 8 Bg2 Qe6+ 9 Be3 Bh6 10 Qd3 Bxe3 11 Qxe3
Qxe3+ 12 fxe3 Nd7) 8 c3 Nd7 9 Bg2 Qc4 10 Qb3 Qa6 11 Bf1 Bd3 12 Bxd3 Qxd3 13 c4
Qe4+ 14 Qe3 is not certain. For me, he is almost there.
2) It is also possible to push the h-pawn with 6 … h5!? as in E.Nakar-L.Paichadze, European Championship, Plovdiv
2012: 7 Nf3 Bg4 8 Bg2 h4 (this looks like an efficient way to get rid of a weak pawn) 9 h3 Bxf3 10 Bxf3 hxg3 11 fxg3
e6 12 Bf4 Bd6 13 Qd2 Bxf4 14 Qxf4 Qa5+ 15 Qd2 Qc7 16 Qf2 Nd7 17 0-0-0 f5 18 Kb1 Nb6 19 g4 fxg4 20 hxg4 0-0-0
(Black is very comfortable) 21 Be4 Rhg8 22 Qf3 Nc4 23 Qe2 Nd6 24 Bd3 Rg5 25 Rh5 Qe7 26 Be4 Nxe4 27 Qxe4 Rdg8
28 Rxg5 Qxg5 29 Rg1 Rg6 30 c4 f5 31 Qe5 fxg4 32 Qxg5 Rxg5 33 Rg3 Kd7 34 Kc2 e5 35 Kd3 Ke6 36 dxe5 Kxe5 37
Ke3 Rg7 38 b3 Kf5 39 Kf2 Rh7 40 Rg2 Rh1 41 Ke2 b6 42 a4 Rb1 43 Rg3 Kf4 44 Rd3 g3 45 Rf3+ Kg4 46 Rf7 g2 47
Rg7+ Kf4 48 Rxg2 Rb2+ 49 Kf1 Rxg2 50 Kxg2 Ke3 0-1.
7 Bg2 Nb6 8 b3 Bf5 9 c4 Qd7 10 a4!?
Hess meets the system with a original set-up. I really like what White does in this
game.
10 … Rd8
So that the rook is not stranded on a8.
11 a5 Nc8 12 a6 b6 13 Ne2 Bh3
13 … h5 certainly comes into consideration.
14 Bxh3 Qxh3 15 Nf4 Qd7 16 0-0!

16 … Bg7
Taking the pawn is suicidal with the king in the middle; e.g. 16 … Qxd4? 17 Qf3! Qd7
18 Nh5! Rg8 19 Re1 Bg7 20 Bf4 Kf8 21 Rad1 Nd6 22 b4, winning.
17 Qf3 0-0 18 d5
 White is much better. Not only has he retained his superior pawn structure, he has also
developed a strong attack, with options on both sides of the board. His kingside chances
are amply seen after 18 … c5 19 Nh5!, and so Stripunsky captures on d5.
18 … cxd5 19 Nxd5 f5 20 Ra2! e6 21 Nb4 Qd4 22 Qb7 Qc3 23 Nc6 Rd3 24 Ra3
It is hard to say why Hess rejected 24 Bb2!; e.g. 24 … Qxb3 (24 … Qxb2 25 Rxb2
Bxb2 26 Nxa7 wins) 25 Ra3! Qc2 (or 25 … Qxc4 26 Bxg7 Kxg7 27 Ne5) 26 Bxg7 Kxg7
(or 26 … Rxa3 27 Bxf8 Kxf8 28 Qxc8+) 27 Nb4. He may have missed 25 Ra3, or he may
have wanted to avoid even the slightest risk, as the game move is also very good for
White.
24 … Bd4 25 b4! Qxc4 26 Rxd3 Qxd3 27 Nxd4 Qxd4 28 Bh6 Rd8 29 Rc1

Bingo! Black’s pieces turn out to be poorly co-ordinated, and if the knight moves, a7
drops. Black is now lost.
29 … Nd6 30 Qxa7 Ne4 31 Be3 Qxb4 32 Qxb6 Qxb6 33 Bxb6 Ra8 34 a7 Nf6 35
Bd4 Nd5 36 Rb1 1-0
I have only provided slight evidence of the struggle after 5 … gxf6, as the line is not as
popular as it used to be. If you play it at master level, you will have to be incredibly well
prepared to make the system work. Meanwhile, club players can still have a lot of fun but,
once again, only after proper research.

We now come to the variation where Black recaptures on f6 with his e-pawn. I would say this is a difficult line to
handle for any level of player.
Game 34
J.Nunn-K.Arkell
Lloyds Bank Open, London 1987

1 e4 c6 2 d4 d5 3 Nd2 dxe4 4 Nxe4 Nf6 5 Nxf6+ exf6!?

Black will get free play for his pieces, but he has handed his opponent the permanent
trump of a queenside pawn majority which is quite a big concession. Observe our first
game, where Arkell deploys his pieces only slightly inaccurately and gets smashed as a
result.
6 c3
White has many sixth moves, but 6 c3 is the one that most masters regard as the best.
White keeps options open for all his pieces and at the same time reinforces his centre.

6 … Bf5?!
What could be more natural in the Caro than developing the bishop to f5, yet 6 … Bd6!
is much better. The bishop is exposed on f5 and White can gain time by attacking it.
7 Ne2! Bd6 8 Ng3 Bg6 9 Bc4 0-0 10 0-0
The advance f2-f4 has now become a serious threat.
10 … Re8
Nunn suggests that the damage can be limited with 10 … Bxg3 11 hxg3 Nd7 12 f4
Be4!.
 11 f4!
Very strong and, as Nunn shows in his analysis, there is very little that Black can do
about the threat of f4-f5.
11 … b5
Others all fail too:
1) 11 … Qd7 12 f5 Bxg3 13 fxg6.
2) 11 … h6 12 f5 Bh7 13 Nh5.
3) 11 … f5 12 Bd3.
All of these are very good for White.
12 Bb3 Be4 13 Qh5! Qd7
13 … g6 leads to the destruction of the kingside after 14 Bxf7+! Kxf7 15 Qxh7+ Kf8
16 Nxe4 Rxe4 17 Qh8+ Ke7 18 Qg7+ Ke6 19 Qxg6. Meanwhile, 13 … Qc7 14 Re1 g6 15
Qh4 f5 16 Nxe4 Rxe4 17 Rxe4 fxe4 18 f5 leaves Black defenceless on the dark squares.
14 Nxe4 Rxe4 15 Bc2 g6 16 Qh4 Re8
 17 f5!
It is worth observing that 17 Qxf6 would have been a mistake, as 17 … Be7 18 Qe5
Bd8 is then equal.
17 … Be7
17 … g5 is taken care of by a not so surprising sacrifice: 18 Bxg5 fxg5 19 Qxg5+ Kh8
20 f6 Rg8 21 Qh6 Rg6 22 Bxg6 fxg6 23 f7.
18 Rf3!
Threatening Rh3. It is just so easy for Nunn to funnel pieces into the attack.
18 … g5 19 Qh5 Bd8 20 Bxg5! 1-0
20 … fxg5 21 f6 Bxf6 22 Rxf6 is game over.
Keith is a grandmaster of great ability, yet he never got into this game. He had one
chance only, on move ten, to rescue the situation. The average club player cannot really
handle such a thin line between a reasonable position and defeat.

I should not give you the idea that 5 … exf6 introduces a merely passive variation.
Quite often Black can advance his kingside pawn majority to create a strong attack.

Game 35
E.Torre-V.Korchnoi
Buenos Aires Olympiad 1978

1 e4 c6 2 d4 d5 3 Nc3 dxe4 4 Nxe4 Nf6 5 Nxf6+ exf6 6 Nf3 Bd6 7 Be2

 At the time Korchnoi played this game, the line with 5 … exf6 was all the rage.
Nothing could be found for White and new ideas were being introduced for Black, one of
which you are about to see. Basically Torre is relying on his queenside pawn majority to
give him the advantage. He develops without ambition and expects this advantage to
gradually give him control of the position. This sort of play is not incisive enough to
challenge Black.

7 … Na6!?
Interesting. The knight redeploys to c7, where it holds up the advance of the pawn
majority.
8 0-0
Clearly not 8 Bxa6? Qa5+ 9 Bd2 Qxa6, when White has given up the bishop pair for no
real compensation.
8 … Nc7 9 c4
Very typical play, but Black is developing easily enough.
9 … 0-0 10 Be3 Re8 11 Qd2?!
As with many games of that time, White is developing routinely. Torre seems
completely unaware that he is setting himself up to be attacked.
11 … Bf5 12 Rad1 Be4 13 Qc1 h6! 14 Nd2 Bh7 15 Bf3 f5 16 Nb3 g5!
 Introducing a situation that could hardly have been envisaged. The black pawns come
straight down on White’s head! It’s very difficult for Torre now. His bishops are targets for
the pawns and Korchnoi is picking up momentum move by move.
17 Na5?
Going out on a limb. White could play either 17 Rfe1 or 17 Nc5! instead, but he seems
genuinely unsettled. Possible lines are: 17 Rfe1 Qf6 18 Bd2 g4 19 Be2 Ne6 20 Bc3 Qg6
or 17 Nc5 Qc8.
17 … g4 18 Be2
Or 18 Nxb7 Bxh2+ 19 Kxh2 Qh4+ 20 Kg1 gxf3.
18 … Bb4!
An excellent move by Korchnoi, sacrificing the b7-pawn.
19 Nxb7
Korchnoi’s analysis shows that White has nothing better: 19 Bd2? Bxa5 wins, as does
19 Nb3 f4 20 Bd2 f3! 21 gxf3 Bd6! 22 Rfe1 gxf3 23 Bf1 (or 23 Bxf3 Qh4) 23 … Bxh2+!
24 Kxh2 Qh4+ 25 Kg1 Kh8 26 Rxe8+ Rxe8 27 Bf4 Rg8+ 28 Bg3 Rxg3+ 29 fxg3 Qxg3+
30 Kh1 f2! and mates.
19 … Qc8 20 Nc5 f4
 21 Bxg4?
Interestingly, computer analysis shows that after 21 Bd3!! fxe3 22 Bxh7+ Kxh7 23 fxe3
White might not be worse, as Black is now the one who is uncoordinated with his king
wide open; e.g. 23 … Kg7 24 Qc2 Rxe3 25 Qf2 Qe8 26 Nd3. Of course, a piece is a piece
and Korchnoi was famous for his defensive skills, so maybe Torre was intimidated.
21 … Qxg4 22 Bxf4 Ne6 23 Be3 Rad8 24 Nd3 Bd6
Three pawns for the piece, but it is the Black initiative that holds sway.
25 Ne5 Bxe5 26 dxe5 Bd3 27 f3
27 Rfe1 Be4 wins.
27 … Qxc4 28 Qxc4 Bxc4 29 Rxd8 Nxd8 30 Rd1 Bxa2 31 Bxa7 Bd5 32 f4 Ne6 33
Be3
Or 33 f5 Nf4 34 Re1 Nd3.
33 … h5 34 h3 Rb8 35 Rd2 Rb3 36 Kf2 h4 37 Rc2 Bxg2 0-1

Summarizing, there is no doubt that the line with 5 … exf6 is playable. Black must be
prepared for a long hard grind, but he must also be ready for tactics. The line we are about
to see represents the most critical test of the variation and brings us up to the present day.

Game 36
B.Savchenko-P.Maghsoodloo
Khomeyn 2017
1 e4 c6 2 d4 d5 3 Nc3 dxe4 4 Nxe4 Nf6 5 Nxf6+ exf6 6 c3 Bd6 7 Bd3 0-0 8 Ne2 Re8 9
Qc2 g6
This looks as though it helps White, but Black intends to block the b1-h7 diagonal by playing a subsequent … f6-f5.
10 h4 Be6
Black generally allows the h-pawn to advance.
11 h5 f5!

Black can play his knight to f8 via d7 and will develop counterplay on the queenside
via … Qa5 combined with the advance of his pawns. This is clearly going to produce a
very sharp position as White has hardly any alternative other than to castle long.
Savchenko has a novelty up his sleeve on move 14.
12 Be3
12 Qd2 Qf6 13 Qh6 is met relatively comfortably by 13 … Bf8!, as in R.Rapport-
I.Salgado Lopez, Camarinas 2013, which continued 14 Qf4 Nd7 15 Qg3 Bd5 16 Bf4 c5!
(creating necessary counterplay) 17 Bb5 Qb6! 18 hxg6 (White has straightened out the
pawn structure to create a tactical melee; Salgado is right on top of the complications as
18 Bxd7 is met by 18 … Rxe2+! 19 Kxe2 Qxb2+) 18 … fxg6 19 Qd3 Qe6 20 Qh3 Re7 21
Bg5 Rf7 22 0-0-0 Bxa2 23 Nf4 Qb6 (or 23 … Qe4 24 Bd3 Qc6 25 d5 Qd6) 24 c4 cxd4 25
Nd5 Qxb5! (a remarkable idea) 26 cxb5 Rc8+ 27 Nc3 (27 Kd2 Bxd5 leaves the white king
very exposed) 27 … dxc3 28 bxc3 Nc5 29 Bh6 Ne4 30 Kb2 Nxf2 31 Qh4 Nxd1+ 32 Rxd1
Bxh6 33 Qxh6 Bc4 34 Qh4 Bxb5 35 Rd8+ Rxd8 36 Qxd8+ Kg7 37 Qd4+ Rf6 38 Qxa7
Bc6 39 g3 g5 40 Qc5 Kg6 41 Qc4 f4. The position is better for Black now, though
Rapport drew in the end.
12 … Qf6
I prefer 12 … Nd7 here, keeping options open for the queen.
13 0-0-0 Nd7

14 c4!?
A new move, but does this encourage Black to play a move he wants to play anyway?
14 Nf4 is worth consideration, when 14 … Bd5 15 Nxd5 cxd5 16 Kb1 Rab8 17 Rh3 b5
18 Rdh1 Re7 19 g3 seems quite unclear to me, U.Atakisi-K.Delaney, European Cup,
Rhodes 2013. Perhaps Black should try 19 … b4 20 hxg6 fxg6 21 c4 h5!? with ongoing
complications.
14 … b5! 15 hxg6 hxg6 16 c5
White must close queenside lines.
16 … Bc7 17 Rh6 Qg7 18 Rh4
Sounding the retreat. 18 Rdh1 Nf6 leads nowhere for White, who faces … Ng4.
18 … Bd5?!
18 … Qf6 19 Rdh1 Bd5 20 Nc3 Bxg2 is a better way of doing things. After 21 Rh8+
Qxh8 22 Rxh8+ Kxh8 23 Bxf5 Nf8 24 Be4! Bxe4 25 Nxe4 Rad8, Black cannot be worse.
19 Nc3 Bxg2 20 Bxf5!
A successful capture, which shows the thin line that Black often walks after 6 c3. As we
know, with kings on opposite sides of the board, the relative speed of the attacks is
everything.
 20 … b4
20 … gxf5 21 Qxf5 Nf6 22 Rg1 is unthinkable for Black.

21 Nb5! cxb5 22 Bxd7 Qf6
22 … Reb8? defends b5 but in all other respects is a poor move. White now has a free
hand to attack; e.g. 23 Rg1 Bd5 24 Rh5! Bf3 25 Rf5 Rd8 26 Be6!! gxf5 27 Bxf7+ Kxf7 28
Qxf5+ Qf6 29 Qh7+ Ke6 30 Rg6+, winning.
23 Bxe8 Rxe8 24 Rg4 Bd5 25 Rdg1 Kg7 26 Rg5
Black is a clear exchange down and does not have enough in return.
26 … b3
He cannot take on a2 as 26 … Bxa2 27 b3 Qa6 28 d5 wins.
27 axb3 Be4 28 Qc3 Qa6 29 b4 Qa2
Or 29 … f6 30 R5g4 Bf5 31 d5! Bxg4 32 Rxg4 Be5 33 Qa3!, winning.
30 d5+ f6 31 d6 Bd8 32 Kd2
White refuses to be bluffed by the apparently active queen and converts easily.
32 … a5 33 Qd4 Kf7 34 R5g4 Bf5 35 Rf4 Re5 36 Ke2 Rd5 37 Ra1! Qxa1 38 Qxd5+
Be6 39 Qb7+ Kf8 40 Qa8 1-0
Nice control by Savchenko. It is unlikely that you will encounter this level of precision
at club level.

3 Nc3 Others
1 e4 c6 2 d4 d5 3 Nc3
We close the section on 3 Nc3 by looking at a couple of uncommon ideas. These lines
may be playable if you possess an open mind.

It has always been fascinating to me that one game by an influential player can affect
the entire popularity of an opening variation. Such is the case here, as Mikhail Tal puts a
sideline of the Caro-Kann to sleep for decades. Today, where anything can be expected in
the opening in order to surprise the opponent, 3 … b5 is beginning to be played a bit more
often.

Game 37
M.Tal-B.Gurgenidze
USSR Championship, Alma-Ata 1968

1 e4 c6 2 d4 d5 3 Nc3
If White does not want to think about what is to come, he may prefer 3 Nd2.
3 … b5!?
Here it is, and it’s really not that bad. Black’s idea is 4 cxd5 b4! and he also threatens
… b5-b4 directly.



4 a3
Tal pauses.
4 … dxe4
4 … a6 seems playable, as the recent game M.Kandic-L.Webb, Gibraltar 2017, shows.
Black lays down a challenge to White to ‘refute this rubbish’, and play continued 5 h3!?
dxe4 6 Nxe4 Nf6 7 Nxf6+ exf6 (we have already looked at a similar line; here, a few extra
pawn moves have been thrown in) 8 Nf3 Bd6 9 Bd3 0-0 10 0-0 Nd7 11 Re1 Nb6 12 Nh4
g6 13 Bh6 Re8 14 Rxe8+ Qxe8 15 Qf3 Nd5 16 c4 bxc4 17 Bxc4 Qd8 18 Re1 Be6 (Black
has emerged into the middlegame in reasonable shape; he has avoided any theory or
preparation the opponent may have made and goes on to win a long ending) 19 g4 Bf8 20
Bxf8 Kxf8 21 b3 Kg7 22 Qg3 a5 23 a4 Qb8 24 Qf3 Qd6 25 Ng2 h5 26 gxh5 Rh8 27 Qg3
Qxg3 28 fxg3 Rxh5 29 h4 Rh8 30 Rc1 Rb8 31 Ne1 Bf5 32 Kf2 Be4 33 Nf3 Rb4 34 Nd2
f5 35 Ke2 Nb6 36 Ke3 Nd5+ 37 Ke2 Kf6 38 Rf1 Ke7 39 Rc1 Nb6 40 Ke3 Bd5 41 Bxd5
Nxd5+ 42 Kd3 f4 43 Re1+ Kd7 44 Rf1 fxg3 45 Rxf7+ Ke6 46 Rf3 g2 47 Rg3 Kf7 48 Ke4
Nc3+ 49 Kd3 Nd5 50 Ke4 Nf6+ 51 Ke3 c5 52 Rxg2 Rxd4 53 Rg5 Nd5+ 54 Ke2 Nf4+ 55
Ke3 Ne6 56 Rxc5 Rxd2 57 Rxa5 Rh2 58 Ra7+ Kf6 59 b4 Rxh4 60 Rb7 Rh3+ 61 Kf2 Ra3
62 a5 g5 63 Rb6 g4 64 a6 Kf5 65 b5 g3+ 0-1.
5 Nxe4 Bf5
Possible is 5 … Nf6!? as seen in B.Belyakov-D.Howell, Doha 2016, where play
continued 6 Bd3 (I presume that after 6 Nxf6+ he was going to play 6 … exf6) 6 … g6 7
Bg5 Bg7 (7 … Qxd4!?) 8 Nf3 0-0 9 Qd2 Bf5 10 Nxf6+ exf6 11 Be3 Bxd3 12 Qxd3 Nd7
(Black is equal) 13 d5 c5 14 Qxb5 Rb8 15 Qa4 Rxb2 16 0-0 Nb6 17 Qxa7 Qxd5 18 Rab1
Ra8 19 Qxa8+ Qxa8 20 Rxb2 Nc4 21 Rb3 Nxe3 22 Rxe3 f5 23 Rfe1 Qa4 24 Re8+ Bf8 25
Rc8 Kg7 26 h4 Qxc2 27 Rd8 c4 28 Rc8 Bxa3 29 Rc7 Bd6 30 Rc6 Bb4 31 Re8 c3 32 Rec8
Qd1+ 33 Kh2 Bd6+ 34 g3 Qxf3 35 Rxd6 Qxf2+ 36 Kh3 Qf1+ 37 Kh2 Qe2+ 38 Kh3 Qe1
39 Kh2 f4 40 gxf4 Qxh4+ 41 Kg2 Qg4+ 0-1. We must not take this too seriously, as the
game was played in the World Blitz Championship, but it is an illustration of the ideas that
may be employed after 3 … b5.

6 Bd3 Bxe4?!
6 … Qxd4 must have seemed terribly risky to Gurgenidze and especially so against Tal,
but there is no reason why this should not have been played; e.g. 7 Nf3 (or 7 Ne2 Qb6 8
Be3 Qc7 9 0-0 Nd7 10 Nd4 Bg6 11 f4 e6 12 Qf3 Rc8) 7 … Qd5 8 Qe2 Nf6 9 Nxf6+ gxf6.
I think if you lay down the gauntlet with a move like 3 … b5!?, you should follow it
through. Gurgenidze shifts down gears and aims for a quiet position where he will tolerate
a small disadvantage, but this is not the way to play against a former world champion.
7 Bxe4 Nf6 8 Bd3
I quite like 8 Bf3! here, aiming at the black queenside.
8 … e6 9 Nf3 Be7 10 Qe2 Nbd7 11 0-0 0-0 12 Re1 Re8 13 Ne5 Nxe5 14 dxe5 Nd5 15
Qg4
White has the attack and two strong bishops, whereas Black has a well-placed knight
on d5 and precious little else to write home about. Tal is better.
15 … a5 16 h4!?
Tal offers a pawn to gain some time, a typical mechanism in his games. This ramps up
the entertainment value. He didn’t need to play riskily as 16 g3, intending h2-h4,
maintained his advantage without any trouble at all.
16 … Bxh4 17 g3 Be7 18 Kg2 g6
18 … Qc7 is an alternative.
19 Rh1 Bf8 20 Bg5 Qc7
 21 Rxh7!
Ouch!
21 … Qxe5
The attack flows smoothly after 21 … Bg7 22 Bf6!! Bxf6 (or 22 … Nxf6 23 exf6 Bxf6
24 Bxg6 Bg7 25 Rah1) 23 exf6 Nxf6 24 Qxg6+ fxg6 25 Rxc7 Rac8 26 Rxc8 Rxc8 27
Bxg6. Note that 21 … Kxh7 is impossible on account of 22 Qh4+ Kg8 23 Rh1.
22 Rxf7! Kxf7 23 Bxg6+!
The utter destruction of Black’s kingside has cost White a rook, but Gurgenidze will
never get to enjoy his extra material.
23 … Kg8
Or 23 … Kxg6 24 Bf4+ Kf6 25 Qh4+! Kf7 (25 … Kf5 26 Qg5+ Ke4 27 Qxe5 mate) 26
Bxe5, winning.
24 Bxe8 Bg7 25 Bd7 Nc7 26 Bxc6 Rf8 27 Rd1 Qc5 28 Bf3 Qxc2 29 Rd7 Rf7 30
Rd8+ Rf8 31 Bf6 Qh7 32 Be4 Qh6 33 Bg5 Qh8 34 Rd7 1-0
I do not think that the line with 3 … b5 is as bad as it was made to look in this game.
As an occasional surprise weapon and as a way to avoid all mainstream theory, it can be
recommended.

Game 38
V.Vladimirov-K.Stupak
Agzamov Memorial, Tashkent 2017
1 e4 c6 2 d4 d5 3 Nc3 a6!?
This should make the opponent sit up and take notice. Black waits before clarifying the
position in the centre. It is devilishly difficult to break down these light-squared
formations, and so White is invited to show how he is going to do it. The idea has
transpositional qualities, as Black may go into a French formation with … e7-e6 and …
c6-c5 at any moment. 3 … a6 looks like the move of a beginner but, as we will see, strong
players are getting involved.
4 Nf3
I guess White can go 4 e5, but then either 4 … Bf5 or 4 … e6 are perfectly playable for
Black.
4 … Bg4 5 h3 Bh5!
Black should not surrender the bishop at this point. Covering the light squares is
important.
6 g4 Bg6

7 Bd3
7 Ne5!? seems challenging, but Black can happily take on e4 with 7 … Bxe4 and now:
1) 8 Nxe4 dxe4 9 c3 e6 10 Qb3 Qc8 11 Bg2 was D.Mastrovasilis-D.Navara, Greek
Team Championship 2016. After 11 … Nd7 12 Bxe4 Ngf6 13 Bg2 Nxe5 14 dxe5 Nd7 15
Bf4 Be7, the position was equal.
2) 8 f3!? can lead directly to 8 … Bg6 (I can offer the weird 8 … f6!? as a possible
improvement: 9 fxe4 fxe5 10 dxe5 e6 or 9 Nxc6 Nxc6 10 fxe4 dxe4 11 d5 Ne5) 9 h4 f6 10
Nxg6 hxg6 11 Qd3 Kf7 12 h5 gxh5 13 gxh5 e6. The black king is safer than it looks and
while White has compensation for the pawn, he still has to prove it. Perhaps 14 Qg6+ Ke7
15 Bh3 is the right way forward.
7 … e6 8 Ne5 Bb4!
Black has also played 8 … Nf6 with success in Z.Borosova-P.Lizak, Kecskemet 2012:
9 Nxg6 hxg6 10 e5 Nfd7 11 Ne2 c5 12 c3 Nc6 13 a3 cxd4 14 cxd4 Qb6 15 Bc2 Be7 16
Qd3 Qa5+ 17 Kf1 Rc8 18 Rb1 Qb5 19 Qxb5 axb5 20 Bd3 b4 21 Bd2 Nb6 22 axb4 Bxb4
23 Bc3 g5 24 Bb5 Be7 25 Ra1 0-0 26 Kg2 Ra8 27 Rhc1 Rxa1 28 Rxa1 Ra8 29 Rxa8+
Nxa8 30 Bxc6 bxc6 31 Ba5 Kf8 32 f4 gxf4 33 Nxf4 c5 34 dxc5 Bxc5 35 Nd3 Be7 36 Kg3
g5 37 b4 Ke8 38 Kf3 0-1.
9 Nxg6 hxg6 10 a3 Bxc3+ 11 bxc3 Nf6 12 e5 Nfd7

The opening has been a success for Stupak. White has the bishop pair but they are not
really a match for the knights in the current blocked position. It remains for Black to start
putting pressure on d4 with the usual plan of … c6-c5 and … Nc6.
13 a4
After 13 c4 dxc4 14 Bxc4 c5! White’s centre is looking very shaky; e.g. 15 c3 Nc6 16
Be3 Qa5.
13 … c5 14 Qf3 Nc6 15 Rb1 Qa5 16 Bd2 Qxa4 17 Rxb7 cxd4
White’s centre is demolished and he is forced to castle into fresh air to try to scramble
out of the mess he is in.
18 0-0 Ndxe5 19 Qg3 0-0 20 cxd4 Nxd3 21 cxd3
Or 21 Qxd3 Qxd4.
 21 … Rac8 22 h4 Nxd4 23 Kg2 Rc2
This is simple fare. Black is material ahead with the more active pieces. He is also
much higher rated and it is no surprise the game is soon over.
24 Bb4 Qc6
24 … Ne2! 25 Qd6 Rfc8 26 Qe7 Qe8 was another possibility, with the white bishop
playing as an onlooker.
25 Rb8 Rxb8 26 Qxb8+ Qc8 27 Qf4 e5 28 Qxe5 Qxg4+ 29 Qg3 Qf5 30 Bd6 Rd2 31
Qh3 Rxd3 32 Qxf5 Nxf5 33 Bg3 Rxg3+ 0-1
I think you can make 3 … a6 work for you with a bit of prior study. Among masters,
Kirill Stupak has played it most frequently, so seeking out his games would be a good
starting point.

 Chapter Three
The Advance Variation

1 e4 c6 2 d4 d5 3 e5
This is the Advance Variation, a very important main line. I always thought that 3 e5
played into Black’s hands, as the white central pawns can often become fixed and rigid,
but the line has become infused with new ideas recently. Black can do many things, but he
should choose either 3 … Bf5 or 3 … c5. We will take a brief look at irregular ideas at the
end of the chapter.
We will start by examining the lines after 3 … Bf5, which I recommend as best.

Game 39
A.Shirov-V.Anand
Leon (rapid) 2011

1 e4 c6 2 d4 d5 3 e5 Bf5 4 g4!?
A questionable idea, yet very direct and aggressive. Black can be caught napping if he
gets the next move wrong.

4 … Bd7!
Showing flexibility of thought. Black may turn the game into something resembling the
French Defence with … e7-e6, following up with … c6-c5, … Nb8-c6, … Qb6 etc. By
playing this way, Black tries to make 4 g4 look out of place.
After 4 … Bg6, 5 e6!? could be an idea. The pawn on e6 is selling its life in return for
hampering Black’s kingside development. If Black captures the pawn, he faces an uphill
task to get his pieces out. Possibly 5 … Qd6 is the best reply, but even there 6 exf7+ Bxf7
7 f4! Nd7 8 Nf3 gives White a grip on the position.
Another try after 4 … Bg6 is 5 h4 h5 6 e6, with 6 … fxe6 (again, 6 … Qd6 is safer) 7
Bd3 Bf7 8 g5 leading to another kingside traffic jam for Black.
5 c4 e6 6 Nc3 c5!!
A brilliant new move from Anand. Although Black was doing well after the alternative,
6 … Ne7, the central break 6 … c5 is far more dynamic. Lines are opened by force and the
pawn on g4 starts to look very poorly placed indeed.
7 cxd5 exd5 8 dxc5 Bxc5

9 Bg2?!
Shirov could have captured on d5 and maybe should have done, but you can understand
why he might not trust his very shaky position. Instead, he tries to stabilize, but runs into
Anand on top form. After 9 Qxd5 Qb6 10 Bc4 Be6 (10 … Bxf2+ is also possible, but
leads to nothing decisive after 11 Ke2 Be6 12 Qb5+ Nc6 13 Bxe6 fxe6 14 Nf3) 11 Bb5+!
Nc6 12 Bxc6+ bxc6 13 Qf3, White manages to protect both f2 and g4, but Black’s chances
are to be preferred due to his superior development, the bishop pair, and White’s very
ragged pawn structure.
9 … Ne7 10 h3 Qb6 11 Qe2 0-0
Black’s development flows easily. It is hard to see where White is going to place his
king safely.
12 Nf3 d4!

TIP: When you are ahead in development – attack!

13 Ne4 Bb5 14 Qd2 Nbc6!

Let’s draw up a balance sheet:

1) King position. Black’s king is completely safe but White’s king is vulnerable in the
centre.
2) Piece Activity. Black’s development is almost complete and he threatens … Bb4.
White’s development, on the other hand, is woeful.
3) Material Balance. White is not even material ahead to compensate him for his
horrible position. Looking at it this way we can understand that Shirov cannot last much
longer.
15 a3 Ng6 16 b4 Be7 17 Bb2 Rfd8 0-1
White sees no hope, despite a full board of pieces. A club player might carry on, but it
is very hard to see a good move, as Black is just going to take on e5, threatening … Nd3+.
The Caro player should be delighted to see 4 g4 on the board, which is just a little bit too
aggressive for White’s own good.

Game 40
A.Motylev-V.Anand
Russia vs. Rest of the World, Moscow (rapid) 2002

1 e4 c6 2 d4 d5 3 e5 Bf5 4 Nc3 e6 5 g4
If White must play with an early g2-g4, then this is a better moment to do it.
5 … Bg6 6 Nge2!

6 Nge2 introduces a very dangerous attacking system. Black must be careful against
this one. White intends a kingside pawn storm, possibly throwing in Nf4 sometime soon.
He makes the bishop on g6 a prime target. As in the first game of this chapter, we will sit
alongside Vishy Anand and let him show us the way.
6 … c5!
Black has tried many moves, but 6 … c5 is generally recognized as the best. With
White limbering up for a big kingside knockout, a counter in the centre seems right.
Classic chess strategy.
7 Be3 Nc6 8 dxc5 a6!
Anand had prepared this move in advance. Obviously Black can take on e5 instead, but
the idea is that he should always be able to do so, and thus taking away the use of b5 is a
higher priority.
9 Nd4
If White tries to punish his opponent by protecting the e5-pawn, he could be in for a
shock, as in O.Hofman-A.Torres Sanchez, correspondence 1998: 9 f4 Qh4+! 10 Bf2 Qxg4
(it is clear that White has to be very careful now; he should try 11 Qd2, but instead he
falters) 11 Rg1? Qf5 12 Ng3 Qxf4 13 Nce2 Qb4+ 14 Nc3 Qxb2 15 Qb1? (an
extraordinarily awful move for a correspondence game) 15 … Qxc3+ 16 Kd1 Bxc2+ 17
Qxc2 Qxa1+ 18 Ke2 Nd4+ 19 Bxd4 Qxd4 20 Rg2 Bxc5 21 Qa4+ Qxa4 22 Kf3 Qa3+ 23
Kg4 h5+ 24 Kh4 g5+ 25 Kxg5 Qe3+ 26 Kh4 Qf4+ 27 Kh3 Qg4 mate. A terrible game of
course, but entertaining. It is definitely a reminder to White that in this variation, he can
expect Black to fight for the initiative.
9 … h5!
I like Anand’s aggressive approach. 9 … h5 punctures a hole in White’s kingside set-
up.
10 Nxc6 bxc6 11 g5
Not a pleasant move to have to make, as Black now has full control of the key f5-
square. The white position is already starting to look ragged.
11 … Ne7 12 Bh3 Rb8

13 Qd4?
Like it or not, White has to play 13 b3 to give himself any chance of equality. Black is
still a bit better after 13 … Bf5, but White might be able to hold it together.
13 … Bf5! 14 Bg2 Bxc2
The time has come to pick off some weak pawns.
15 Qd2 Bg6 16 Na4 Nf5 17 0-0 Be7 18 f4 Nxe3 19 Qxe3 Qa5 20 b3 Rb4 21 Kh1
White seems to have recuperated well and has parked his king in a safe place, yet Black
is still much better. Let’s detail his advantages:
1) He has the superior pawn structure.
 2) He has more active pieces.
3) He has the bishop pair.
4) After castling, he will have the safer king.
With all the positive aspects of the position operating in Anand’s favour, it will be hard
for Motylev to save the game.
21 … 0-0! 22 Rac1 Rfb8 23 f5
Pure desperation. Motylev sees Anand’s game improving move by move while White’s
position is just getting worse. In that situation, you might as well try something.
23 … Bxf5 24 Rxf5 exf5 25 e6 Qc7
25 … Qd8 26 exf7+ Kf8 27 Qe5 Bxg5 is also winning for Black.
26 exf7+ Kxf7 27 Re1 Re4!

Very nice. By returning the exchange, Anand completely blocks Motylev’s attack and
sets up a powerful central pawn chain.
28 Bxe4 fxe4 29 Qh3 g6 30 Nb6
30 Rf1+ Kg7 31 Qe6 Bxg5 leaves White with no good way to continue.
30 … Rd8 31 Rf1+ Kg7 32 Qe6 Bxg5 33 Rg1 Bf6 34 Rf1 Qe7 35 Qxc6 e3 36 Nxd5
e2 0-1
Touchdown approaches. After 37 Re1, the reply 37 … Qe4+ is terminal. Very nice play
by Anand.
The sharp system 5 g4 and 6 Nge2 has accumulated a lot of theory over the past 30
years, yet I don’t think you need to know that much as Black. Firstly, you have to get into
the right frame of mind: you must come out and fight! Secondly, you must form the view
(probably correctly) that all these systems with an early g2-g4 are inferior and that a well-
timed counter in the centre will always give Black a good game.

TIP: Positive thinking is essential in chess. Believe that you are playing good moves

and executing good plans.

Game 41
V.Onischuk-R.Dimitrov
Corsica 2017

1 e4 c6 2 d4 d5 3 e5 Bf5 4 h4

4 h4 introduces one of White’s main weapons in the Advance Variation. With the centre
blocked, White begins aggressive flank operations and gains some space. He may make
the bishop on f5 a target, while reserving the alternative option of playing in the centre
with c2-c4.
There are quite a few dangers that Black has to negotiate. 4 Ne2 is another interesting
move, a favourite of Boris Spassky in the dim and distant past. The game A.Kaczmarek-
M.Bartel, Suwalki 2017, continued 4 … e6 5 Ng3 Bg6 6 h4 h5! (the best move, throwing
a spanner in the works; after 6 … h6 Spassky patented 7 h5 Bh7 8 Bd3 Bxd3 9 cxd3!?,
when White’s plan is to get a kingside attack via moves such as 0-0, Qg4, f4-f5 and he
may even sacrifice the b2-pawn in order to achieve this; e.g. 9 … Qb6 10 Be3! Qxb2 11
Nd2 and the scene is set for strong white pressure; I don’t think a pawn is enough to make
this prospect attractive to Black) 7 Ne2 (7 c3 c5 8 Bb5+ Nc6 9 Be3 Qb6 10 a4 is okay for
White, but not very inspiring) 7 … c5! 8 Nf4 Bf5 9 Nxh5 Nc6 (White has wasted a lot of
time picking off the h5-pawn; meanwhile, Black counterattacks in the centre and I already
prefer his position) 10 Ng3 Rxh4 11 Rxh4 Qxh4 12 Bb5 (White can take on f5, but then
the whole centre collapses: 12 Nxf5 exf5 13 c3 cxd4 14 cxd4 Qxd4 15 Qxd4 Nxd4 16 Bd3
Bb4+ 17 Nc3 Ne7 and Black holds the advantage) 12 … cxd4 13 Nd2 Bg4 14 Nf3 Bxf3
15 Qxf3 Nge7 16 Bd2 a6 17 Be2 Ng6 18 Qb3 0-0-0 (with relatively simple moves, Black
has refuted White’s odd opening play) 19 Qa4 Ngxe5 20 b3 Kb8 21 Rb1 Bd6 22 Nf1 Qe4
23 b4 d3 24 Ng3 dxc2 25 Rc1 Nd3+ 26 Kf1 Bxg3 0-1.
If Black meets 4 Ne2 carelessly he can get into hot water. Fortunately for him, 4 … e6
5 Ng3 Bg6 6 h4 h5! just seems strong.
4 … h6
Botvinnik’s favourite long ago. 4 … h6 is a good move which does not make any
positional concessions.
4 … h5 is a reliable and more common alternative, which has the drawback of letting
White use the g5-square.

Play can continue:

1) 5 Bd3!? is the modern reply, playing only for a very small edge. White argues that he
owns a bit more space and that will be enough. A.Morozevich-I.Rozum, Sochi (rapid)
2016, now saw 5 … Bxd3 6 Qxd3 e6 7 Bg5 Be7 (I prefer 7 … Qb6!) 8 Nf3! Nh6 9 Nc3
Nf5 10 0-0-0 Nd7 11 Kb1 Rc8 12 g3 b5 13 Ne2 b4 14 Nf4 g6 15 Bxe7 Qxe7 16 Qa6 Nb6
17 Nd3 0-0 18 Rhg1 Nc4 19 g4 (White changes gear, increasing the tempo of the game;
we now see the realization of the opening edge with Black under serious attack) 19 …
Nxh4 20 Nxh4 Qxh4 21 gxh5 Qxh5 22 Qxa7 Rb8 23 Qe7 Qf5 24 Rg3 Rfe8 25 Qf6 Qxf6
26 exf6 Nd6 27 f3 Rec8 28 Ne5 Ne8 29 Ng4 c5 30 Rh1 Rb7 31 Rgh3 (the attack has
continued, despite the queen exchange) 31 … Nxf6 32 Rh8+ Kg7 33 Rxc8 cxd4 34 Ne5
Nh5 35 Nc6 Kf6 36 Nxd4 Ke5 37 Nc6+ Kf4 38 Rh4+ Kxf3 39 Rxb4 Rxb4 40 Nxb4 g5 41
Nd3 1-0. It is hard to believe that 5 Bd3 gives any sort of advantage at all, although results
have been good. Black must show discretion as to which pieces to exchange and which
not.
2) 5 c4 has been pretty popular over the years, but with a well-timed capture on c4,
Black can control the central light squares, which should see him through. For example,
the game P.Ponkratov-A.Shimanov, St. Petersburg 2017, continued 5 … e6 6 Nc3 Ne7 7
Nge2 Nd7 8 Ng3 Bg6 9 a3?! (I prefer 9 Bg5 Qb6 10 Qd2 from White’s perspective, but
again it is … d5xc4 that will take the pressure off: 10 … dxc4! 11 Bxc4 0-0-0, when Black
threatens to take on e5 and stands very well indeed) 9 … dxc4! 10 Bxc4 Nf5 11 Nge2?
(far too ambitious; White should settle for 11 Nxf5 Bxf5 12 Be3 and an equal game) 11 …
Nb6 12 Ba2 Nxh4 13 Nf4 Bf5 14 Kf1 g6 15 Be3 Be7 16 Rc1 Rc8 17 f3 g5! (a confident
move which pushes White back; throughout this game Ponkratov is plagued by his
inability to get his pieces working together) 18 Nfe2 Nd5 19 Bf2 g4 20 Nxd5 exd5 21 Rc3
gxf3 22 gxf3 Ng6 23 Ng3 Be6 24 Rxh5 Kd7 25 Qb3 Kc7 26 Bb1 Nf4 27 Rf5 Bxf5 28
Bxf5 Ne6 29 Bg4 Bh4 30 Qb4 Kb8 31 Ke2 a5 32 Qa4 Qb6 33 Rb3 Qa6+ 34 Kd2 b5 35
Rb4 Ka7 0-1.
5 g4
This is what they do now. Black should drop back to d7 to avoid any e5-e6 issues.
5 … Bd7! 6 c4 Be6!?
 A new and very interesting move, designed to entice White forward. 6 … e6 7 Nc3 c5!
is another good way to go, with the idea 8 cxd5 exd5 9 Nxd5?? Bc6 and wins, while 9 Be3
cxd4 10 Qxd4 Ne7 11 Nf3 Nbc6 is rather unclear.
7 c5
Otherwise, Black will get his usual control of d5.
7 … Qc8
7 … b6 8 b4 a5 is very sharp and White may be down to 9 b5 bxc5 10 dxc5 d4 11 Bb2
just to keep himself on the board. What an odd position!
8 Be2 h5
Black should again play 8 … b6!.
9 g5 g6 10 Nc3 Bg7
The bishop is looking at a brick wall, so one would assume Black is going to try to
break down the wall by playing a subsequent … f7-f6.
11 Be3 Nd7 12 b4 f6 13 f4 fxe5 14 fxe5 Bf5 15 Bd3 Nf8 16 Bxf5 Qxf5 17 Nge2 Qd7
18 b5 e6 19 bxc6 bxc6 20 0-0
 Extra space really does mean something on the chessboard and Black has not really
timed his attack on the white pawn chain that well. He can manoeuvre a knight to f5, but
in every respect his position is passive.
20 … Ne7 21 Nf4 Nf5 22 Bf2 Kf7 23 Qa4 Qc7 24 Rab1 Nd7 25 Qa6 Rab8 26 Nb5!
Onischuk has found a very nice way of increasing his advantage.
26 … Qc8
26 … Rxb5 27 Rxb5 cxb5 28 Qxe6+ Kf8 29 Nxg6 mate is not what the doctor ordered.
27 Qxc8 Rhxc8 28 Nd6+ Nxd6 29 cxd6 Rxb1
Black’s king is in the wrong place. He would like to play 29 … c5, but after 30 Rxb8!
Rxb8 (or 30 … Nxb8 31 Rb1! cxd4 32 Rb7+ Kg8 33 Bxd4) 31 dxc5 Nxe5 (or 31 … Bxe5
32 Nd3 Ke8 33 Re1) 32 Bg3, a crushing discovered check may follow.
30 Rxb1 Rb8 31 Rxb8 Nxb8 32 Nd3
Normally, one would classify Black’s bishop as the better of the two, but the bishop on
g7 is completely walled in, which is why White is winning.
32 … Nd7 33 Bg3 Ke8 34 Kf2 Nb6 35 Nc5 Kf7 36 Ke2 Bf8 37 Be1 Nc4 38 d7 Be7 39 Nb7 c5 40 Bf2 1-0

Game 42
A.Oberhofer-R.Dautov
German Bundesliga 2017

1 e4 c6 2 d4 d5 3 e5 Bf5 4 Bd3
4 Bd3 is completely harmless, but you do see it from time to time. It looks as if White
is changing his mind, launching complications after 3 e5 and then immediately retracting
by offering Black easy exchanges.
 4 … Bxd3 5 Qxd3 e6 6 Nf3
Giving up any hope of advantage. Grandmaster Dautov now plays an idea which I
believe dates back to the days of Capablanca. Slightly more complexity is offered after 6
Qg3.
6 … Qa5+ 7 Bd2 Qa6!

By exchanging queens, Black nullifies White’s tiny edge.
8 b4?!
I’m not sure I understand what is going through White’s mind when he plays a move
like this. It may not be an outright blunder, but voluntarily inheriting doubled pawns is
very risky. Positionally, White is trying to stop … c6-c5, but Black can also lever open the
position with … f7-f6.
8 Qxa6 Nxa6 9 a3 Rc8 10 b4 Ne7 11 0-0 Nf5 12 g3 is dull but would be better than
what happens in the game.
8 … Qxd3 9 cxd3 Nd7 10 0-0 f6! 11 a3 Ne7 12 exf6 gxf6 13 Nc3 Nf5 14 Rfb1 Bd6 15
Na4
I must admit that I thought the idea was 15 a4 Kf7 16 b5, but then 16 … c5! is just very
good, despite allowing White to undouble his pawns; e.g. 17 dxc5 Nxc5 18 d4 Ne4 19 Be1
Nxc3 20 Bxc3 Rac8 21 Rb3 Rc4.
15 … b6 16 Nc3 Kf7 17 Ne2 Rac8 18 a4 h5 19 h4 e5
Grandmasters do not fear a queenless middlegame as there are usually plenty of
opportunities to outplay a lower-ranked opponent.

TIP: Exchanging queens early gives you the opportunity to test your technique.
If you find yourself avoiding these positions, you need to study strategy more closely.

20 g3 exd4 21 Nexd4 Nxd4 22 Nxd4 Be5 23 Nf5 Ke6 24 Ng7+ Kf7 25 Nf5

25 … Bxa1!
Black definitely should not repeat. We come to an ending where Dautov’s king
becomes very active and this proves to be the decisive factor.
26 Nd6+ Kg6 27 Nxc8 Rxc8 28 Rxa1 Ne5 29 Ra3 Nf3+ 30 Kf1 Nxd2+ 31 Ke2
Interestingly, Black cannot save the knight, but he can use the time that White spends
regaining the piece to improve his position.
31 … Nb1 32 Rb3 Re8+ 33 Kd1 Kf5 34 Rxb1 Kg4! 35 Rc1 Re6 36 d4 Kf3 37 Rc2
a6
White can do nothing. If he waits, then … f5-f4 will smash up his pawns. If he acts, his
position will deteriorate further. He chooses the latter course.
38 b5 cxb5 39 axb5 axb5 40 Rb2 b4 41 Rxb4 Kxf2 42 Rb5 Rd6 43 Rb3 f5 44 Kd2
b5! 45 Rxb5 Kxg3 46 Rb8 f4 47 Rg8+ Kxh4 48 Rg7 Re6 49 Rd7 f3 50 Rxd5 f2 51 Rf5
Kg3 52 d5 Re1 0-1

TIP: The worst way to try to get a draw against a stronger opponent is to exchange

pieces and to play a simple position. White did not create any pressure or manufacture
any tactics to worry Dautov and that was his downfall.

Game 43
I.Nepomniachtchi-V.Laznicka
European Team Championship, Heraklio 2017

1 e4 c6 2 d4 d5 3 e5 Bf5 4 Nf3 e6 5 Be2

The Short Variation, named after grandmaster Nigel Short. In 2017 this is probably
White’s most popular choice after 3 e5 Bf5. Outwardly unpretentious, White is
concentrating on development for the time being and waiting to see what will turn up. He
is confident that his small advantage in space will be enough to secure an edge. He will
castle and then decide on a concrete course of action.

5 … c5
Logical, but by no means the only move.
5 … Ne7 can lead to some very obscure positions. A.Shirov-H.Nakamura, Douglas
2017, continued 6 0-0 (6 c3 may be played first; e.g. 6 … Nd7 7 Nbd2 c5 8 dxc5 Nxc5 9
0-0 Nc6 10 Nd4 Bg6 11 Nxc6 bxc6 12 b4 Nd3 13 Nf3 Nxc1 14 Rxc1 Be7 was equal in
S.Karjakin-V.Anand, St. Louis blitz 2017) 6 … h6!? 7 b4 Nd7 8 a4 g5 (flank operations
seem justified with the centre blocked, but this is nevertheless a very strange position) 9
a5 Bg7 10 Nbd2 0-0 11 Nb3 f6 (Nakamura had to find a freeing pawn break somewhere or
the queenside would have been opened) 12 exf6 Bxf6 13 Nc5 Nxc5 14 bxc5 Ng6 15 Be3
Qc7 16 c4 a6 17 cxd5 exd5 18 Bd3 Bxd3 19 Qxd3 Kg7 20 Rfe1 Rae8 21 h3 Qd7 22 Bd2
Rxe1+ 23 Rxe1 Re8 24 Rxe8 Qxe8 25 Qf5 Ne7 26 Qg4 Qc8 27 Qh5 Qf5 28 Qe8 Qc8 29
Qh5 Qf5 30 Qe8 Qc8 ½-½.
6 Be3!?
Development takes priority! 6 0-0 was Nigel’s original thought, but I very much like
the way Yasser Seirawan handled the black position in N.Short-Y.Seirawan, Euwe
Memorial, Amsterdam 1992: 6 … Nc6 7 c3 cxd4 8 cxd4 Nge7 9 b3 Be4! (very good;
Black is playing in thematic Caro style, aiming at undermining the d4-pawn with help of
moves such as … Nf5 and … Qb6) 10 Nbd2 Nf5 11 Bb2 Bb4 12 g4 Nh4 13 Nxe4 dxe4 14
Nxh4 Qxh4 15 Qc2 h5! 16 Qxe4 (16 gxh5 0-0 sees the black position coming together; the
rooks will come to the centre with a fine game) 16 … hxg4 17 Qg2 0-0-0 18 Bxg4 Kb8 19
Rfd1. Now, with 19 … Ba5 20 Rac1 Bb6, Black keeps the position in balance.
6 … cxd4
Black takes the opportunity to break up the white centre.
7 Nxd4 Ne7 8 0-0 Nbc6 9 Bb5 a6 10 Bxc6+ bxc6

11 c4
If White does not open the position right away, Black will gradually get around to
castling and have a slight edge thanks to his more compact pawn structure.
11 … Qd7 12 Nc3 dxc4!
 A surprising move, but Black plans to entrench a knight on d5. It would have been
much more risky to play a move like 12 … Rb8 as White’s superior development would
come to the fore; e.g. 13 Na4! Ng6 14 Nxf5 exf5 15 cxd5 Qxd5 16 Qxd5 cxd5 17 Rfd1.
From this small variation we see one of the main attractions of this line from White’s
perspective. He has castled and Black has not.
13 Na4 Nd5 14 Nxf5 exf5 15 Rc1 c3 16 Qc2
16 Nxc3 Nxe3 17 Qxd7+ Kxd7 18 fxe3 Ke6 leads nowhere for White after 19 Ne2 (or
19 e4 Bc5+ 20 Kh1 f4! 21 Ne2 Be3 22 Rxc6+ Kxe5) 19 … c5 20 b4 Kxe5 21 bxc5 Rc8.
16 … Be7 17 Nxc3 Nxe3 18 fxe3 Bg5
18 … g6 first was more circumspect.
19 Rce1 g6

20 g4!?
No expense is spared to open up the game.
20 … fxg4 21 Ne4 Be7 22 Rd1 Qc8 23 e6
More of the same.
23 … fxe6??
23 … Qxe6 required a bit of calculation but turns out satisfactorily; e.g. 24 Nd6+ Bxd6
25 Qxc6+ Ke7 26 Rxd6 Qxd6 27 Qb7+! (27 Rxf7+ Kxf7 28 Qxd6 h5 is about equal) 27
… Qd7 28 Rxf7+ Kxf7 29 Qxd7+ Kf6 30 Qxg4 Rhd8 and White should not win this.
24 Qc3!
Maybe Black overlooked this. At any rate, 24 Qc3 is suddenly crushing.
 24 … Rf8 25 Rxf8+ Bxf8 26 Qh8 1-0
With 4 Nf3, 5 Be2 and quick castling, White argues that development really does mean
something and that he is setting up a platform from which to attack. I think White is right,
yet the lack of immediate pressure should give Black time to sort himself out. There was
no need for Laznicka to lose this game.

Game 44
M.Tal-M.Botvinnik
World Championship (Game 8), Moscow 1961

1 e4 c6 2 d4 d5 3 e5 c5
3 … c5 is usually chosen by aggressive players who want confrontation. Black tries to
turn the game into an improved French Defence where he can get all the advantages of
that opening and develop his bishop on c8 too. Today 3 … c5 is very popular, although I
am convinced that 3 … Bf5 is more correct. We start our journey with a classic game.
4 dxc5 e6
4 … Nc6 is a favourite of English GM Keith Arkell, although 5 Bb5! (for 5 Nf3 Bg4,
see the next game) seems better for White; for example, 5 … Qa5+ 6 Nc3 e6 7 Be3 Bd7 8
Nf3! (stronger than 8 Bxc6 Bxc6 9 Qd4 Ne7 10 g4 h5) 8 … a6 (8 … Nge7 is met by 9
a3!) 9 Bxc6 Bxc6 10 a3 Qc7 11 0-0 and, despite his two bishops, Black has yet to regain
his pawn.
5 Qg4!?
In keeping with his superb attacking style, Tal deploys his queen so as to tie down the
bishop on f8. Will the white queen become a target?




5 … Nc6 6 Nf3 Qc7 7 Bb5
This is a key move in many of the variations after 3 … c5, making it as awkward as
possible for Black to get his pawn back.
7 … Bd7 8 Bxc6 Qxc6 9 Be3 Nh6 10 Bxh6
Tal understands that he cannot keep the pawn forever and sets up a very unbalanced
position.
10 … gxh6 11 Nbd2 Qxc5 12 c4
3 … c5 certainly produces a lot of interesting positions and this is one of them. Black
has a potentially powerful bishop pair and a possible strong attack along the g-file, but
will he ever get to utilize these strengths? The problem is that there is no real safe haven
for the black king and with 12 c4! Tal shows that he understands he needs to open the
position immediately.
12 … 0-0-0 13 0-0 Kb8 14 Rfd1
14 Qf4! might have been a bit better, as Black then struggles to defend f7.
14 … Qb6 15 Qh4
 15 … a5?!
Trying to restrain White on the queenside and secure the c5-square, but one cannot help
but feel that this plays into Tal’s hands. 15 … Rg8 was more active.
16 Rac1 Rg8 17 Nb3 a4 18 c5! Qc7 19 Nbd4 Rc8 20 b4
So much for restraint.
20 … axb3 21 axb3 Qd8 22 Qxd8 Rxd8 23 b4
White is now much better. Black’s bishops are passive and Tal still has a big queenside
attack, despite the exchange of queens.
23 … Rg4 24 b5 Rc8 25 c6 Be8 26 Rc2 Bg7 27 Ra1 Bxe5 28 Nxe5 Rxd4 29 Nd7+ 1-
0
Games such as this one did nothing for the early popularity of 3 … c5.

Bringing our investigation into the present day, let us demonstrate some of the positive
sides to 3 … c5.

Game 45
D.Vocaturo-P.Prohaszka
Benasque 2017

1 e4 c6 2 d4 d5 3 e5 c5 4 dxc5 Nc6 5 Nf3

As stated before, 5 Bb5! is better.
5 … Bg4! 6 c3 e6
 If Black is allowed to play … Bxc5 in peace he will immediately have the better game
and so White must simply stop or at least delay that.

7 Be3
7 b4 looks critical, as White establishes a dangerous queenside pawn majority. Black
must concentrate on pointed development and the e5-pawn. V.Topalov-H.Nakamura,
London Classic 2016, continued 7 … a6! (I prefer this to 7 … Bxf3 8 Qxf3 Nxe5 9 Bb5+
Nc6, which we saw back in Game 2; and definitely not 7 … Nxe5? 8 Qa4+! Nd7 9 Ne5
Ngf6 10 Bb5, which is an immediate trap) 8 Nbd2 (8 Qe2 cannot hold on to the e5-pawn
for long; e.g. 8 … Nge7 9 Nbd2 Ng6 10 h3 Bxf3 11 Nxf3 Qc7) 8 … Nxe5! (an important
difference; Black has effectively gained a tempo on the last fragment of analysis and can
negotiate the tactics) 9 Qa4+ Nd7 10 Ne5 Ngf6 11 c4 a5! 12 Nb3 axb4 13 Qb5 Be7 14 c6
bxc6 15 Nxc6 Qc7 16 f3 Bf5 17 Nxe7 Rb8! 18 Nxf5 Rxb5 19 Nxg7+ Ke7 20 cxb5 Nc5 21
Bb2 Nxb3 22 axb3 Qf4 (22 … Rg8 23 Bd4 Qf4 24 Bc5+ Kd7 25 Be2 Rxg7 26 0-0 Rg8 27
b6 looks dangerous as the b6-pawn is very powerful, while 23 … Qc2 might lead to a
draw after 24 Ra7+ Nd7 25 b6 Qc1+ 26 Ke2 Qc2+) 23 Be2 Rc8 24 Rd1 Qg5 25 b6 Rc2
26 Bxf6+ Qxf6 27 Nh5 (27 b7 is met by 27 … Qc3+ 28 Kf1 Qc7) 27 … Qc3+ 28 Kf1 Qe3
29 Re1 Qxb6 30 Nf4 Qe3 31 g3 Qxb3 32 Kg2 Kf8 33 Kh3 Qb2 34 Rb1 Qf6 35 Rhe1 e5
36 Nxd5 Qe6+ 0-1.
7 … Nge7
 I guess Black can play 7 … Bxf3 8 Qxf3 Nxe5 9 Bb5+ Nc6 10 0-0, but White would be
pleased to see this with two bishops and the queenside majority.
8 Nbd2 Nf5
Black stations his pieces as actively as he can.
9 Bd4 Be7
9 … Qc7 10 Bd3 does not really further Black’s cause.
10 Qa4 Bh5 11 Bd3 Bg6 12 Bc2 0-0 13 0-0 Qc7 14 Rac1 b6!
Note that Black has still not regained his pawn, so he forcibly opens lines on the
queenside, trying to exploit the awkwardly placed white pieces.

15 b4
White does not wish to play 15 cxb6 axb6 16 Qb5 Rxa2.
15 … Rfb8! 16 Bb1 bxc5 17 bxc5 a5
The co-ordination of his forces is Vocaturo’s main problem. He has an overgrown pawn
on d4 that is preventing him from getting any active play, and this means that Black can
handle the position calmly, just building up.
18 Qd1 Qa7 19 h3 Bxc5
Finally, Black gets his pawn back and with it immediately obtains the preferable game.
20 Bxc5 Qxc5 21 c4 Nfd4 22 Bxg6
Or 22 cxd5 Qxc1 23 Qxc1 Ne2+.
22 … hxg6 23 Re1 Rd8 24 Nxd4 Nxd4 25 cxd5 Qxd5 26 Nc4 Rac8 27 Qd3 Nf5 28
Qe2
 28 Qxd5 Rxd5 29 Re4 Rdc5 is awkward for White.
28 … Qb5 29 Qf1 Rc5 30 Na3 Qb4 31 Rxc5 Qxc5 32 Nc4??
Time trouble, perhaps?
32 … Ng3 0-1
Black’s system tends to work very well if White permits him to play … Bg4 early.

We have seen that Black has two good and very different choices after 3 e5. He can go
for the stable 3 … Bf5, or play much more sharply and riskily with 3 … c5, which can
lead to difficult and challenging positions. You should take your pick on that basis. Here
are a couple more games with the latter move.

Game 46
M.Kravtsiv-I.Cheparinov
Turkish Team Championship 2017

1 e4 c6 2 d4 d5 3 e5 c5 4 dxc5
4 Ne2!? is novel but hardly threatening. M.Vachier-Lagrave-V.Anand, Stavanger 2017,
continued 4 … Nc6 5 c3 Bg4 6 h3 Bxe2 7 Bxe2 e6 8 0-0, when taking on d4 appears to be
the most accurate choice: 8 … cxd4 9 cxd4 Nge7 10 Nc3 Nf5 with equality.
4 … e6
Perhaps the most solid move, enabling the pawn to be regained by force.
5 a3!?
A very popular move. If instead 5 Be3 then 5 … Qc7!? is a good way forward; e.g. 6
Nf3 Nd7 7 Bb5 Bxc5 8 Bxc5 Qxc5 and Black is equal.
5 … Bxc5
 6 b4!?
Kravtsiv comes to the board determined to sharpen things up. Instead, 6 Nf3 Ne7 7 Bd3
Ng6 8 0-0 is a much more stable continuation for White, reaching a position that looks
like it started life as a French. S.Karjakin-V.Topalov, Shamkir 2017, continued 8 … Nc6 9
b4 Bb6 10 Bb2 Nf4 11 c4 Nxd3 (castling seems like a good option: 11 … 0-0 12 Nc3 Bd7
13 cxd5 exd5 with an unclear position; Black is less cramped than in the game, though he
has his isolated pawn to worry about) 12 Qxd3 dxc4 13 Qxc4 Ne7 14 Nc3 Bd7 (White has
more freedom of movement and this gives him the advantage) 15 Qg4 Bc6 16 Rad1 Qc7
17 Ng5 Qxe5 18 b5 h5 19 Qh4 Bxb5 20 Rfe1 Qf5 21 Nxb5 Qxb5 22 Bxg7 (easy to
understand; Black has engaged his opponent in a pitch battle with the position of his king
unresolved – no wonder he loses quickly) 22 … Nf5 23 Nxe6! fxe6 24 Rxe6+ Kf7 25
Qf6+ Kg8 26 Bxh8 Bxf2+ 27 Kh1 Qa4 28 Red6 Rf8 29 Qg6+ Kxh8 30 Rd7 1-0.
6 … Be7
Rather than 6 … Bb6, when 7 Qg4 might be annoying.
7 Nf3 a5 8 bxa5?!
Although White wins this game, his whole opening strategy seems very dubious.
Breaking up his pawns like this cannot be good – but it does lead to a messy situation,
which it looks like was what he wanted. Naturally, 8 b5 is the standard and probably better
move. Why you would prefer 8 bxa5 - ?
8 … Nc6 9 Bd3 Qxa5+
9 … f6! was also possible, taking the fight to White. Tactically, Black can handle
anything that White throws at him; e.g. 10 Bb2 fxe5 11 Nxe5 Nf6 (or 11 … Bf6 12 Qh5+
g6! 13 Bxg6+ hxg6 14 Qxg6+ Ke7 15 Qf7+ Kd6 16 f4 Qe7) 12 Nd2 Nxe5 13 Bxe5 Rxa5
14 0-0 0-0, when I am not sure I like White’s poor queenside pawn structure.
10 Bd2 Qc7 11 0-0

11 … f6
Was 11 … Nxe5 playable? It seems so after 12 Nxe5 Qxe5 13 Bc3 Qg5 14 f4 (or 14
Bb5+ Kf8 15 Qd4 Nh6 16 f4 Qg6 17 Bd3 f5) 14 … Bc5+ 15 Kh1 Qh6 16 Bb5+ Kf8 17
Nd2 Nf6. One can see why Cheparinov doesn’t go for these lines, He just prefers a move
which he thinks is less risky and good at the same time.
12 exf6 Nxf6 13 c4 0-0
13 … d4! is also okay. White will have a hard job surrounding the pawn which
certainly restricts him. For example, 14 a4 0-0 15 Na3 Bxa3 16 Rxa3 e5 17 Re1 (or 17
Qc2 Qe7) 17 … Bg4 and Black is slightly better.
14 Nc3 Ng4
Threatening … Rxf3.
15 Nb5 Qd8
15 … Qb8 kept the threat up and introduced some tactics; e.g. 16 g3 Nce5! 17 Bf4? (17
Nxe5 is perhaps a little better but runs into the same exchange sacrifice: 17 … Nxe5 18
Bf4 Rxf4 19 gxf4 Nxd3 20 Qxd3 Qxf4 21 cxd5 Bd7 22 a4 Bf6) 17 … Rxf4 18 gxf4 Nxd3
19 Qxd3 Qxf4 20 cxd5 Bd7 21 Nc3 Rf8 and Black has a strong attack.
16 Qe2 Bf6
Throughout this game Cheparinov neglects tactical ideas in favour of safer
continuations. Here 16 … Rxf3!? was still interesting; e.g. 17 Qxf3 (or 17 gxf3 Nge5 18
cxd5 Nxd3 19 Qxd3 Ne5 20 Qe4 Qxd5 21 Qxd5 exd5 22 Kg2 Ra6) 17 … Nge5 18 Qh3
Nxd3 19 Qxd3 Ne5 20 Qg3 Nxc4 21 Bc3 Qf8. Playing the exchange down is not to
everyone’s taste.
17 Rad1 Nge5 18 Nxe5 Bxe5 19 Bb1 g6 20 Bb4 Nxb4 21 Qxe5 Nc6 22 Qg3 Kg7 23
Rfe1
I think this is the phase of the game where Kravtsiv starts to get on top. He has
weakened the black king position and has central pressure as well.
23 … Ra4 24 Nd6 Ra5 25 h4!

Softening up the kingside still more.
25 … Kh8 26 cxd5 Rxd5 27 Rxd5 exd5 28 Re8!
Very nice!
28 … Qf6
28 … Rxe8 29 Nf7+ Kg7 30 Nxd8 Rxd8 31 Qc7+ Bd7 32 Qxb7 does not hold out any
hope for Black.
29 Rxc8 Rxc8 30 Nxc8 Qe6 31 Nd6! Qe1+
Black gets the piece back, but there is a sting in the tail.
32 Kh2 Qxb1 33 Qf4! 1-0

Game 47
M.Rodshtein-K.Arkell
Hastings 2014/15

1 e4 c6 2 d4 d5 3 e5 c5 4 dxc5 e6 5 a3!?
Again this move. If Black now plays 5 … Bxc5, White might reply with 6 b4 (as we
saw in the previous game) or even 6 Qg4. Arkell tries to steer around this idea.
5 … a5!? 6 Nf3
Already the engines are starting to favour White. Here 6 Be3! seems a very stiff test of
Black’s opening plan. If now 6 … Nd7 then 7 Nc3! keeps the option open of Qg4 for as
long as possible. For example, 7 … Bxc5 8 Bxc5 Nxc5 9 Qg4 Qb6 10 0-0-0 Kf8 11 Nf3 is
a bit uncomfortable for Black.
6 … Bxc5 7 Bd3 Ne7 8 0-0 Ng6

9 Bg5
Keith had fond memories of this position and so is happy to repeat the variation.
Perhaps 9 Bg5 was Rodshtein’s improvement over A.Longson-K.Arkell, Ascot 2013,
which saw 9 Nc3 Nc6 10 Qe2 0-0 11 Nb5 f6 12 exf6 Qxf6 13 Bg5 Qf7 14 Bd2 Nf4 15
Bxf4 Qxf4 16 c4 Be7 17 Rae1 g6 18 Qc2 Kg7 19 cxd5 exd5 20 Re3 Bf6 21 g3 Qb8 22
Rfe1 Bg4 23 h4 d4 24 Nh2 dxe3 25 Nxg4 exf2+ 26 Qxf2 0-1.
9 … Be7
9 … Qb6 10 Nc3 Bd7 11 h4 gives White a strong initiative, so Arkell feels obliged to
swap bishops.
10 Bxe7 Qxe7 11 Nc3 Bd7 12 Qd2 0-0 13 Rae1 Na6 14 Nd4
The self-imposed weakness of the b5-square does not fill one with confidence when
one considers Black’s chances (even if it is not especially relevant right at this moment)
and he is also rather passive. All in all this position is solidly better for White.
14 … Nc5 15 f4 Nxd3 16 cxd3
16 Qxd3 is a little more fluid and White maintains the edge after 16 … Qc5 17 Rf2 f6
18 exf6 Rxf6 19 f5! Raf8 20 Qd2 b6 (or 20 … exf5 21 Nb3) 21 Ref1.
16 … Qc5 17 Qe3 f6 18 exf6 Rxf6 19 g3 Re8 20 Nf3 Rc8

21 Rf2
Perhaps 21 Qxc5! Rxc5 22 Rc1 is more precise with a really unpleasant grip on this
French-like ending. After 22 … Rf8 23 Ne2 Rfc8 24 Rxc5 Rxc5 25 Ned4 White is clearly
better.
21 … b6! 22 Rfe2 h6 23 Qf2 Rcf8 24 Qd4 Bc8 25 Qxc5 bxc5 26 Na4?
Careless. He has been dictating play and underestimates the black resources. Instead,
26 h4! keeps Black penned in, and if he tries the same sacrifice with 26 … Nxf4 27 gxf4
Rxf4 then 28 Ne5! Rxh4 29 Rg2 is very good for White.
26 … Nxf4!
Turning things around. The difference is that when the rook comes to f4, the knight on
a4 is attacked.
27 gxf4 Rxf4 28 Nxc5 Rxf3 29 Re3 R3f4 30 h3 R8f6
The long grind begins. Can Black win with his extra pawn? Keith has the patience and
the technique to make every effort to convert.
31 R1e2
 31 Nxe6 Rg6+! 32 Kh1 Bxe6 33 Rxe6 Rg3 34 R6e3 Rff3 results in a rook ending
which is very promising for Black. Quite understandably, Rodshtein is reluctant to give up
his knight.
31 … Kf7 32 Kg2 Ke7 33 Kg3?!
33 Rd2 Kd6 34 d4 Rg6+ 35 Kh2 was a little more tenacious. Now Keith forces the
knight back.
33 … Kd6 34 Nb3
Or 34 Rc2 e5.
34 … a4 35 Na5 d4
I’m not sure why Black doesn’t try 35 … e5 36 Rc2 h5. Possibly Arkell was short of
time.
36 Re4?!
36 Nc4+ Ke7 37 Re4 Rf3+ 38 Kh2 Rxd3 39 Ne5 Rd1 40 Nc6+ Kd6 41 Rxd4+ Rxd4 42
Nxd4 Ba6 43 Rd2 was an alternative defensive try.
36 … Rf3+ 37 Kg2 Rxd3 38 Rc2 Bd7
After 38 … Ba6!, 39 Rc6+ Kd5 40 Rxa6 is not possible on account of 40 … Kxe4.
39 Nc4+ Ke7 40 Nb6

40 … Kd6
This has to be time trouble. 40 … Rg6+ is clearer; e.g. 41 Kh2 Kd8 42 h4 Bb5 and
White is hampered by the vulnerability of his king.
41 Nc4+ Ke7 42 Nb6 Kd8
 Again 42 … Rg6+! is very strong.
43 Rc4 e5 44 Nxd7 Kxd7 45 Rxe5 Rc6 46 Rd5+ Rd6 47 Rxd6+ Kxd6 48 Rxa4 Kd5
49 Rb4 Re3 50 Rb7 Kc4 51 a4 d3 52 Kf2
52 Rd7 was an alternative defence; for example, after 52 … Re2+ (or 52 … Kb3 53 a5
Kc2 54 Rc7+ Kxb2 55 Rd7 Kc2 56 Rc7+ Kd1 57 a6) 53 Kf1 Rxb2 54 Ke1 Re2+ 55 Kd1!
g5 56 Rc7+ Kb4 57 Rd7 Re3 58 h4 gxh4 59 Kd2 Rh3 60 Rd4+ Kc5 61 Rf4, the position is
not going to be easy to win for Black.
52 … d2 53 Rc7+ Kb3 54 Rd7 Rxh3

55 Ke2
Over to Keith for his explanation: “I think it was a very unusual endgame, and difficult
for us both to fully understand at the time. I felt that 55 Kg2! would draw, and now believe
that to be the case. My understanding of the position after 61 … h3 was that I was almost
certainly winning, chiefly because his king wants to be on both d1 and c2 at the same
time. The problem for White is that he must keep his pawn on b2. Otherwise I win,
simply, with … Rh1+ when Kxd2 is met by … h2, followed by a rook move picking up
his rook wherever he has his king. Therefore he has to keep my king out of b3. I didn’t
work out all the analysis over the board – not by a long way, of course. However, this
turns out to be the case, though there is some intricate stuff with best play. Keeping my
king out of b3 requires him to put his king on c2, but this enables me to gain time by
manoeuvring my king to e1, threatening to both queen the pawn or block check with …
Re2 winning. That’s about it. I missed his move 69 Re3, but quickly saw that 69 … Kg2
wins fairly trivially, as his b-pawn can be stopped in time. Strangely, only just in time
though. For instance, if his king was on d4 instead of c3 in the final position then it is a
draw!. He did indeed place it there, but only because we have to place the kings on d4 and
e5 at the end of the game to indicate a Black win for the live transmission.”
55 … Rh2+ 56 Kd1 Kxa4 57 Rxg7 h5 58 Kc2 Kb5 59 Rc7 h4 60 Rc8 Kb6 61 Rc4
h3 62 Rh4 Kc5 63 Kd1 Kd5 64 Kc2 Ke5 65 Rh8 Ke4 66 Re8+ Kf3 67 Rf8+ Ke2 68
Re8+ Kf1 69 Re3 Kg2
69 … Rh1! also works: 70 Kxd2 Kg2 71 Re2+ Kf3 72 Re3+ Kf4 73 Re2 h2 74 Kc2
Rc1+ 75 Kxc1 h1Q+.
70 Kxd2
70 Rd3 Kf1 71 Kd1 Rh1! 72 Rxd2 Rg1 is a nice winning variation.
70 … Rh1 71 Re8 h2 72 Rh8 Rb1 73 Kc3 h1Q 74 Rxh1 Rxh1 0-1
Not a premature resignation. If 75 b4, then 75 … Rh4! is the instructive winning move,
whereas if 75 Kd4 then 75 … Rb1! is deadly. Typical perseverance by Arkell.

At the end of this chapter we turn our attention to some oddball lines, which are not as
bad as they might seem.

Game 48
P.Giacomini-A.Adly
Menton 2016

1 e4 c6 2 d4 d5 3 e5 g6!?

The grandmaster avoids theory, steering his lower-rated opponent into unfamiliar
territory. 3 … g6 introduces unorthodox play and White must find his own moves from
now on.
Other unusual tries are:
1) 3 … h5!? is a move I recommended in a Foxy Openings video a while back, which
dealt with offbeat Black lines of the Caro-Kann. 3 … h5 looks extravagant, but Black is
aiming to set up a chain of light-squared pawns and the pawn is supporting … Bg4. It’s
not that bad. K.Griffith-D.Fernandez, Kecskemet 2015, continued 4 c4 g6 5 Nc3 Nh6 6
cxd5 cxd5 7 Qb3 e6 8 Nf3 Be7 9 Bxh6 Rxh6 10 h4 a6 11 Bd3 Nc6 12 Rc1 Bd7 13 g3 Kf8
14 Kf1 Na5 15 Qc2 Kg7 and was later drawn in 85 moves. The opening experiment has
not turned out too badly. The position is equal with all to play for. Black has a rather slow-
moving game but owns the bishop pair and can hope to gradually take charge of the dark
squares.
2) 3 … Na6!? is my least favourite line covered here. Black is obviously hoping for 4
Bxa6?! Qa5+, but White is not going to be that naive. The problem is one of time. Black’s
knight on a6 is not well placed and will have to come to c7 shortly, but is it any better on
that square?
Once again, we are encountering a line which can be launched as a surprise. With the
blocked centre, White must work hard to prove that 3 … Na6 is bad and will probably
have to show some originality. T.Rydstrom-E.Rozentalis, Gothenburg 2017, continued 4
Nf3 g6 5 c4 Nc7 6 Nc3 Bg7 7 Qb3 Nh6 (more extravagance) 8 cxd5 Nxd5 9 Bc4 e6 10
Ne4 0-0 11 h4 Qb6 (with an attack brewing, Black hastens to get the queens off) 12 Qd3
Nf5 13 a3 h5 14 Bg5 Qa5+ 15 Bd2 Qc7 16 Rc1 Rd8 17 Bg5 Nde7 18 Nf6+ Kf8 19 Nh7+
Kg8 20 Nf6+ Kf8 21 Nh7+ Kg8 22 Nf6+ ½-½. White was very happy to take a draw
against the veteran grandmaster and so 3 … Na6 didn’t really have the desired effect.
4 c4
Reacting in the centre seems reasonable.
4 … Bg7 5 Nc3 f6!
Sharpening the game considerably and making perfect sense of the bishop on g7.
6 f4 Nh6 7 Nf3 0-0 8 Be3 Ng4


9 Qd2?
Adly’s experiment has been successful. White should not give up his dark-squared
bishop so easily. Instead, he should play 9 Bg1!, not fearing 9 … fxe5 10 fxe5 Qb6 (or 10
… e6 11 h3 Nh6 12 Bd3 c5!? 13 dxc5 Nc6) 11 Qd2. Obviously, this position is very
unclear, but White is in there fighting. As the game goes, he regrets losing the bishop on
e3.
9 … Bh6! 10 exf6
The ugly 10 g3 was forced, but Black is a bit better after 10 … Nxe3 11 Qxe3 Na6! 12
0-0-0 (or 12 cxd5 Nb4) 12 … Nc7 13 Kb1 Be6 14 c5 Qd7.
10 … Nxe3 11 Qxe3 exf6 12 0-0-0 b5!?
Trying to finish him off here and now. There was nothing wrong with 12 … Re8 13
Qd2 Qc7 14 g3 Bg4, which looks to me like a more correct way of developing the
advantage.
13 cxb5 a6 14 g3?
He should brave 14 bxa6, when Black obviously has a powerful attack, but White has a
pawn for his woes and may survive to enjoy it!
14 … axb5 15 b3 Re8 16 Qd2 Bf8 17 h4 Bg4 18 Qd3 Bf5 19 Qd2 Ba3+ 0-1
We must keep an open mind about all of these third move alternatives. None of them
are bad, all are relatively solid, and White must show them respect.

 Chapter Four
The Panov-Botvinnik Attack

1 e4 c6 2 d4 d5 3 exd5 cxd5 4 c4

The Panov-Botvinnik Attack became popular after World War Two, when at first the
Soviet master Panov and later World Champion Botvinnik started to analyse the
possibilities. 4 c4 leads to a more open game than most of the other main lines. In many
cases it leads to isolated queen’s pawn positions. Transpositions to the Queen’s Gambit,
the Nimzo-Indian, and even the 2 c3 Sicilian are quite common.
4 … Nf6
This is nearly always played.
5 Nc3
The path now diverges into 5 … g6, 5 … Nc6 and 5 … e6, all of which are decent
replies. It will be a question of taste as to which you prefer.

Game 49
L.Portisch-R.Wade
Palma de Mallorca 1966

1 c4 c6 2 e4 d5 3 exd5 cxd5 4 d4 Nf6 5 Nc3 g6!?

Let’s start by taking a look at 5 … g6, which if it works, is definitely the move Black
wants to play. He deploys his dark-squared bishop as actively as possible and hopes to
attack the d4-pawn in due course. He can continue easily with … 0-0, maybe … d5xc4
and then … Nbd7-b6, blockading d5. If White just sits there and plays standard
developing moves, then this plan works extremely well. The problem arises after 6 Qb3!.
6 Qb3! Bg7
Since 6 … dxc4 7 Bxc4 gives White a strong attack, Black is obliged to sacrifice a
pawn.
7 cxd5 0-0
An important position. Can Black get his pawn back or not? Is the extra pawn, doubled
and isolated in the centre, useful to White anyway?
Well, yes, the pawn is extremely useful because it is inhibiting Black’s development
and so we come to the main struggle in this line. If Black does not get the pawn back soon
he will end up in trouble. We start with a game from a while back, where Portisch, a
world-class grandmaster, plays a rather unusual move, but one that poses a threat to the
black set-up.
 8 Bg5!?
Either 8 Be2 (Game 9) or 8 Nge2 (Game 50) is much more common.
8 … Nbd7 9 Be2 Nb6
Wade goes straight for the pawn in the usual way.
10 Bf3!

It’s positions like this one that make me a bit uneasy about the whole black idea. Yes, if
he gets the pawn back he will stand well, but can he get it back? White’s plan continues
with Nge2, 0-0, then rooks to the middle and Nf4 where necessary. He will take on f6 if
necessary to keep that pawn on d5. Black has to play really inventive chess from here to
show compensation; that is the issue. He cannot relax for a moment.
10 … a5?!
I am not sure that this is the way. 10 … Bf5 was better, intending 11 Nge2 (or 11
Rd1!?) 11 … Bd3 12 Rd1 Bc4 and the bishop finds a nice square.
10 … h6 is not quite as good, met as it is by 11 Bxf6 Bxf6 12 Nge2 Bf5 13 Rd1!
(preventing … Bd3) 13 … Rc8 14 Ng3 and White is a still a pawn up.
11 Bxf6 Bxf6 12 Nge2 a4 13 Qb4
Portisch is careful not to put his queen in the way of Black’s light-squared bishop.
13 … Bf5
The machines are suggesting 13 … Qd6!? 14 Qxd6 exd6 15 0-0 Bd7 and claiming
equality, as Black has positional pressure for the pawn. However, White may also play in a
slightly unorthodox way with 16 Be4! Rfe8 17 Bd3 Ra5 18 b4 axb3 19 axb3 Rea8 20
Rxa5 Rxa5 21 Be4 and he is still a pawn to the good, even though it might not be much of
a pawn.
14 0-0 Bd3 15 Rfd1 Bxe2 16 Nxe2 Qd6 17 Qb5!

Holding on to the d5-pawn for dear life.
17 … a3 18 b3 Bg5 19 Nc3
Black may place a rook on c8, but then comes Ne4. Wade is running out of ideas and
plays a very loose move.
19 … f5?!
19 … Bf4 20 g3 Bh6 21 Ne4 is still very nice for White, but perhaps Black could last
longer than he does in the game.
20 Re1 Rf6 21 Qd3 Qb4 22 Nb5 Ra5 23 Nc7 Rd6 24 Rad1 Kf7 25 g3 Bf6 26 Qe3
The self-inflicted weaknesses in the black camp start to show, with Qh6 always an idea
from now on.
26 … Rd7 27 Ne6 Qd6 28 Qh6 1-0
There are many games like this buried in the archives. White takes a cold-blooded
approach, keeps his pawn, and refutes each trick as it comes along. We learn from this
brief adventure that 6 Qb3 poses a real threat.

Game 50
S.Harini-B.Amin
Dubai 2017
1 e4 c6 2 d4 d5 3 exd5 cxd5 4 c4 Nf6 5 Nc3 g6
I would say that this is a good weapon if you are facing a lower-rated opponent or
someone who feels uncomfortable defending. It’s not a line to play against a good
technician, as White will find a way of keeping the extra pawn for a while, thus making
life awkward. In this next game we see a grandmaster risking 5 … g6 in an all-out effort to
win with Black.
6 Qb3 Bg7 7 cxd5 0-0 8 Nge2
This and 8 Be2 are the main choices. The difference is that after 8 Nge2 White intends
to place the bishop on g2, again reinforcing d5.

8 … Na6!?
Perhaps the most natural idea for Black is 8 … Nbd7 9 g3 Nb6 10 Bg2 Bf5 11 0-0 Qd7
12 Rd1 Rfd8 13 Nf4 Rac8. Here we have a typical position for this line. Black is a pawn
down, but he has developed actively. Frankly, I think most masters would prefer to be
White, as the onus is on Black to prove it. 14 a4! would be a good start, with White
making some progress.
9 g3 Qb6 10 Qxb6
Surprising! Black offers exchanges a pawn down and allows White to double his
pawns. It’s all about activity. Black is going to place his knight on b4.
10 … axb6 11 Bg2 Nb4 12 0-0 Rd8 13 d6
An advance which Harini plays to improve the co-ordination of her pieces. In this case,
the bishop on g2 is unleashed. Note that 13 Nf4?! would be met by 13 … Nc2.

13 … Rxd6
13 … exd6 14 Bg5! did not fare very well for Black in the recent grandmaster game
A.Naiditsch-V.Burmakin, Zürich 2015: 14 … Re8 15 a3 Nc6 16 Rfe1 h6 17 Bxf6 Bxf6 18
Ne4 (18 Nd5! seems very strong; e.g. 18 … Bd8 19 Nec3 Rxe1+ 20 Rxe1 Kf8 21 Nb5) 18
… Bg7 19 Nxd6 Rd8 20 Nb5 Ra5 21 Nec3 Nxd4 22 Rad1 Kf8? (22 … Bg4) 23 b4 Ra6 24
Re4! Bf5 25 Rexd4 Bxd4 26 Bxb7 1-0. Throughout, the weak queenside pawns hampered
Burmakin’s ability to drum up any meaningful counterplay. He may have been okay if he
had played 22 … Bg4, but equally 18 Nd5! was stronger than 18 Ne4.
14 Bf4 Rd7
At first sight, this position definitely looks better for White. Black’s pawns are wrecked
and the white bishops rake the queenside. It is deceptive though. White has an isolated
pawn on d4 which can be attacked and Black is active on the light squares. Amin has
chosen his weapons well. He backs himself in this unbalanced situation and expects that
Harini will go wrong first.
15 Rfd1 Nfd5 16 Bd2 Rd8 17 a3 Nxc3 18 bxc3 Nc6 19 Bf4 Bg4 20 f3 Be6 21 Bc7
Rd7
Extraordinarily, this is the first new move. It seems like a lot of preparation went into
this game. 21 … Rdc8 had been played previously in M.Kobalia-G.Kamsky, World Blitz
Championship, Khanty-Mansiysk 2013, which continued 22 Bxb6 Bc4 23 Nf4 e5 24 dxe5
Bxe5 25 Nd5 Bxd5 26 Rxd5 Bxc3 27 Rb1 Rxa3 28 Rd7. This position is about equal,
despite Black’s extra pawn.
22 Bxb6 Bc4 23 Kf2 Ra6 24 Bc5?
 24 Rdb1 was correct, keeping the bishop on b6 for the time being. After 24 … e5 25
Bf1 it is, again, about equal.

24 … b6!
Surprisingly strong. The bishop becomes a target.
25 Bb4 Rda7! 26 Rab1 Nxb4 27 axb4 Ra2 28 Bf1 e5
The game opens up to Black’s advantage.
29 dxe5
The pawn advance can be stopped by 29 d5 Bh6 30 d6 Bd2.
29 … Bxe5 30 Rbc1 R7a3 31 Ke3 Bxe2 32 Bxe2 Bxc3
Black regains the pawn and will win more material soon, maintaining a strong attack.
33 Bd3 Bxb4 34 h4 h5 35 Ke4 Bc5 36 Bc4 0-1
I guess White lost on time. At any rate, after 36 … Rf2, her position is resignable
anyway.

The coming game does not start life as a Caro-Kann, but soon transposes.

Game 51
A.Riazantsev-D.Jakovenko
FIDE Grand Prix, Sharjah 2017

1 c4 c5 2 Nf3 Nc6 3 e3 Nf6 4 d4 cxd4 5 exd4 d5 6 Nc3 Bg4

 This is one of the most popular replies to the Panov Attack. Essentially, Black
concentrates on development. Over time, White has found that there is only one way to
apply pressure and that is to attack b7.
7 cxd5 Nxd5 8 Qb3! Bxf3 9 gxf3 e6
9 … Nb6 is an important alternative, which we will look at next.
10 Qxb7 Nxd4 11 Bb5+!

This is all very well-known theory and has been played countless times. I am not sure
whether this stuff is suitable for any player below master level because of the amount of
information that one has to know. As it happens, in our featured game, a strong
grandmaster actually forgets what to do at a critical point and loses straight away!

TIP: Think twice about entering highly theoretical lines if you wish to improve.
Chess should not be a battle of memory.

11 … Nxb5 12 Qc6+
An important zwischenzug that forces the king to e7.
12 … Ke7 13 Qc5+!?
An interesting departure from the common 13 Qxb5, when an ending can be reached
after 13 … Qd7 14 Nxd5+ Qxd5 15 Qxd5 exd5 which is a pretty good drawing weapon
for Black, if recent statistics are anything to go by. Now comes 16 Be3 or 16 0-0, neither
of which could exactly be described as thrilling. This ending has been played out so often,
that one could conduct a 50 move game, or even longer, and not play a single original
move. To me, that is a reason to back off.
13 … Ke8
13 … Nd6 keeps the piece but gives White a very dangerous attack after 14 Nxd5+
exd5 15 0-0. Black can try to defend after 14 … Kd7 15 Nc3 Qc8 16 Qa5 Qc6 17 Be3, but
I do not like his position very much.
14 Qxb5+ Qd7 15 Nxd5 exd5

16 Qb3
The whole point of 13 Qc5+ is to play with more risk, keep the queens on, and avoid
the dreary ending mentioned earlier. It is an unusual position with mutual pawn
weaknesses and both kings are exposed.
16 Qd3 has been scoring well for White in recent times, despite the fact that it allows
… Bb4+. The game A.Grischuk-S.Vidit, World Rapid Championship, Doha 2016,
continued 16 … Bb4+ 17 Kf1 f6 (it is worth remembering that Black cannot castle!) 18
Kg2 Kf7 19 Bf4 Ba5 (19 … Rhd8 20 Rac1 Kg8 21 Rc7 Qe8 22 Bg3 Qg6 23 Qb3 was
V.Grinev-D.Mischuk, Omelnyk 2017, where White retains the more active position, as he
would after 23 Qd4 or 23 Qa6 too) 20 Rhd1 Rad8 21 Rac1 Bb6 22 Rc2 d4 23 Rdc1 Rc8
24 Qb3+ Kg6 25 Qd3+ f5 26 b4 Rxc2 27 Rxc2 Rc8 28 Rxc8 Qxc8 29 a4 Qd7 30 Qa6 Kh5
31 a5 Bd8 32 b5 (here 32 Qd6 Qxd6 33 Bxd6 a6 34 Bc5 seems winning to me; Black’s
central pawn is in danger and he faces b4-b5 at any moment) 32 … d3 33 Bd2 Bg5? (33
… h6) 34 Qc6 Qd4 35 f4! Bd8 (or 35 … Bxf4 36 Qf3+) 36 Be3 Qb2 37 Qe8+ g6 38 Qxd8
Qxb5 39 Qg5 mate.
 16 … Bd6 17 0-0 Rb8 18 Qe3+ Kf8 19 Rd1??
Essentially, this position is fine for Black whatever White plays. For instance, after 19
f4 comes 19 … h5 with … Rh6 in the offing. Meanwhile, 19 Qg5 is met by 19 … f6 20
Qg4 Qxg4+ 21 fxg4 h5! and Black is suddenly better in the ending, as White suffers from
such poor development. However, 19 Rd1 is just a blunder. The fact that the rook is
unprotected is decisive.
19 … Qh3 0-1
Seeing that 20 f4 Qg4+ picks up the rook.

Game 52
B.Sambuev-W.Hernandez Gonzalez
Capablanca Memorial, Matanzas 2017

1 Nf3 Nf6 2 c4 c5 3 Nc3 Nc6 4 e3 d5 5 d4 cxd4 6 exd4 Bg4

Again we see a Caro-Kann emerging from a completely different opening sequence.
7 cxd5 Nxd5 8 Qb3 Bxf3 9 gxf3 Nb6!?

9 … Nb6 is much sharper than 9 … e6. Black walks a tightrope in this line, as White is
allowed (indeed encouraged) to play d4-d5 immediately. Given his awful pawn structure,
White is forced to take action.
10 d5!
White can certainly try 10 Be3 e6 11 0-0-0 Be7 12 d5 exd5 13 Nxd5 Nxd5 14 Rxd5
Qc7, but then Black has a stable position.
10 … Nd4 11 Qd1 e5 12 dxe6
12 Be3 Bc5 13 f4 is an unsuccessful attempt to knock Black off his central perch.
However, after 13 … 0-0! 14 fxe5 Qh4 15 Bg2 Nc4, Black has a tremendous initiative,
this being the sort of position players of 9 … Nb6 are hoping for.
12 … Nxe6
12 … fxe6!? is clearly playable and there is a piece sacrifice in the air. We now follow
the game P.Mazzilli-P.Basso, Italian Team Championship 2017: 13 Be3 Bc5 14 b4 0-0 15
bxc5! Nxf3+ 16 Ke2 Qf6 17 Bg2?! (17 cxb6 Rad8 18 Qc2 Nd4+ 19 Bxd4 Qxd4 leaves
Black two pieces down, but with White having to defend for his life; best now is 20 Ke1
Rxf2 21 Rd1 Rxc2 22 Rxd4 Rxd4 23 bxa7 Rd8 24 Nb5 Rxa2, reaching a rather difficult
position to play) 17 … Nc4 18 Rc1 Qf5 19 Qd3 Qh5 20 Kf1 Nfe5 21 Qe2 Nxe3+ 22 Qxe3
Ng4 23 Qxe6+ Kh8 24 Ne4 Rae8 25 Qb3 Nxf2 0-1. A crushing win and a game which
demonstrates the vitality of the whole variation. White must play precisely.
13 Bb5+ Nd7 14 0-0 a6

15 Bxd7+
Keeping the tension with 15 Ba4 seems a better way to try for a win, although if Black
plays sensibly, he should be able to maintain a decent position. Play can continue 15 …
Be7! (a nice tactical sequence follows after 15 … b5?! 16 Bb3 Ndc5 17 Bd5 Rc8 18 b4
Nd7 19 Qb3 Qf6 20 Ne4! Qxa1 21 Bb2 Nd4 22 Nd6+! Bxd6 23 Qe3+ Be5 24 Bxa1 Nc2
25 Qe4 Nxa1 26 f4 and White wins) 16 Be3 b5 17 Bb3 0-0. This position is about equal.
15 … Qxd7 16 Qxd7+ Kxd7 17 Rd1+ Ke8 18 Be3 Rd8!
 White is more active, whereas Black has the superior pawn structure and so exchanges
make sense.
19 Nd5 f5
Making room for the king to come to f7, a common location in this variation. Black is
readying himself for more exchanges and any ending that may follow.
20 Bb6 Rd6 21 Rac1 Kf7! 22 a4 g6 23 Bc7 Rd7 24 Nb6 Rxd1+ 25 Rxd1 Nxc7 26
Rd7+ Ke6 27 Rxc7 Bd6 28 Rxb7 Rb8!
Sambuev has even won a pawn, but the centrally placed black king is the most
important feature of the position. In fact, it is White who is struggling to draw.
29 Rxb8 Bxb8 30 b4 Bc7 31 Nc4 Kd5 32 Ne3+ Kd4 33 Kg2 Bd6 34 b5 axb5 35 axb5
Kc5 36 b6 Kxb6 37 f4 Kc6 38 Kf3 Bc7 39 h3 ½-½
The line with 5 … Nc6 and 6 … Bg4 is still reliable and very popular. Stylistically, I
prefer the sharp interpretation that we have just witnessed, but there is no reason why 9 …
e6 should not be played. You just have to be in the right frame of mind.

Game 53
A.Stebbings-F.Trois
World Seniors Championship, Acqui Terme 2017

1 e4 c6 2 d4 d5 3 exd5 cxd5 4 c4 Nf6

Here 4 … dxc4 5 Bxc4 Qc7!? is an unusual sequence, usually reached via the Queen’s
Gambit Accepted with the move order 1 d4 d5 2 c4 dxc4 3 e3 c5 4 Bxc4 cxd4 5 exd4 Qc7.
Tony Miles used to like playing this way but I am not convinced. At some stage the queen
on c7 could become a target for attack. Nevertheless, 5 … Qc7 is not refuted by any
means. The game R.Vaganian-C.Gabriel, German Bundesliga 2005, continued 6 Be2 Nf6
7 Nc3 e6 8 Nf3 Be7 9 Ne5 0-0 10 0-0 Rd8 11 Nb5 (Vaganian has not tried to exploit 5 …
Qc7 at all, just settling for steady development; once castled, he now starts to pick up a bit
of steam) 11 … Qb6 12 a4 Rf8 (a weird move, but it does make room for the queen) 13 a5
Qd8 14 Bf3 Nd5 15 Bxd5 exd5 16 Bf4 Na6?! (16 … a6) 17 Bd2 Nb4 18 Qb3 Nc6 19 a6!
Nxe5 20 dxe5 bxa6? 21 Ba5 1-0. A brief game which helps us to understand that White
does not need to try to smash 5 … Qc7.
5 Nc3 Nc6
I once had a bad experience against Mark Hebden after 5 … Be6?! 6 Nge2! with the
strong idea of Nf4. This has some relevance to our main game.
6 Bg5
 Rather than 6 Nf3, White makes an aggressive bishop move, threatening to capture on
f6. This was Botvinnik’s favourite. Black has tried many ideas now. Perhaps 6 … e6 is
most common.
6 … Be6!?
After 6 … e6 7 Nf3 Be7 White can change the dynamics of the pawn structure by
playing 8 c5. This may give him an edge, as he establishes a queenside pawn majority.
The game M.Adams-J.Speelman, Staunton Memorial, London 2008, proceeded with the
usual reply 8 … Ne4, aiming for exchanges. What Black cannot do is just sit there
passively and let White roll the queenside pawns up the board. Play continued 9 Bxe7
Qxe7 10 Qd3 Nxc3 11 Qxc3 0-0 12 Be2 b6 13 b4 e5 (Speelman takes what he thinks is
necessary action) 14 Nxe5! Nxe5 (14 … Re8! would have justified Speelman’s play and
I’m not sure why he didn’t go for it: after 15 0-0 Nxe5 16 dxe5 Qxe5 17 Qxe5 Rxe5 18
Bd3 bxc5 19 bxc5 Bf5 a rather level position has arisen; 15 Bb5 is normally the right idea,
but White’s king is still in the middle and after 15 … Nxe5 16 dxe5 Qxe5+ 17 Qxe5
Rxe5+ 18 Kd2 Rf5! 19 f3 bxc5 20 Rhe1 Be6 21 bxc5 d4, Black has an active position) 15
dxe5 a5 16 Qd4! (as usual, Mickey Adams is very precise) 16 … axb4 (16 … bxc5 17
bxc5 Re8 18 f4 Rb8 19 a3! keeps the black rooks at bay; the queen is incredibly strong on
d4) 17 cxb6 Re8 18 0-0 Qxe5 19 Qxb4 Bb7 (19 … Qxe2 20 Rae1) 20 Bf3 Qe7 21 Qd4
Qa3 22 h3 (White has consolidated his extra pawn; against Adams, one can write the
game off) 22 … Ra4 23 Qd2 Qd6 24 Rfb1 h6 25 Rb5 Bc6 26 Rb2 Qf6 27 Rab1 Rd4 28
Qc1 Bb7 29 Rc2 Rf4 30 Rb3 Rc4 31 Rb1 Rf4 32 Qd1 Rd4 33 Rd2 Rxd2 34 Qxd2 Re6 35
Qb4 Qg6 36 a4 Rf6 37 Rb3 Qg5 38 a5 Rf4 39 Qd2 Qf6 40 Re3 d4 1-0.
Instead, 6 … Qa5!? is another one of those sharp tries, which look a bit dubious but
which White has not refuted. The capture on f6 is now a must, and 7 Bxf6! exf6 (7 …
gxf6 8 cxd5 Nb4 9 Bb5+ is just good for White) 8 cxd5 Bb4 9 Qd2! is the critical line,
when Black must examine 9 … Bxc3 (9 … Ne7 is met by 10 Bb5+! Kf8 11 Bc4 or 10 …
Qxb5 11 Nxb5 Bxd2+ 12 Kxd2 Nxd5 13 Ne2 0-0 14 Rac1 with a lead in development for
White) 10 bxc3!? Qxd5 11 Ne2 0-0 12 Nf4 Re8+ 13 Be2 which definitely looks better for
White to me. Practical results bear this out.
7 a3
A very odd-looking move, but then Black had left the main highways when he played 6
… Be6. The main danger to Black in this line is an advance of the queenside pawns, and
so Stebbings starts the ball rolling. 7 Nge2 looks similar to the Hebden fragment
mentioned earlier, but the difference is that Black can now play 7 … dxc4 and he is hitting
d4. Thus 8 Nf4 Qxd4 is absolutely fine for Black.
7 … g6 8 Bxf6 exf6 9 Be2 Qd7 10 c5

Stebbings makes his move forward. We can see what’s coming, basically b2-b4-b5.
Black must demonstrate that his dark-squared bishop, pressure against d4 and a possible
kingside attack can offset White’s advance.
10 … Bg7 11 Bf3 0-0 12 Nge2 Rfd8 13 0-0 g5 14 g3 f5! 15 Qd2 Bf6 16 b4 Kg7 17
Bg2 h5
Going for it. I like his style!
18 f4
It is irritating for White that 18 b5 is met by 18 … Na5, eyeing c4 and b3.
18 … g4 19 Nd1 b6!?
 This was unexpected. Black should probably just get on with it with 19 … h4. Out of
the blue, Trois challenges White on the queenside. Stebbings does not react in the best
way.
20 Ne3
Surely, he should play 20 b5! Na5 21 c6 Qd6 22 Qd3. Maybe 20 … Nxd4!? 21 Nxd4
bxc5 was the idea, though after 22 Nxe6+ fxe6 23 Rb1, a piece is a piece.
20 … bxc5 21 bxc5 Rab8 22 Rab1 Ne7 23 Qa5 Nc6 24 Qc3 Ne7 25 Qa5 Nc6 26 Qc3
Ne7 ½-½
An equilibrium has been reached. The databases give this as a win for Black, but it was
clearly a draw, as can be deduced from the recent repetitions.

Summarizing, after 6 Bg5, I recommend 6 … e6 as the most stable move, although
there are interesting alternatives, as we have seen. Of the minor lines. 6 … Be6 is
guaranteed to lead to a sharp position, which is ideal if Black needs to win.

A direct transposition to another opening can often result from the Panov-Botvinnik
move order. Before we enter our main game, let’s consider the way we get there using the
Caro: 1 e4 c6 2 d4 d5 3 exd5 cxd5 4 c4 Nf6 5 Nc3 e6(!). This is my favourite move here.
Black plays solidly, bolsters his centre, and prepares … Bb4 or … Be7 to be followed by
castling. It’s a reliable method and scores well. After 6 Nf3 Bb4 (6 … Be7 is also possible,
but why not play actively?) 7 Bd3 dxc4 8 Bxc4 0-0 9 0-0, hey presto, we have reached our
main game.
Game 54
B.Itkis-L.Van Wely
Romanian Team Championship 2000

1 d4 Nf6 2 c4 e6 3 Nc3 Bb4 4 e3 0-0 5 Bd3 c5 6 Nf3 d5 7 0-0 cxd4 8 exd4 dxc4 9 Bxc4
We have come to a position where Black has a wide choice of ideas. Let’s take a look at
some of the best.

9 … Nbd7
Others:
1) 9 … b6 is extremely common. Black is trying to find the best way to get his
queenside pieces out and b7 seems a good square for the bishop. Importantly, Black’s
queenside pawn structure is compact and difficult to attack. 9 … b6 is popular among
world-class players. Play might proceed as in M.Kravtsiv-Ding Liren, FIDE World Cup,
Tbilisi 2017: 10 Qe2 Bb7 11 Rd1 Bxc3! (saddling White with hanging pawns) 12 bxc3
Qc7 13 Ne5 Nbd7 14 Nxd7 Qxd7, when an equal position has arisen.
2) 9 … a6 intends … b7-b5. Black finds another way to structure his queenside. This
way is a bit more aggressive, but of course the pawns are more vulnerable, say to a strike
such as a2-a4. Y.Boidman-I.Khenkin, Wiesbaden 2017, continued 10 Bd3 (White can play
10 a4, but then Black can use the b4-square; e.g. 10 … Nc6 11 Bg5 Be7 12 Re1 Nb4!) 10
… b5 11 Ne4 Nbd7 12 Ne5 Bb7 13 Nxd7 Nxd7 14 Nc5? (White has tried to do without
a2-a4 and it hasn’t worked; he can do much better than 14 Nc5, though the black position
is excellent in any case) 14 … Nxc5 15 dxc5 Qd5! 16 Qg4 Bxc5 17 Rd1 f5 18 Qg3 Bd6
19 Qh3 Qe5 20 a4 (far too late) 20 … b4 21 Be3 Rac8 (I don’t understand why he cannot
take on b2: 21 … Qxb2! 22 Bc4 Qf6! 23 Rac1 Rad8 24 Bb6 Rc8! 25 Rxd6 Rxc4) 22 Bd4
Qxd4 23 Bxa6

TIP: If you are losing, you can try anything in an effort to get back into the game.

Unleashing a barrage of tactics, whether unsound or not, is often the way to go.

23 … Qb6 24 Bxb7 Bxh2+ (missing 24 … Rc2!; the grandmaster seems confused) 25
Qxh2 Qxb7 26 Qe5 Rf6 27 Rac1 Rxc1 28 Rxc1 Rg6 29 f3 h5 (Black is still a bit better,
but he has made life tough for himself – a very familiar scenario for the professional chess
player)

TIP: If you want lifelong enjoyment from chess, don’t play the game for a living.

30 Rc7 Qa6 31 Qb5 Qd6?? (horrible; after the better 31 … Qa8 the game may have
continued for a very long time) 32 Qb8+ 1-0. Thanks to 32 … Kh7 33 Rxg7+ or 32 …
Qf8 33 Rc8. An outright swindle!
10 Bg5 Bxc3 11 bxc3 Qc7
It looks to me as though 9 … Nbd7 is an options-preserving move. Black saddles White
with hanging pawns as usual, but he does not yet reveal how he will organize the rest of
his queenside development. It’s the variation that scores best for Black at all levels, but
this is not to say that the other lines we have covered are bad at all.
12 Qe2 b6 13 Rac1 Bb7 14 Bd3 Ng4!?

Threatening … Bxf3. Instead, 14 … h6 is rock solid, with Black able to move his queen
and rooks into decent positions, getting ready for a complex middlegame; e.g. 15 Bh4 (or
15 Bd2 Rac8 16 Rfe1 Rfd8) 15 … Qd6 16 Bg3 Qa3.
15 Be4 Bxe4
When you hold hanging pawns, you don’t want to exchange too many pieces, as the
pawns generally become more vulnerable as the endgame approaches. So Van Wely is
pleased to exchange.
16 Qxe4 Ngf6 17 Qe2 h6 18 Bd2 Rac8 19 c4 Qb7
Eyeing e4 and a6, with a possible plan of … a7-a6 and … b6-b5 in mind.
20 a4 Qa6 21 Bb4 Rfe8
We have on the board an absolutely typical position for this line. The players have
equal chances. White has the potentially worse pawn structure, so he must show the way
forward, staying as active as he can.
22 Qa2 Nb8 23 Qb3 Nc6 24 Bc3 Na5 25 Bxa5 Qxa5 ½-½
A very reasonable conclusion. You cannot really do better as Black than to choose 5 …
e6 against the Panov.

Game 55
J.Anderson-J.Hawkins
Hastings 2014/15
Isolated queen’s pawn positions statistically favour the attacker, as it is simply more
pleasant to attack! In our featured game, Jonathan Hawkins takes a step towards
redressing the balance.
1 e4 c6 2 d4 d5 3 exd5 cxd5 4 c4 Nf6 5 Nc3 e6 6 Nf3 Bb4 7 cxd5
White does not want to lose the battle for the tempo (7 Bd3 dxc4) and thus he often
employs 7 cxd5.
7 … Nxd5 8 Qc2

Although this is a main line, the early placement of the white queen on c2 isn’t
dangerous and in fact in the game, Anderson soon moves the queen again.
8 Bd2 is the other main line here, but Black is very comfortable after 8 … Nc6 9 Bd3 0-
0 10 0-0 Be7 11 a3 Bf6 with equality in T.Kantans-L.D.Nisipeanu, European Cup, Antalya
2017. There are many such similar sequences. The issue for White is that the bishop on d2
is passively placed and he will soon have to move it again, which loses impetus.
8 … Nc6 9 Bd3 h6 10 0-0 0-0 11 a3 Be7 12 Qe2 Nf6
12 … Qd6 13 Bc2 Bd7 14 Qd3 f5 15 Nxd5 exd5 16 g3 g5 is a rather loose way to
handle the black position, and after 17 Re1 Rae8 18 h4! White is starting to put the
pressure on.
13 Be3
Hawkins had had this position at the Isle of Man tournament the year before. That
game, B.Dahl-J.Hawkins, Douglas 2014, continued 13 Rd1 (more or less committing the
other rook to go to c1, which does not seem optimal) 13 … b6 14 Ne4 (14 Bc2 Bb7 15
Bf4 Rc8 16 Rac1 Bd6 is perfectly acceptable for Black) 14 … Nd5 15 Nc3 Bb7 16 Nxd5
Qxd5 17 Be4 Qd6 18 g3 Bf6 19 Bf4 Qe7 20 Rac1 Rac8 (this is the issue with placing the
white rooks as they are; it is all too easy for Black to trade along the c-file and White
should not be trading pieces) 21 Be5 Rfd8 22 b4? (22 Bxf6 Qxf6 23 Qe3 is equal) 22 …
Nxd4! 23 Bxd4 Rxc1 24 Rxc1 Bxd4 25 Bxb7 Qxb7 26 Nxd4 Rxd4 and Black had an extra
pawn.
13 … b6 14 Rad1 Bb7 15 Bb1 Bd6

16 Ne5
This was the moment for a tactical opportunity with 16 d5. One has to admire
Anderson’s aggressive approach, but I think he misjudged the potential of his position at
this point. White should be thinking of equality! After 16 d5, a possible line is 16 … exd5
17 Nxd5 Nxd5 18 Qd3! (the point) 18 … Bxh2+ 19 Nxh2 Nf6 20 Ng4 Qxd3 21 Nxf6+
gxf6 22 Bxd3 Kg7. It is not easy to judge this position, but I would say it is approximately
equal.
16 … Ne7 17 Ng4
Or 17 f3 Rc8 18 Ne4 Nxe4 19 fxe4 Bxe5 20 dxe5 Qc7 21 Rd6 Ng6, which is leading
White nowhere.
17 … Nf5 18 Nxf6+ Qxf6 19 Qg4 Qh4
Nice and easy, defusing any attack. 19 … Nxe3 20 fxe3 Qe7 was also very sensible,
when White doesn’t really have any pressure and Black can look forward to using the two
bishops.
20 Qxh4 Nxh4 21 d5 Rad8 22 Ba2 e5 23 Bb1 f5
Hawkins has made the defence look effortless.
 24 f3 Ba6 25 Rfe1 Rf6 26 Bf2 Ng6

27 g4?!
27 g3 was maybe more prudent.
27 … fxg4 28 Ne4?!
28 fxg4 Nf4 29 Ne4 Rff8 30 Bg3 Bb8 is a bit better for Black, but nothing like as big
an advantage as Hawkins gets in the game.
28 … Rxf3 29 Nxd6 Rxd6 30 Bxg6 Rxg6 31 d6 Bd3! 32 Rxe5 Rxd6 33 Re7 a6 34
Bg3 Rd5 35 Rc1
35 Be5 Rf7 36 Re8+ Kh7 37 Re6 b5 38 Bc3 Rfd7 39 Kf2 (or 39 Rxa6 b4) 39 … h5
sees Black making typical progress. There is every chance of a winning attack against the
white king.
35 … Bb5 36 a4
It’s possible that White was short of time. 36 Bc7! was the best chance to hold,
although 36 … Kf8! 37 Re4 h5 38 Bxb6 Rd2 keeps up the pressure.
Not, however, 36 Rcc7? Rf1+ 37 Kg2 Rd2+.
36 … Bxa4 37 Rc8+ Kh7 38 Be5?
38 Rcc7 was surely the idea, after which Black should run his king up the board with
38 … Kg6! 39 Rxg7+ Kh5 40 Rce7 Rdf5!, when White’s brief initiative is over and he
now faces a slow death.
38 … Rd1+ 39 Kg2 Kg6 40 Rxg7+ Kh5 41 Rxg4 Kxg4 42 Rc4+ Kf5 43 Kxf3 Bb5!
0-1
Patience and attention to detail were on show, two key defensive skills.

 Chapter Five
2 c4

1 e4 c6 2 c4

The lines after 2 c4 can easily transpose back into the Panov-Botvinnik Attack, but they
may not and this variation often has an independent flavour. Isolated queen’s pawn
positions are the order of the day.
2 … d5
Black can play many moves here, but in this book we concentrate on the standard Caro-
Kann approach.
3 exd5 cxd5 4 cxd5 Nf6
The first parting of the ways. 4 … Qxd5 is a bit more risky. Nevertheless, the position
after 5 Nc3 Qd6! is quite okay, as we will see. 5 … Qd8 is also playable.
5 Nc3
Unique variations occur after both 5 Bb5+ and 5 Qa4+. We will take a look at each.
5 … Nxd5 6 Nf3
One glance at this position will show us that a transposition to the Panov could be about
to take place. We will take a look at some options where it might not. All in all, 2 c4 is an
interesting move leading to a complex game, where Black must stay alert.
Game 56
J.Carstensen-K.Rasmussen
Danish Championship, Skorping 2017

1 e4 c6 2 c4 d5 3 exd5 cxd5 4 cxd5 Qxd5

4 … Qxd5 is definitely playable, but it is obvious that Black will have to exercise care,
as he is presenting White with a free tempo. Positions reached can resemble the 2 c3
Sicilian or the Queen’s Gambit. I prefer 4 … Nf6.
5 Nc3

5 … Qd6!
The most active square. Black can also go back to base and try to snipe at the IQP but,
to me, 5 … Qd8 seems negative. White finds himself in a favourable transposition to the
QGA, with an important extra tempo and that counts; e.g. 5 … Qd8 6 d4 Nf6 7 Nf3 e6 8
Bd3 Nc6 9 0-0 Be7 10 Qe2 and now:
1) 10 … Nxd4 11 Nxd4 Qxd4 12 Rd1 is White’s idea, and the pawn grab seems to be
ill-advised. I’m not sure I see a satisfactory defence:
1a) After 12 … Qg4 13 Bb5+ Kf8 14 Qe5 a6 15 Be2 Qf5 16 Qxf5 exf5 17 Bf3 Black may be able to survive, but he
will give himself headaches completing his development, despite the extra pawn.
1b) 12 … Qh4 13 Bb5+ Bd7 14 Bxd7+ Nxd7 15 Be3 a6 16 Qf3 Rb8 17 Ba7!.
1c) 12 … Bd7 13 Bb5 Qc5 (or 13 … Qb6 14 Bxd7+ Nxd7 15 Be3 Qa6 16 Qg4 with an
attack) 14 Be3 Qf5 15 Rxd7! Nxd7 16 Rd1 0-0-0 (or 16 … Rd8 17 Bxa7) 17 g4! Qe5 18
Rc1 and Black is in trouble as 18 … Kb8 is met by 19 Bxd7 Rxd7 20 Bxa7+.
 A close look at these lines is advised, as they demonstrate how easily a White initiative
may flow after the pawn is sacrificed and accepted.
2) 10 … 0-0 11 Rd1 Nb4 12 Bb1 Nbd5 13 Ne5 Bd7 14 Qd3 Bc6 15 Bg5 g6 16 Qh3
Rc8 17 Bc2 Nxc3 18 bxc3 Be4 19 Bb3 Nd5 20 Bd2 Bg5 21 c4 Bxd2 22 cxd5 Rc3 23 g3
Bf5 24 Qg2! Qg5 25 h4 Qh6 26 g4 (26 dxe6! is convincing: 26 … Bxe6 27 Qe4! Bxb3 28
axb3; Black cannot extricate her bishop and faces the winning threat of Ng4) 26 … Rfc8
27 gxf5 exf5 28 d6 Bf4 29 d7 Rf8 30 Nxf7 Qxh4 31 d8Q Rxd8 32 Nxd8+ 1-0 was
E.Pähtz-H.Sahin, European Cup, Antalya 2017. A typical IQP game, where Black never
quite equalized.
6 d4 Nf6 7 Nf3 e6 8 Bd3 Be7 9 0-0 0-0 10 Qe2 Nc6 11 Rd1 b6 12 Bg5 Bb7 13 Rac1
Nb4 14 Bb1 Rac8
An absolutely standard IQP position has arisen, with both players marshalling forces.
Black could easily get the upper hand here if White does not demonstrate enough activity.
Note the queen on d6 is radiating power.
15 Ne5 g6

16 Re1
I am not sure that White knows what to do. 16 Re1 looks rather vague, although it is
not that easy to find the right plan. Maybe White should just play 16 a3 Nbd5 17 Nb5 Qb8
18 Bd3, though it is not clear he is getting anywhere after 18 … Rfd8.
16 … Nh5?
16 … Nc6! was indicated, with at least equality for Black.
 17 Bh6
17 Bh6 is not bad, but 17 Nb5! is much better. Perhaps caught in two minds about how
to proceed, White misses his chance. 17 Nb5! is surprisingly strong. Once again, the
variations are worth examining:
1) 17 … Rxc1? 18 Nxd6 Rxe1+ 19 Qxe1 Bxd6 20 g4! Ng7 21 Nc4 wins outright.
2) 17 … Qd5 18 Be4 Qxe4 19 Qxe4 Bxe4 20 Bxe7 Nxa2 21 Rxc8 Rxc8 22 Nd6 Rc7 23
Bd8 is also winning.
3) 17 … Qd8 18 Rxc8 Bxc8 19 Bh6 Re8 20 Qf3 Bf6 (after 20 … Nf6 21 a3 Nbd5 22
Nxa7, White wins a pawn) 21 Rc1! Nd5 (or 21 … Bxe5 22 dxe5 Nd5 23 Nd6) 22 Nxf7!,
setting up for a crushing fork on d6.
17 … Rfe8 18 Rcd1 Bf8
Black is happy to offer exchanges.

TIP: When you have the superior pawn structure, consider trying to exchange

pieces and to angle for a better endgame.

19 Bxf8 Rxf8 20 a3 Nd5 21 Ne4 Qe7 22 g3 f5 23 Nd2 Ng7 24 Ndf3 g5!?

Wow! I am not sure anyone expected that. Black has ambitious intentions of opening
the long diagonal and checkmating on g2 or h1. 24 … Ne8 25 Ba2 Nd6 was a more
modest plan.
25 Ba2 g4 26 Nh4 Rc7 27 Nd3 Qf6 28 Ng2 Rfc8 29 Ne3 Ne7 30 Nf4?
For the second time, White misses the moment. He should play 30 d5!, breaking
Black’s control of the long diagonal. As far as I can see, White is then better: 30 … Nxd5
(or 30 … Bxd5 31 Bxd5 Nxd5 32 Nxd5 exd5 33 Nb4 Re8 34 Nxd5 Qd6 35 Qd2 Rxe1+ 36
Qxe1) 31 Nxd5 exd5 32 Nf4 Rc2 33 Qd3. In these variations, White has the upper hand,
with the black king looking very exposed.
30 … Bf3! 31 Qd2 Bxd1 32 Nxd1 Kh8 33 Bxe6 Rd8 34 d5 Ng6 35 Nc3 Nxf4 36
Qxf4 Re7 37 Ne4 Qg6 38 Nd6 Nxe6 39 dxe6 Rxe6 40 Rxe6 Qxe6 41 Nxf5 Rd1+ 42
Kg2 Qc6+ 0-1

Most players prefer the initiative over structure in isolated queen’s pawn positions and
would be happy to take White in this variation. However, as we have seen, Black has
opportunities to station his pieces sensibly and attain a good game. I could be persuaded to
give 5 … Qd6 a try myself.

Game 57
J.Granda Zuniga-O.Leon
Panama City (blitz) 2017

1 c4 c6 2 e4 d5 3 cxd5 cxd5 4 exd5 Nf6 5 Bb5+!?



This is our first sight of 5 Bb5+. White’s intentions are disruptive.
5 … Nbd7
5 … Nbd7 is the best move. White is hoping for 5 … Bd7, when the retreat 6 Bc4!
leaves Black in an awkward situation. It is not so easy to develop the queenside pieces to
ideal squares, and White may be able to keep his pawn on d5 with simple moves such as
Qb3 and Nc3.
6 Nc3 g6
Instead, 6 … a6!? 7 Qa4!? (7 Be2 would be my choice) 7 … Rb8 8 Bxd7+ Bxd7 9 Qf4
leads to a very weird situation, typical of this line. White is happy to play an unorthodox
game of chess. My own feeling is that Black is okay, despite the extra white pawn, and
that with … g7-g6, … Bg7 and … 0-0 he can complete development in comfort.
7 d4 Bg7 8 d6!
If Black is allowed to regain his pawn in peace, for instance with … 0-0, … Nb6 and …
Nbxd5, then life will be very easy. White must not give him that time.

TIP: Disrupt enemy plans whenever you can!

8 … exd6
Maybe Black should just turn it into a real sacrifice by castling. I’m not sure whether he
would have enough compensation after 8 … 0-0! 9 dxe7 Qxe7+ 10 Nge2 a6, but it would
be more fun than the game and he would be much more active too.
9 Qe2+! Qe7 10 Bf4 d5 11 Bd6 Qxe2+ 12 Ngxe2



Suddenly, Black is in a serious mess with his king stuck in the middle. He
underestimated the subtle strength of White’s opening plan.
12 … a6 13 Ba4 b5 14 Bb3 Bb7
Not the greatest square, but how else to get his rooks into the game?
15 a4!
Jabbing at the precarious black pawn structure. In effect, this is the winning move.
15 … bxa4 16 Bxa4 Ne4
16 … 0-0-0 17 Rc1 is curtains for Black.
17 Nxe4 dxe4 18 0-0
White has all the time he needs with the black king so exposed.
18 … Bf8 19 Be5 Rg8 20 Rfc1 Bd5 21 Nf4 1-0
A rout that makes us think a little bit more about 5 Bb5+. Black must certainly show
more care than he did here.

Now comes a much better representation of the Black cause.

Game 58
J.Rapport-J.Houska
Isle of Man 2017

1 c4 c6 2 e4 d5 3 exd5 cxd5 4 cxd5 Nf6 5 Bb5+ Nbd7 6 Nf3 a6 7 Ba4 g6 8 Nc3 Bg7

Black is simply aiming to castle and to regain the pawn on d5 via … Nb6, … Nbxd5, or
… b7-b5 and … Bb7 followed by the same idea. If she is allowed to carry out this plan,
White will have no positional compensation for the isolated pawn on d4. So White must
find a way of taking the initiative.
9 0-0
9 d6 is not as effective as it was in the previous game: 9 … b5! 10 Bb3 Nc5! 11 d4
(after 11 dxe7 Qxe7+ 12 Qe2 Qxe2+ 13 Kxe2 Nxb3 14 axb3 Bb7, White is in a shocking
mess) 11 … Nxb3 12 Qxb3 Qxd6 and Black stands better.
9 … 0-0 10 Re1 Nb6!
10 … b5 is equally effective, even though the c6-square is weakened. V.Zigo-E.Gibiec,
Czech League 1996, continued 11 Bb3 Nb6 12 Ne5 Bb7 13 Nc6! Bxc6 14 dxc6 Rc8 15
Qf3 Qd6 (Black is trying to get the pawn back, but Houska’s method seems much simpler
to me) 16 a4! b4 17 a5 Na8 18 Nd5 Nxd5 19 Qxd5 Qxc6 20 Qxc6 Rxc6 21 Rxe7 Rfc8
(Black has sufficient activity for the pawn) 22 Re1 Nc7 23 d3 Nb5 24 Ba4 Rc5 25 Be3
Nd4 26 Kf1 Rxa5 27 Rac1! (White is sensible enough to realize he can only make a draw
at best from such a position and plays accordingly) 27 … Rac5 28 Rxc5 ½-½. Not an
especially enjoyable experience for White, and perhaps it is the discovery of such games
that is enough to put players off 5 Bb5+.
11 Bb3 Nbxd5 12 d4 b5 13 Nxd5 Nxd5 14 Bg5 Bf6 15 Qd2 Bb7

The game is equal. White is active enough and Black has the usual long-term pressure
against d4. There is enough tension in the position for the better player to win. White
kicks off the action.
16 h4!?
16 Rac1 is what I would classify as the ‘normal move’ here. Maybe Rapport didn’t
think this was good enough to apply pressure.
16 … e6 17 Rac1 Be7 18 a3
Preventing … Bb4.
18 … Re8 19 Ne5 Qd6 20 Ng4
White continues to play well, forcing Houska to weaken her king position and the e6-
pawn.
20 … f6 21 Bh6 Rad8?!
I think 21 … Rac8 is a better move, not fearing exchanges.
22 Ne3
22 Qe2! turns the heat up. After 22 … Kf7 23 Qf3! Rc8 24 Rxc8 Rxc8 25 h5, White
has a serious initiative, with the serious threat of h5xg6 followed by Ne5+.
22 … Rd7 23 Qe2
Only slightly too late, but late enough to enable Black to defend.
23 … Bf8! 24 Bxf8 Kxf8 25 Bxd5 Bxd5 26 Qg4 Kg7 27 h5 g5 28 Rc5
28 h6+!? could lead to a confusing position: 28 … Kxh6 (28 … Kg6 is safer) 29 Nf5+
exf5 30 Qh3+ Kg6 31 Rxe8 h5 32 Rh8 h4. In time trouble, this would have been difficult
for both sides to play.
28 … Ba8 29 Rd1
Now 29 h6+!? Kf8! 30 Rec1 Red8 31 d5 exd5 32 Nf5 Qf4 33 Qxf4 gxf4 34 Nd4 Kf7
gives Black enough counterplay, but White should have preferred this to the game.
29 … Qf4 30 Qe2 Red8

31 h6+?
This time it is imperative to play 31 d5 right away, with a draw most likely after 31 …
Bxd5 32 Nxd5 Rxd5 33 Rcxd5 (33 Rdxd5 Rxd5) 33 … Rxd5 (or 33 … exd5? 34 Qe7+)
34 Rxd5 exd5 35 Qe7+ Kh6 36 Qf8+ Kxh5 37 Qf7+ Kg4 38 Qe6+ Kh5 39 Qf7+.
31 … Kg6!
Black’s king is now safe and she is getting ready to take on d4.
32 d5 Bxd5 33 Nxd5 exd5
Can this be won against perfect defence by White? Probably, but the next move makes
it a bit easier.
34 Rc6
34 Rd3 is better.
34 … Qe4! 35 Qd2 d4 36 Rxa6 d3 37 Re1 Qf4 38 Qxf4 gxf4 39 Rc6??
And here 39 Rd1 Rd5 40 a4. The text move is just a blunder.
39 … d2 40 Rd1 Re7 0-1
41 Kf1 Rde8 is curtains.
 A tense, complicated game, quite typical of the variation.

Game 59
A.Zhigalko-R.Hovhannisyan
Asrian Memorial, Yerevan 2016

1 c4 c6 2 e4 d5 3 cxd5 cxd5 4 exd5 Nf6 5 Qa4+

A disruptive check, in the manner of 5 Bb5+.

5 … Nbd7!
After 5 … Bd7, White intends 6 Qb3 which scores quite well in practice.
6 Nc3 g6 7 Nf3 Bg7 8 Bc4
White stations himself ready to meet the inevitable … 0-0 and … Nb6. If he allows an
easy recapture on d5, the excursion of his queen will have been wasted.
8 … 0-0
8 … a6! looks rather good to me, intending … 0-0, … Rb8 and then … b7-b5. This
could be the best way to meet White’s system.
 9 d3 Nb6 10 Qb3 Bg4

We must take stock. White is holding the pawn on d5 for all he is worth and Black is
developing activity. I guess Black’s plan is to move his queen to either c7 or d7 and bring
his rooks to the centre. There is also the threat of … Bxf3.
This position seems quite difficult for White, as he is yet to castle, but plenty of strong
players have entered this line thanks to the next move.
11 Ne5! Qc7
Keeping the bishop might seem like a good idea, but after 11 … Bf5 12 0-0 Nfd7 13
Nxd7 Qxd7 14 Be3 Black is yet to demonstrate how he will regain his pawn and, if he
can’t, White is catching up on development fast.
12 f4 Qc5 13 h3 Bc8 14 Ke2!? Nh5 15 Kf3!? e6
Believe it or not, this game had a predecessor: S.Kozhuharov-M.Butuc, Marianske
Lazne 2008, continued 15 … Qc7 16 Re1 Bxe5 17 fxe5 Qd7 18 Kf2 Qf5+ 19 Kg1 (a
strange sequence of moves has resulted in White castling artificially and Black failing to
regain his pawn; Black is clearly much worse now) 19 … Nf4 20 Bxf4 Qxf4 21 Re4 Qg5
22 h4 Qh6 23 Rf1 Bf5 24 Ref4 Nd7 25 d4 (throughout, White keeps his central pawns
intact) 25 … Nb6 26 Be2 Rad8 27 a4 Rb8 28 a5 Na8 29 Qb4 Rfe8 30 Bb5 Rf8 31 Qxe7
b6 32 Ba4 b5 33 Bb3 Qg7 34 d6 h5 35 Nd5 Kh8 36 Nf6 1-0.
16 g4 exd5 17 Nxd5 Nxd5 18 Bxd5 Bxe5 19 gxh5 Bd4 20 hxg6 hxg6 21 Bd2
 After all the previous exchanges, White is still a pawn up. Hovhannisyan throws
another pawn on the fire to fan the flames.
21 … Rd8! 22 Bxf7+ Kg7 23 Rae1 b5 24 Re4?

24 … Bb7?
If only he had found 24 … Qh5+! White would have been in serious trouble: 25 Kg3
(or 25 Kg2 Bf5!) 25 … Bb7 26 Be6 Bxe4 27 dxe4 Qe2.
25 Be6!
Recovery is now possible …
25 … Qh5+ 26 Bg4
… because White has this resource.
26 … Qh4 27 Rh2 Bd5 28 Qc2 Bb6 29 Rhe2 Rd6 30 Be1
We are witnessing some serious grovelling by White, and there is hope, as the black
king is also rather open.
30 … Qd8 31 Kg3 Bxe4 32 dxe4 Kh7 33 Kh2 Rd3 34 Rg2 Qd6 35 Bg3 Rxg3 36
Kxg3 Be3 37 e5 Qd4 38 Qc7+ Kh6 39 Qf7 Qc5 40 Bh5!
White fires the final shots.
40 … Rg8
Or 40 … Qc6 41 e6 Qe8 42 Bxg6.
41 Qxg8 Bxf4+ 42 Kxf4 Qd4+ 43 Kf3 Qd3+ 44 Kf2 Qd4+ 45 Kf1 Qc4+
The internet one-minute addict plays such moves.
 46 Qxc4 1-0
To be sure, both 5 Bb5+ and 5 Qa4+ introduce interesting and often obscure play,
though objectively Black has no reason to fear them.

Game 60
L.Trent-K.Arkell
British Championship, Liverpool 2008

1 e4 c6 2 c4 d5 3 cxd5 cxd5 4 exd5 Nf6 5 Qa4+ Nbd7 6 Nc3 g6 7 Bc4 Bg7 8 d3 0-0 9
Nf3 a6

Keith is happy to play these gambits and 1 e4 c6 2 d4 d5 3 exd5 cxd5 4 c4 Nf6 5 Nc3
g6!? is another of his favourites. It isn’t easy for White to keep control of the position and
if Black regains the pawn, he will usually retain the superior pawn structure.
10 Qa3
To prevent … b7-b5.
10 … Nb6 11 Qb3 Nxc4 12 dxc4 b5!
He has to destroy the white central pawns at all cost, or else White will simply build up
on the e-file and nail Black to e7.
13 0-0 bxc4 14 Qxc4 e6 15 Bg5

15 … h6!?
15 … exd5 16 Qh4 Qb6 17 Rad1 is a model of what White should be doing and so
Arkell comes up with an improvement. The game O.Gschnitzer-J.Kick, Oberwinden 2007,
continued 17 … Bb7 18 Be3 Qd8 19 Qb4 Qb8 20 Bd4 a5 21 Qa4 Rc8 22 Be5 Qa7 23 Qh4
Rc4 24 Nd4 Ne4 25 Bxg7 Kxg7 26 Nxe4 dxe4 27 Ne6+ Kg8 28 Qf6 1-0.
16 Bxf6 Qxf6 17 Rfe1
17 d6 would have been the first move to consider but, of course, this unleashes the
bishop pair after 17 … Bb7 18 Qg4 Bxf3 19 Qxf3 Qxf3 20 gxf3 Rab8. A typical line
where now only Black can be better.
17 … Rb8 18 Rad1
18 Rab1 would have been more testing.
18 … exd5 19 Nxd5 Qxb2 20 Ne7+ Kh7 21 Nc6 Rb6 22 Nd8?!
22 Ncd4 Rb4 23 Qc5 was a better chance. The knight on d8 goes into the hole, never to
return.
22 … Qc3!
Keith knows exactly when to trade and when not. We now enter the technical phase,
where he excels.
23 Qd5
23 Qxc3 Bxc3 24 Re7 is answered by 24 … Rxd8 25 Rxd8 Rb1+ 26 Ne1 Be6!,
winning. Meanwhile, 23 Qe4 Qf6 24 Qe7 Bg4 is just clearly better for Black.
 23 … Qf6 24 h3 Rb2 25 a4 Rc2
Cleverly keeping the knight locked in.
26 Re3 Qb2 27 Rf1 Rc7 28 Nc6 Bb7 29 Nfe5 Qc2!

The pressure on the white knights is too great to withstand.

30 Qd6 Rxc6 31 Nxc6 Qxc6 32 Qg3 Qxa4 33 Rc1 Bd5 34 Qd6 Be6 35 Rc7
White plays on, but without any hope at all. He is just cheesed off.
35 … Qa1+ 36 Kh2 Qa2 37 f4 Qd5 38 Qxa6 Qd4 39 Rf3 g5 40 Rc1 gxf4 41 Qb5
Be5 42 Re1 f6 43 Qb1+ Kg7 44 Rd1 Rb8 45 Qc1 Qc4 46 Qa3 Bd5 47 Qe7+ Kg6 48 h4
Bxf3 49 Rd7 Be4 0-1
This was Keith Arkell at his best, creating complications on his own terms and then
ultra-precise in the technical phase.

Game 61
V.Nevednichy-J.Grant
European Team Championship, Heraklio 2017

1 e4 c6 2 c4 d5 3 exd5 cxd5 4 cxd5 Nf6 5 Nc3 Nxd5 6 Nf3

White rejects a spoiling check in favour of straightforward development. Delaying or
even omitting d2-d4 adds an original element to the position.
6 … Nxc3!?
Not at all a bad move, by which Black usually angles for a Grünfeld-like situation after
7 bxc3 g6. White inherits hanging pawns rather than the regular IQP.
 7 bxc3 Qc7!?
A bit unusual – but, of course, Black can still play … g7-g6.
8 Bb5+!?

White is determined to make the game original. Naturally, 8 Bb5+ cannot be a bad
move, developing and preparing to castle
8 … Bd7 9 a4 e6 10 0-0 Bd6 11 Ba3
A new move, which I do not believe to be dangerous. Exchanges do not help the player
with hanging pawns and here White is positively encouraging exchanges.
11 Re1 0-0 12 Qb3 Nc6 13 d4 looks a little bit more like it for White, but he was
outplayed in the game R.Schoorl-A.Dreev, European Blitz Championship, Warsaw 2013:
13 … Rac8 14 Bd3 (unfortunately for White, 14 c4? is a blunder due to 14 … Na5) 14 …
Na5 15 Qc2 h6 16 Bb2 Rfd8 17 Qe2 Bc6 (Black has posted his pieces in an ideal way and
stands better already; White relies on the pawn advances c3-c4 and d4-d5 to make the
position work and it is just not happening for him, so we can conclude that the idea
starting with 7 … Qc7 is a pretty good one) 18 Bc2 Bd5 19 Qd3 f5 20 h3 Bf4 21 g3?
(maybe 21 Rab1 was the best of a bad job)

TIP: If you have hanging pawns, they need to be mobile to be effective. White’s
pawns are nailed down in this game and that is why he is losing.

21 … Bxg3 (naturally!) 22 Ne5 (22 fxg3 Qxg3+ 23 Kf1 Bc4 or 23 Kh1 Bxf3+) 22 …
Bxe5 23 dxe5 Nc4 24 Bc1 Nxe5 25 Qg3 Nf3+ 0-1. White should be making his
opponent’s life a lot harder than this.
11 … 0-0
After 11 … Bxa3 12 Rxa3 0-0 13 Qb1 (or 13 d4 Bc6 14 Ne5 Nd7 15 Bxc6 bxc6 16
Nxd7 Qxd7) 13 … h6 14 Rb3 Bxb5 15 Rxb5 (or 15 axb5 Nd7 15 … b6 16 a5 Nd7, White
has no advantage at all.
12 Qe2 Bc6
I guess he is not taking on a3 because he thinks White can get some pressure down the
b-file with Rb3 and Rfb1. I don’t see the pressure: thus 12 … Bxa3! 13 Rxa3 Bc6 14 Rb3
Rd8 15 Rfb1 Bd5! 16 Rb4 b6 and Black is OK.
13 Bxd6 Qxd6 14 Nd4 Nd7 15 a5 Qd5 16 Nxc6 bxc6 17 Bc4 Qd6 18 Rfb1 Rab8

Black is defending well. White is trying for a microscopic edge, but that is the limit of his
ambition.
19 Qe3 c5 20 g3 Ne5 21 Be2 Qd5 22 a6 Nc4 23 Bxc4 Rxb1+ 24 Rxb1 Qxc4 25 Rb7
Qxa6 26 Rc7
I think we can all agree that Black should not lose this position. It may have been down
to the clock or the intimidating difference in rating.
26 … Rb8 27 Kg2 Rc8
27 … h6 seems okay, not fearing 28 Rxf7 Kxf7 29 Qf4+ Kg6 30 Qxb8 Qc6+ 31 Kg1 a5
which is still equal, but Black is the one playing on.
28 Qd3 c4 29 Qd7 Rf8 30 Rxa7 Qc8 31 Qd6 h6 32 Kf1 Qd8 33 Qxd8 Rxd8 34 Ke2
Rc8 35 h4 Rc5
 As White can force a passed pawn by playing d2-d3, he is able to create some problems
for Black. I think Grant has to go for kingside counterplay rather than passively defend.
Thus 35 … g5! looks correct, when I don’t think Black is in too much danger; e.g. 36
hxg5 hxg5 37 d3 (or 37 Ra5 f6 38 Ra7 Rc5 39 g4 Re5+) 37 … cxd3+ 38 Kxd3 Rd8+ 39
Ke4 Rc8 40 Kd4 Rd8+ 41 Kc4 Rd2 and so on.
36 f3 g6 37 Rd7 Kg7 38 d4 cxd3+ 39 Kxd3 Kf6 40 c4 e5 41 Rb7 Rc8 42 Kc3 Ke6 43
g4 Ra8 44 Rb6+ Ke7 45 c5 h5 46 g5
The fixing of the pawns is unwelcome for Black.
46 … Ra4 47 Rb4 Ra3+ 48 Kc4 Rxf3 49 Kd5 f6 50 Rb7+ Kd8 51 Rg7 fxg5?
Maybe a check was right: 51 … Rd3+! 52 Kc4 (52 Ke6 fxg5 53 hxg5 e4 54 Rxg6 e3 or
52 Kc6 fxg5 53 hxg5 e4) 52 … Rd4+ 53 Kb5 fxg5 54 hxg5 e4 55 Rxg6 e3 56 Re6 Rg4 57
Rxe3 Rxg5, in all cases looking drawn.
52 hxg5 e4 53 Rxg6 e3 54 Kd6! Rf8 55 Rg7 Rh8
55 … e2 is met by 56 Rd7+ Kc8 57 Re7 Rd8+ 58 Kc6.
56 Ra7 Ke8 57 Ra8+ 1-0
Extraordinarily frustrating for Grant.

 Chapter Six
The Exchange Variation

1 e4 c6 2 d4 d5 3 exd5 cxd5 4 Bd3 Nc6 5 c3

This is the Exchange Variation, a firm favourite with club players. White’s development
plan is simple, straightforward and promises an advantage if Black is at all careless.
Having said that, the line is rather quiet and does not pose any real problems. Let us first
see White’s standard plan, which is pretty devastating if he is allowed to carry it out.

Game 62
R.J.Fischer-T.Petrosian
USSR vs. Rest of the World, Belgrade 1970

As soon as one of the strongest players in the world shows a liking for an opening
variation, the popularity of that variation increases. After Fischer’s famous victory in the
game to come, the Exchange Variation became all the rage.
1 e4 c6 2 d4 d5 3 exd5 cxd5 4 Bd3 Nc6 5 c3 Nf6 6 Bf4
White often develops his bishops before his knights in this line. His overall plan is a
kingside attack. He will build up further with Nd2 and Ngf3, castle and station a knight on
e5, behind which he will mass his forces. Then comes a kingside pawn storm. Countless
games have been decided in White’s favour using these simple ideas and that is what the
average player likes.
6 … Bg4 7 Qb3! Na5
I’m not going to go too deeply into the theory just yet or even at all, suffice it to say
that Black has other moves here.
8 Qa4+ Bd7 9 Qc2 e6 10 Nf3 Qb6
Preparing … Bb5.

11 a4!
An important new move at the time, preventing … Bb5. The immediate thought is that
the b3-square has become weakened, but Fischer has everything covered.
11 … Rc8
Black can occupy b3 in two ways. After 11 … Qb3 12 Qe2! is good; for example, 12 …
Bxa4? 13 Rxa4 Qxa4 14 Bb5+. Meanwhile, if 11 … Nb3 then 12 Ra2 Rc8 13 0-0 Be7 14
Nbd2 ejects the knight from b3, after which White holds an advantage in development.
12 Nbd2 Nc6 13 Qb1
Black was threatening … Nb4.
13 … Nh5 14 Be3 h6 15 Ne5!
White puts his knight into e5 a little earlier than usual in this line, but of course there is
the knight on h5 for Black to consider. If he can, White will follow up with f2-f4-f5,
opening up the kingside.
15 … Nf6 16 h3 Bd6 17 0-0




17 … Kf8?!
Black should certainly have castled and invited White to show how he is going to
attack; e.g. 17 … 0-0 18 Ndf3 Qc7 19 Re1 Be8 20 Bf4 a6 and Black is very solid.
18 f4 Be8
If 18 … Nxe5, White has a tactic: 19 fxe5 Bxe5 20 a5!, winning.
19 Bf2! Qc7 20 Bh4 Ng8 21 f5
Fischer rarely missed an opportunity. 21 f5 is very well timed, is exactly what White is
hoping for in this line, and puts Black under severe pressure. Petrosian invited the pressure
by playing moves like 17 … Kf8.
21 … Nxe5 22 dxe5 Bxe5 23 fxe6 Bf6 24 exf7 Bxf7 25 Nf3!
Transferring another piece into an attacking position. With his rook on h8 completely
bottled up, Black struggles to defend.
25 … Bxh4 26 Nxh4 Nf6 27 Ng6+ Bxg6 28 Bxg6 Ke7
Running for the hills, but the hills are a long way away.
29 Qf5 Kd8 30 Rae1 Qc5+ 31 Kh1

31 … Rf8
Black can try to defend in two other ways.
1) Firstly 31 … Rc6, which should be met by 32 b4! Qxc3 33 b5 Rc7 34 Qe6, when I
think White is just winning as it is so hard to get the black pieces working together.
2) Secondly, 31 … Rc7, which should again be met by 32 b4! Qxc3 (or 32 … Qb6 33
Re6 Rc6 34 Rfe1) 33 Qf4! Qc4 (or 33 … Kc8 34 Rc1) 34 Qd6+ Rd7 35 Qb8+ Qc8 36
Qxa7.
One cannot blame Petrosian for not liking either of these lines, but what he does in the
game leads to a quick defeat.
32 Qe5! Rc7 33 b4! Qc6 34 c4!
Opening lines decisively.
34 … dxc4 35 Bf5 Rff7 36 Rd1+ Rfd7 37 Bxd7 Rxd7
Or 37 … Nxd7 38 Rfe1, winning.
38 Qb8+ Ke7
38 … Qc8 39 Rxd7+ Nxd7 40 Qd6 again wins.
39 Rde1+ 1-0
The prospect of this type of attack and the easy way it can be set up keeps 3 exd5 and 4
Bd3 popular today.

For every standard plan by White, there is a standard plan for Black. The coming game
shows what might happen when the standard plans in this variation clash.

Game 63
V.Kovalev-A.Riazantsev
Russian Rapid Team Championship 2017

1 e4 c6 2 d4 d5 3 exd5 cxd5 4 Bd3 Nc6 5 c3 Nf6 6 Bf4 Bg4 7 Qb3 Qd7!

A very solid move, defending b7 without moving a knight to the edge. It’s an idea
which pretty much pours a cold shower over the white set-up.
8 Nd2 e6 9 Ngf3 Bxf3!
Black is very happy to exchange, reducing any attacking force.
10 Nxf3 Bd6! 11 Bxd6 Qxd6 12 0-0
12 Qxb7 Rb8 13 Qa6 0-0! is one of the key points of the line. By giving up a pawn,
Black assumes a temporary initiative, quite enough to gain equality at least and maybe
more against an opponent who overestimates the position. Now:
1) 14 0-0 is the ‘common sense’ move, when play can continue 14 … Rb6 15 Qa3
Qxa3 16 bxa3 Rc8 17 Rab1 Na5 18 Rxb6 axb6 19 Rc1 g6, reaching a situation which is
about equal. Black has all the positional trumps, but White has an extra pawn.
2) 14 Bb5 Rb6 15 Qa4 Ne7 16 b3 (otherwise … a7-a6! wins the b2-pawn) 16 … Ne4
17 Rc1 Rc8 and Black again holds the initiative and this time it’s serious.
3) It must be pointed out that 14 Rb1 Rb6 15 Qa3 Qxa3 16 bxa3 Rxb1+ 17 Bxb1 Rb8 is
no good for White at all.
 12 … 0-0 13 Rae1 Qc7 14 Qc2 Rac8 15 Qe2 a6 16 g3 b5!

Black’s standard plan is the minority attack against the white queenside. You often see
Black playing … Rab8 and then … b7-b5 in this line. In our current game White has
slowed the process down a bit. Nevertheless, the minority attack is still effective:
essentially, … Rb8, … a6-a5 and then … b5-b4! will provide counterplay.
17 a3 Rfe8 18 Kg2 Rb8 19 h4 h6 20 Rh1 Nd7 21 Bb1 a5 22 Qd3 Nf8 23 b4
White has made some progress towards creating a kingside attack, but the knight on f8
is doing a sterling defensive job and Kovalev decides he must now pause to block the
queenside. It’s hard to judge whether 23 b4 is necessary or not, and it is certainly a move
that plays into Black’s hands.
23 … Rec8 24 Re3 Rb7 25 Rhe1 Qb6 26 Nd2
The knight heads for c5 via b3 and Black has to put a stop to that.
26 … a4! 27 Nf3 Rbc7 28 Qd2 Qb8 29 h5 Rb7 30 Nh2 Qd6 31 Ng4 f5 32 Ne5 Nxe5
33 Rxe5 Rb6 34 R1e3 Qd8
 35 g4?
I guess that with more time, Kovalev might have been more circumspect. Basically, 35
g4 is too risky. Just 35 Qe2, maintaining the balance, was called for.
35 … Qg5! 36 f3 Nd7
White has managed to trap his own rook.
37 Rxf5 exf5 38 Bxf5 Rd8 39 Qe2 Nf8 40 Re5 Qf4 41 Re7 Rf6 42 Qd3 Rxf5 43 Qxf5
Qxf5 44 gxf5 Nd7 45 Kf2 Kf8 46 Re6 Rc8 47 Re3 Nf6 48 Ke2 Re8 49 Re5 Nxh5 50
Ke3 Nf6 51 Kf4 Re7 0-1
The opening set-up commencing with 7 … Qd7 is relatively simple to learn and
execute, and in the present day it has taken a bit of the lustre out of the Exchange
Variation.

Game 64
V.Kramnik-V.Ivanchuk
FIDE World Cup, Tbilisi 2017

As soon as White takes on d5 and plays c2-c3, he fixes the pawn structure, which limits
his options. Today, we see many of the top players rejecting this type of clarification early
on in the game, preferring elastic pawn structures, which can change as the game
progresses. Thus the game becomes more difficult to predict. Obviously, even at the
highest level, if White just wants to keep it rock solid he can still choose this line.
1 e4 c6 2 d4 d5 3 exd5 cxd5 4 Bd3 Nc6 5 c3 Nf6 6 Bf4 Bg4 7 Qb3 Qc8!


This is just as good as 7 … Qd7 and casts doubt on White’s ability to gain any
advantage out of the opening.
8 Nd2 e6 9 Ngf3 Be7 10 0-0 Bh5!
Re-routing to g6, snuffing out the active bishop on d3.
11 Rae1 Bg6 12 Bxg6 hxg6


13 h4!?
A new move from Kramnik. White needed a boost, as the normal idea with Ne5 just
leads to an equal game; for example, S.Teichmeister-B.Zhak, correspondence 2008,
continued 13 Ne5 Nxe5 14 Bxe5 Qc6 15 a4 a6 16 h3 b5! (Black responds thematically,
gaining the usual queenside counterplay) 17 Ra1 0-0 18 Bf4 Rfc8 19 Nf3 Ne4 20 Ne5
Qb7 21 Rfd1 g5 22 Bh2 Nd6 23 Re1 Nf5 24 Rec1 Qb6 25 Qd1 bxa4 26 Rcb1 a3 (White
has to struggle to keep on the board here, but with plenty of time for thought and no doubt
a 3500 computer program to back him up, he manages to do so) 27 Qh5 Nh6 28 bxa3 Qd8
29 Rb3 Bd6 30 Rab1 Rab8 31 Nd3 Rxb3 32 Rxb3 Be7 33 Qe2 Nf5 34 Be5 Nd6 35 Bxd6
Bxd6 36 g3 Rb8 37 Qb2 Rxb3 38 Qxb3 Qc7 39 Kg2 ½-½. Games like this one reaffirm
the effectiveness of Black’s opening plan.
13 … a6 14 c4!?
White is trying to make something out of nothing and has to make the radical decision
to change the pawn structure in order to do so. I am not totally sure about White’s play; it
looks most un-Kramnik-like.
14 … a5 15 a3 a4 16 Qd3 0-0 17 c5 b6!
Breaking the cramp quite comfortably, as Black is always going to regain his pawn.
18 cxb6 Qb7 19 Bc7 Nd7 20 Rc1 Rfc8 21 b4 axb3 22 Qxb3 Bxa3
Black is a bit better, but what did Kramnik expect after playing on the queenside? This
is Black’s usual theatre of action. It is noticeable how hard it is for White to start anything
on the kingside at all after the odd 13 h4!?.
 23 Rc3 Be7 24 Rfc1 Nb4 25 Ne5 Nxe5 26 Bxe5 Rxc3 27 Qxc3 Qxb6 28 Nf3 Qd8
The scene is set for a very long grind and I would not like to be Kramnik. Ivanchuk is
never going to agree a draw and it is unlikely that he will make even the slightest mistake
in this simple situation. What are White’s drawing chances? Reasonable, I would say, but
suffering will be the order of the day.

29 Qb3 Rc8 30 Ra1 Rc4 31 g3 Qc8 32 Kg2 Nc6 33 Qb5 Bf8 34 Ra4 Rxa4 35 Qxa4
Nb4 36 Qb5?!
Letting the black queen come to c2 is not a good idea. 36 Bf4 was a little better,
keeping the status quo. If then 36 … Qc2, then 37 Qe8 Qc6 38 Qb8 keeps the white queen
active.
36 … Qc2! 37 Qf1
37 Qe8? Nd3 wins for Black.
37 … Qe4 38 Qe1 f6 39 Bc7 Nc2 40 Qxe4 dxe4 41 Nd2 f5 42 Nb3 Ne1+ 43 Kf1 Nf3
44 Ke2 Kf7 45 Bb6 Bb4 46 Ke3 Ke8 47 Bc5 Bc3
Thanks to the pawn fixed on d4, Black’s bishop is a better piece than its white
counterpart, and this is why Ivanchuk does not exchange.
48 Ba3 Kf7 49 Bd6 Kf6 50 Bc7 Ke7
The task for Black in this endgame is to create a passed pawn. Once that happens,
matters will clarify.
51 Bb8 Bb2 52 Bf4 Kd8 53 Bd6 Kd7 54 Bf8 Kc6! 55 Bxg7 Kd5 56 Bf6 Bc3
White is only able to enjoy winning his pawn back for a short time, as the black pieces
have taken up dominating positions. There is now a direct threat of … Kc4.
57 Ke2 Kc4 58 Nc5 e5 59 Nd7 exd4 60 Bd8 d3+
It’s all over. The end came quickly once the passed pawn was created.
61 Kd1 Bd4 62 Bb6 Bxb6 63 Nxb6+ Kd4 64 Nc8 Ne5 65 Ne7 Ng4 66 Ke1 Kc3 67
Nd5+ Kb3 68 Kd2 Nxf2 69 h5 gxh5 70 Ne3 h4 71 Nxf5 h3 0-1
Well played, Vassily Ivanchuk!

Having seen Black enjoy success by placing the queen both on d7 and c8, there is yet
another good square to which the queen may deploy.

Game 65
V.Fedoseev-Y.Bacallao Alonso
FIDE World Cup (rapid tiebreak), Tbilisi 2017

1 e4 c6 2 d4 d5 3 exd5 cxd5 4 Bd3 Nc6 5 c3 Qc7!?

As this is a Queen’s Gambit Exchange Variation reversed (if you can wrap your head
around that). Black plays the queen to a most active square and prevents Bf4. This line is
yet another issue for White if he wishes to play for the advantage.
6 Ne2
6 h3!? prevents … Bg4, but it is pretty slow. Black may employ a fianchetto to get to a
good game; for example, 6 … Nf6 7 Nf3 g6! 8 Na3 a6 9 Qe2 Bf5! (note this move; Black
inherits doubled pawns but gets rid of a problem bishop and increases his control of the
centre) 10 Bxf5 gxf5 11 Nc2 e6 12 Ne3 Ne4! 13 Nd2, and now Black is absolutely fine
after either 13 … 0-0-0 or 13 … h5 (V.Fedoseev-Y.Bacallao Alonso, FIDE World Cup,
Tbilisi 2017). Yes, the players are the same as featured in our main (but different) game.
6 … Bg4 7 0-0
White mostly plays 7 f3 Bd7 8 Bf4 e5 9 dxe5 Nxe5 10 0-0 Bd6 but does not score very
well with it. Black is active.
7 … e6
7 … Nf6 is another very reasonable approach. A.Bachmann-A.Dreev, FIDE World Cup,
Tbilisi 2017, continued 8 Qe1 g6 9 Bf4 Qb6 10 h3 Bxe2 11 Qxe2 Bg7 12 a4 0-0 13 Re1
Rfe8 14 Na3 Nh5 15 Be3 Qc7 16 Qd2 Rad8 17 Nb5 Qb8 18 Be2 Nf6 19 Bf4 e5!
(activating his entire position) 20 dxe5 Ne4 21 Qe3 Bxe5 22 Bxe5 Rxe5 23 Bf1 Rde8 24
Qf4 Qd8 25 f3 Qb6+ 26 Kh2 Ng5 27 Rxe5 Rxe5 28 Qg3 Ne6 29 a5 Qe3 30 a6 bxa6 31
f4? (Bachmann doesn’t like 31 Rxa6 Rg5 32 Qh4 Ne5, but then 33 Nd4! was playable for
White) 31 … Nxf4 32 Rxa6 Qc1 33 Rxc6 Qxf1 34 Rf6 Re1 35 Qxf4 Qg1+ 36 Kg3 Re3+
37 Kh4 0-1.
8 Qe1 Bxe2 9 Qxe2 Bd6 10 g3 Nge7
A new move, which I am not sure improves over 10 … Nf6 or 10 … h5!?, but the truth
is that all three moves are good. Black has a wide choice.
11 Nd2 0-0-0 12 f4 Nf5 13 Nf3 h5 14 Bd2 Kb8 15 b4

15 … Rhe8?!
For me, the right place for this rook is h8, supporting … h5-h4. Black can play 15 …
h4 right away, but it does not appear very good after 16 Bxf5! exf5 (16 … hxg3 17 Bd3
gxh2+ 18 Kh1) 17 Nxh4 g6 18 Nf3. Instead, the steady 15 … Be7! was the way forward,
preparing … h5-h4 and making room for the knight to drop back to d6 if necessary; e.g.
16 a4 h4 17 g4 Nd6 18 a5 a6.
16 a4 f6 17 a5 g6 18 Rae1 Re7
Black’s plan appears to be 18 … e5, but he finds he cannot play it as 19 Bxf5! gxf5 20
fxe5 fxe5 21 dxe5 Nxe5 22 Nd4! leaves White well on top.
19 Qf2 Rc8
I think it was a good idea to hold up the pawn advance with 19 … a6! 20 Re2 Rc8 21
Rfe1 Nd8. The computers show a small edge to White, but I believe the black position to
be very tough and resourceful. An interesting middlegame lies ahead.
20 a6 b6

21 Ne5!?
Ten out of ten for shock value, and a move which provokes a mistake. Objectively, 21
Re2 was better, steadily increasing White’s positional edge.
21 … Rg8?
A poor response. He has to take that knight, and after 21 … fxe5 22 dxe5 Bxb4! 23
cxb4 Ncd4 24 Rc1 Qd8 25 Rxc8+ Qxc8 26 Rc1 Rc7, we have two well-entrenched
knights versus the bishop pair and an interesting struggle ahead.
22 Nxg6!
White takes his tactical opportunity.
 22 … Rxg6 23 Rxe6 Nh4??
Black has been completely thrown by the events of the last few moves and crashes to
defeat. He has to try 23 … Ng7 24 Rxe7 Nxe7 25 Bxg6 Nxg6 26 Qf3 Qc4 27 Re1 Ne7
and take his chances. There is a long way to go in this position.
24 Bxg6 Nxg6 25 Rxf6 1-0
We have so far noted three separate decent defensive systems for Black in the
Exchange Variation. There are yet more.

My own White experiences with 4 Bd3 are few, far between, and not that memorable.

Game 66
A.Martin-D.Kumaran
British Championship, Dundee 1993

1 e4 c6 2 d4 d5 3 exd5 cxd5 4 Bd3 Nc6 5 c3 g6!?

This can be separated from the line where Black plays … Nf6 first and only then … g7-
g6. In this variation Black’s plans are more ambitious. He wants to complete development
with … Bg7, … Nh6 and … 0-0, after which he will turn his attention to preparing a
central advance via … Nf7, … f7-f6 and … e7-e5. It all sounds rather slow, but White’s
structure is so static that he has problems preventing the plan. Maybe White has to go c3-
c4 somewhere, but it is not so easy to arrange.
6 Nf3
6 Bf4 is also very common; e.g. 6 … Bg7 7 Nf3 Nh6 (or 7 … Nf6 8 0-0 0-0 9 Nbd2
Nh5 10 Be3 Qd6) 8 0-0 0-0 9 Re1 Bf5! (plans can and should be changed; it is worth
pointing out that 9 … f6 10 c4! e6 11 Nc3 is clearly better for White) 10 Bxh6 Bxh6 11
Bxf5 gxf5 12 Nbd2 e6 and there is nothing wrong with Black’s game.
6 … Nh6 7 0-0 Bg7 8 Re1 0-0 9 Bxh6!?
I could see what was coming up and wanted to prevent it once and for all, but
surrendering the bishop is a high price to pay. To be honest, I wasn’t exactly sure what to
do. Today, I would settle for 9 Bf4 f6 10 c4!; back then it was a sheer lack of familiarity
with the line that prevented me from playing this way.
9 … Bxh6 10 Nbd2 Bg7 11 Nb3 Qd6 12 Rc1 Re8 13 h3 Bd7 14 Rc2 Rad8 15 Rce2

Well, White has completed a picture postcard development designed to dissuade Black
from playing … e7-e5. The bad news is it doesn’t prevent that move at all, and perhaps
Dharshan should now have advanced.
15 … Bc8
He does not want to play with an isolated pawn, but 15 … e5! 16 dxe5 Nxe5 17 Nxe5
Bxe5 18 Nd4 a6 19 Bc2 is about equal.
16 Bb5 e6 17 Nc5 f6 18 Qa4 Re7 19 b4 Nb8
Black keeps the tension. I must admit that I had expected 19 … e5 and, again, there is
nothing wrong with this move yet: 20 dxe5 Nxe5 21 Nd4 (or 21 Nxe5 fxe5 22 Rd1 b6 23
Ne4 Qc7) 21 … a6 22 Bd3 b6 23 Ncb3 Rc7.
20 Qc2 b6 21 Nd3 Kh8 22 Qd2 Bd7 23 Bxd7 Nxd7 24 Nf4?
The strategy of keeping the central tension pays off and, with both players getting into
time trouble, I make a mistake. White should add to his positional trumps by playing 24
b5! Nf8 25 Nb4 Rc7 26 Nc6, when the knight cramps Black slightly.
24 … Bh6! 25 Nxg6+
I have to confess that I missed 29 … Qd3! in the coming sequence. Dharshan was very
strong tactically and now he shows it.
25 … Kg7! 26 Nxe7 Bxd2 27 Rxe6 Qf4 28 Nxd5 Qf5 29 Ne7 Qd3!

Nicely played, getting the queen into an active location. By contrast, 29 … Qc2? is met
by 30 R6e2 and Black loses.
30 R1e4 Nf8! 31 Nf5+ Kh8 32 Re7 Bg5?
32 … Qxc3 was much better. I can only think the flags were hanging.
33 Nxg5 fxg5 34 Nh6 Rd7 35 Re8?
Time trouble. Of course I should try 35 Nf7+ Kg7 (35 … Kg8 36 Nh6+) 36 Rxd7!
Nxd7 37 Re7 Qb1+ 38 Kh2 Nf6 39 Ne5+ Kg8 40 Rxa7 Qe4 41 a4, when this might still
be a win for Black, but at least I could have made him work for it.
35 … Kg7 36 Nf5+ Kg6 37 Ne3 Rf7 38 Ng4 Kg7 39 Re3 Qb1+ 40 Kh2 Qxa2
Black consolidates and it is all over.
41 R8e5 h6 42 Re2 Qc4 43 R2e3 Ng6 44 Re6 Nf4 0-1
5 … g6!? can produce complicated positions, but it is not quite as reliable as the other
main lines that we have discussed.

Game 67
G.Murugappan-E.Winslow
San Francisco 2016

Let’s take a look at a final system now, where Black strongpoints e4.
1 e4 c6 2 d4 d5 3 exd5 cxd5 4 Bd3 Nc6 5 c3 Nf6 6 Bf4 g6 7 Nd2 Bg7 8 h3 0-0 9 Ngf3

9 … Bf5!
Black may delay … Bf5 for one move, but it is more or less the same; for example, the
game V.Ivanchuk-A.Riazantsev, Russian Team Championship 2008, continued 9 … a6 10
0-0 Bf5 11 Bxf5 gxf5 12 Ne5 e6 13 Nxc6 bxc6 14 Nb3 Nd7 15 Re1 Kh8 16 Re3 a5 (this
move makes me believe that 9 … Bf5 is more accurate) 17 a4 Qb6 18 Rg3 Rg8 19 Rb1
Bf6 20 Rxg8+ Rxg8 21 Bd6 Qd8 22 Qe2 Nb6 23 Bf4 Nc4 24 Nc5 Bg5 25 Bh2 Bf6 26 b4!
(this was always going to be the issue with the pawn structure; White creates an outside
passed pawn) 26 … axb4 27 Rxb4 Qf8 28 Rxc4!? dxc4 29 Nd7 Qg7 30 Qf3 Qh6 31 Bd6
Rd8 32 Qxc6 Qc1+ 33 Kh2 Bg5 34 Qc7 (the pressure is firmly on Black) 34 … Rxd7? (34
… Kg8) 35 Qxd7 Bf4+ 36 g3 Bxd6 37 Qxd6 Kg7 38 Qe5+ Kg6 39 d5 and White won.
10 Bxf5 gxf5 11 Ne5 Qb6 12 Qb3 Nxe5 13 Bxe5 Qe6
13 … Qc6 is also very reasonable. It looks like Winslow is tempting his lower-rated
opponent to take the b-pawn, or at least spend some time thinking about it. That’s often
the way strong players beat lesser players. They give them things to think about, and
decisions to make, and seem to win without doing anything at all. By making bad calls,
the opponent beats himself.
14 0-0
14 0-0-0 was probably a step too far against the IM.
14 … Ne4
 When Black sets up a structure with pawns on d5 and f5, he is almost always going to
follow up with … Ne4.

15 Bf4!?
It might have been time to trade everything; e.g. 15 Nxe4 fxe4 16 Bxg7 Kxg7 17 f3 and
Black will struggle to win this one.
15 … Qc6 16 Nxe4 fxe4 17 Rfe1 Kh8
Now that the g-file is half-open, a Black attack on that file becomes a real possibility.
18 Rad1 Rad8 19 a4 Rg8 20 Qb5 Qc8 21 Kh1 a6 22 Qb6 Qf5 23 Bg3 Rd7 24 Qb3
Bh6 25 Qc2 e6 26 Kh2 f6!
Move by move, Black has improved his chances, but there was no need for what
happens next.
27 Re3??
Hard to explain. Panic? Time trouble? An out and out blunder? A desire to get to the
bar? 27 Re3 certainly brings the game to a speedy conclusion. The calm 27 Qe2 Rdg7 28
Rg1 was called for. Passive, yes, but Winslow would still have to break through.
27 … Bxe3 28 fxe3 Rdg7 0-1
The Exchange Variation is commonly played, but does not pose a threat to the Caro-
Kann. Black has the choice of several reliable systems of defence. Take your pick!

 Chapter Seven
The Two Knights Variation

1 e4 c6 2 Nc3 d5 3 Nf3

Periodically popular among grandmasters, the Two Knights line has never really
become one of the main Caro-Kann highways. The reason for this is down to 3 … Bg4!,
which is a very solid move and which makes it difficult for White to demonstrate any
advantage whatsoever. However, Black must not underestimate this sensible line.

Game 68
A.Stukopin-I.Nyzhnyk
Kenner 2016

Casual play by Black, or attempts to transpose back into main lines, can often come
unstuck against the Two Knights. Observe here, as Black plays an early … Bf5, after
which the best he can get is a solid, but very passive set-up.
1 e4 c6 2 Nf3 d5 3 Nc3 dxe4 4 Nxe4 Bf5?!
Playable, but not very promising.
5 Ng3 Bg4
It is worth pointing out that 5 … Bg6? is a clear mistake. The analysis is not
complicated either: 6 h4! h6 (or 6 … Nf6 7 h5 Be4 8 Nxe4 Nxe4 9 d4) 7 Ne5 Qd6 (if 7 …
Bh7 then 8 Qh5 forces 8 … g6) 8 Nxg6 Qxg6 9 d4.
6 h3 Bxf3 7 Qxf3 Nf6 8 Bc4 e6 9 0-0 Nbd7 10 Re1 Be7 11 c3 0-0 12 d4 Re8

The best one can hope for as Black is a draw from this type of quiet situation. Basically,
White has the bishop pair and all the prospects, so Black has to settle for a long period of
passive defence. It takes a certain cast of mind to play this type of stuff and is probably
only to be recommended against weaker opposition, where some strategic mistakes can be
expected.
13 Be3
Exchanging pieces does not seem like the right approach, unless one is happy to split
the point. That is exactly what happened in the following, brief snippet: 13 Nh5 Nxh5 14
Qxh5 Nf6 15 Qf3 Nd5 16 Qg4 Nf6 17 Qf3 Nd5 18 Qg4 Nf6 19 Qf3 ½-½, R.Antoniewski-
E.L’Ami, German Bundesliga 2011.
13 … Qc7 14 Rad1 Rad8 15 Bb3 Bd6 16 Bg5 Be7 17 h4
Stukopin starts to inch forward. The computers will show equal, but we all know this is
a bit better for White and it is an edge which cannot be shaken off easily.
17 … a6
17 … h6 18 Bc1 Bd6 19 Ne2 b5 seems okay but is still a bit passive.
18 Re2 b5 19 Rde1 h6 20 Bc1 Bd6 21 Ne4

Now that both … c6-c5 and … e6-e5 have been prevented, White feels he can
exchange off a pair of minor pieces, subtly weakening the black kingside.
21 … Nxe4 22 Rxe4 Nf6 23 R4e2 Be7
It is worth noting that 23 … c5? is met by 24 Bxh6!.
24 Bc2 Nd7
24 … c5 was better.
25 Re4 Bd6 26 g4 Bf8 27 Rf4 Re7 28 h5 e5 29 Qd3!
Black thought he could shuffle around forever, but 29 Qd3 shows him the error of his
ways.
29 … g6 30 hxg6! exf4 31 Rxe7 Bxe7

32 gxf7+??
Extraordinary. Perhaps under time pressure, Stukopin misses 32 Qf5! Nf6 33 gxf7+,
which would have finished the attack nicely. The lines are as follows: 33 … Kxf7 (33 …
Kf8 34 Bxf4 Bd6 35 Bxh6+ Ke7 36 g5! Nd7 37 Qe4+ or 34 … Rd5 35 Bxc7 Rxf5 36
Bxf5 Kxf7 37 Bf4 wins) 34 Bxf4 Qd7 (34 … Qb7 gets mated after 35 Qg6+ Ke6 36 Bf5+
Kd5 37 Qf7) 35 Qg6+ Ke6 36 Bf5+.
32 … Kf8
But not 32 … Kxf7?? 33 Qg6+ Kf8 34 Qxh6+ (or just 34 Bb3) 34 … Kf7 35 Bb3+ Ke8
36 Qg6+ Kf8 37 Qf7 mate.
33 Qh7 Bg5!
Black keeps the f-pawn and this should make all the difference.
34 Bg6 Qb8
After 34 … Qc8!, I do not see how White breaks through; e.g. 35 b3 b4 36 c4 Nf6 37
Qh8+ Ke7.
35 b3 a5 36 c4 b4
36 … Nf6! still wins for Black!
37 Bb2 f3?? 38 d5 Nf6 39 Bxf6 Bxf6 40 d6! 1-0
Blocking the king’s escape.

Game 69
J.Van der Wiel-A.Miles
Brussels 1986

1 e4 c6 2 Nc3 d5 3 Nf3 dxe4 4 Nxe4 Nf6

This usually transposes to main lines. Attempts by White to avoid d2-d4 have not been
that successful.
5 Nxf6+ gxf6
5 … exf6 has been popular with world-class players, but there again White almost
always plays 6 d4. I’ll confine myself to one example, the game I.Nepomniachtchi-
M.Carlsen, World Rapid Championship, Doha 2016: 6 d4 Bd6 7 Be2 0-0 (Korchnoi
played 7 … Na6!? in Game 35) 8 0-0 Re8 9 c4 Nd7 10 Bd3 Nf8 11 h3 Ng6 12 Qc2 Qa5
13 c5 Bb8 14 Bd2 Qd8 15 Rfe1 Be6 16 Bc4 Nf8 17 Rad1 Bc7 18 Qb3 Bxc4 19 Qxc4 Qd7
20 b4 Ng6 21 a4 (White has a tiny edge, dependent on whether he can create a passed
pawn on the queenside and then push it home; this seems unlikely, with Black in full
command of the d5-square) 21 … Ne7 22 b5 Nd5 23 bxc6 bxc6 24 Qa6 g5 25 Nh2 Bxh2+
26 Kxh2 Kg7 27 Kg1 Rxe1+ 28 Rxe1 Rb8 29 Qd3 Rb2 30 a5 h6 31 Bc1 Ra2 32 Bd2 Qb7
33 Re8 Qb2 (Carlsen has outplayed his opponent) 34 Be1 Nf4 35 Qa6 Qxd4 36 Qc8 Ng6
37 Qxc6 Ra1 38 Qe4 Qxc5 39 Kh2 h5 40 Qe3 Qc6 41 Bc3 Ra4 42 f3 Ra2 43 Kh1 Rc2 44
Be1 Qb5 45 Bg3 h4 46 Bh2 Qb1+ 47 Bg1 Qf1 0-1.
6 g3 Bg4 7 Bg2 Qd7
Setting up ideas of … Bh3 or … Qf5 followed by … Nd7-e5.
7 … e6 is a little more passive, after which 8 b3! Nd7 9 Bb2 is a good reply. White is
now a bit better, not only due to the pawn structure, but thanks to his prospects on the
kingside. Without a pawn on d4, Black has little to aim at. P.Keres-L.Pachman, Moscow
1967, continued 9 … Bg7 10 h3 Bh5 11 Qe2 0-0?! (very careless; 11 … Qa5 followed by
castling long was the best way to go) 12 g4 Bg6 13 h4 f5 14 Bxg7 Kxg7 15 h5 fxg4 16
hxg6 gxf3 17 Rxh7+ Kxg6 18 Qd3+ f5 19 Rxd7 Qf6 20 c3 Qe5+ (20 … fxg2 21 Qg3+
Kh6 22 Qh2+ Kg6 23 Qxg2+ Qg5 24 Rg7+ ends the game) 21 Kf1 fxg2+ 22 Kxg2 Kf6 23
Rh1! (Black’s king is turning out to be much the more vulnerable) 23 … Rh8 (or 23 …
Rg8+ 24 Kf1 Rh8 25 Qf3! Rh2 26 Rxb7, winning) 24 Rhh7 Rxh7 25 Rxh7 Rg8+ 26 Kf1
Rg7 27 Qd8+ Kg6 28 Rh3 Qd5?? (28 … f4) 29 Qe8+ Rf7 30 Qh8 1-0. White’s d-pawn
remained nailed to the spot.
8 h3 Bf5 9 Nh4 Be6 10 d3 Na6 11 Be3 h5 12 Qd2 Bd5!

Once the light-squared bishops are exchanged, Black will not feel any real pressure.
13 0-0 Bxg2 14 Kxg2 e6 15 f4 f5 16 Nf3 Bg7 17 c4
He could have tried 17 Rab1, intending b4-b5 if Black castles long.
17 … 0-0-0 18 Rad1 c5 19 h4 Kb8 20 Qe2 Bd4 21 a3 Ka8 22 Bxd4 cxd4 23 b4 Nc7
24 b5 f6 25 a4 Rhg8 26 Rg1 Rde8!
A quick … e6-e5 is going to give Miles all the counterplay he needs.
27 Kh2 e5! 28 fxe5 fxe5 29 Qf2
29 Nxe5!? Qg7 30 Rde1 Re6 31 a5 a6 32 b6 Ne8 may have been okay for White, but
one can see why Van der Wiel would not want to risk it. Even in the final position of this
variation White is very tied up and faces … Nd6-f7.
29 … Rg4 30 Rde1 Qg7 31 a5 Qf6 32 Rg2 f4!
 33 Re2?
Here 33 Kg1 Reg8 34 Rxe5 fxg3 35 Qe2 would be a much better attempt to break the
Black attack.
33 … Rf8 34 Re4 Ne6 35 gxf4 Rxf4 36 Rxf4 Qxf4+ 37 Rg3 Nc5 38 Qe2 e4 39 dxe4
Nxe4 0-1
4 … Nf6 is clearly playable against the Two Knights Variation, as long as you are
happy with the 3 Nc3 main lines. That applies to both sides.

Game 70
K.Dragun-B.Macieja
US Masters, Greensboro 2017

1 e4 c6 2 Nc3 d5 3 Nf3 Bg4

We now move on to the main line and 3 … Bg4. This is a very solid and harmonious
continuation, which has scored well. Black intends to push his pawn to e6 and weave his
pieces around this rock-like central pawn chain.
4 h3
Trying, and usually succeeding, to obtain the bishop pair.
4 … Bh5!?

But what is this? In fact, 4 … Bh5 is a departure from the normal, solid set up and
offers sharp complications, should White want them. There is nothing to be gained from
robotic development such as 5 Be2, so it is a good idea to go for it.
5 exd5 cxd5 6 Bb5+ Nc6 7 g4 Bg6 8 Ne5 Rc8!
Better than either 8 … Qd6 or 8 … Qb6, as Black does not yet know the correct square
for his queen. Having struck out with 7 g4, White must try to keep the pressure on.
9 d4 e6 10 Qe2 Bb4!
Nice and aggressive in reply. The alternative 10 … Bd6 is also okay.

11 Bxc6+
With 11 h4 White makes a target of the bishop on g6, and might even be able to win
this piece, but falls dangerously behind in development. Interesting play was seen in
D.Sadzikowski-R.Ruck, Poland-Hungary match, Katowice 2016: 11 … Nge7 12 h5 Be4
13 f3 0-0! 14 Nxc6 (14 fxe4? Nxd4 is no good at all for White, whereas 14 Bxc6 Nxc6 15
Nxc6 Rxc6 16 fxe4 Rxc3 17 Kf2 Rc6 18 c3 Be7 may be playable for White, but he has
relinquished any advantage) 14 … Nxc6 15 Be3 Qf6! 16 fxe4 Nxd4 17 Bxd4 Qxd4 (the
positive side to 10 … Bb4 is now revealed and the pressure on the knight on c3 is
considerable) 18 Rd1 Bxc3+ 19 bxc3 Qxc3+ 20 Kf1 dxe4 21 Qxe4 Qxc2 22 Qxc2 Rxc2.
Lest you think this is brilliant play, I have found 14 previous examples of this position
in master play and Black has failed to win a single one! I think the position is about equal.
Three pawns for a piece is just about enough here. White’s rook on d1 can become active,
but his king is poorly placed. These factors balance things up.
23 a4 (23 Rd7 Rxa2 24 Rxb7 Rd8 25 Be2 h6 26 Rh3 is the way that White can to try to
play for a win – good luck!) 23 … a6 24 Be2 Ra2 25 Rh3 Rxa4 26 Rc3 h6 27 Rd7 b5 28
Rcc7 a5 29 Kg2 Rb4 30 Kf3 a4 31 Ra7 Rb3+ 32 Bd3 a3 33 Ke4 ½-½. Rather an odd
moment for both players to back out. Black has four pawns now, but he is tied up, while
White may have feared 33 … b4 34 Rdb7 Rb2.
11 … bxc6 12 h4 Ne7 13 h5 Be4 14 0-0
It is worth pointing out that White does not win the bishop now, as after 14 f3 Black has
14 … f6!, a counter which is more than sufficient; e.g. 15 fxe4 (or 15 Nd3 Bxd3 16 Qxd3
c5!) 15 … fxe5 16 dxe5 d4 17 Qc4 Qa5.
 14 … Bxc3 15 bxc3

15 … h6
15 … Qa5 is a reasonable alternative; for example, 16 Bd2 h6 17 Rfb1 0-0 18 Rb7 Rfe8
19 Re1 c5 20 f3 cxd4 21 cxd4 Qxa2 22 c3 f6 23 Ng6 Nxg6 24 hxg6 Bxf3 25 Rxg7+ Kxg7
26 Bxh6+ Kxh6 27 Qxa2 Kxg6 28 Qxa7 Rxc3 29 Qd7 Rh8 30 Qxe6 Rh1+ 31 Kf2 Rh2+
32 Kg3 Rg2+ 33 Kf4 Rxg4+ 34 Qxg4+ Bxg4 35 Kxg4 Rd3 36 Re8 Rxd4+ 37 Kf3 Kf5
was J.Srbis-T.Cop, Bjelovar 2017, and Black duly won this endgame, which the tablebases
declare as mate in 37.
16 f3
After 16 Rb1 0-0, Black can defend with 17 Rb7 Rc7 or 17 Ba3 Re8.
16 … Bh7 17 Ba3 0-0 18 Rab1 Rc7
Black is solid as a rock with white pawn weaknesses on the queenside that he might
attack later, except that the game now fizzles out.
19 Rb2 Re8 20 Rfb1 f6 21 Rb8 Rc8 22 R8b7 Rc7 23 Rb8 Rc8 24 R8b7 Rc7 25 Rb8
½-½
The current state of play after 4 … Bh5 is satisfactory for Black and what you have
seen is more or less state of the art. As usual with sharp lines, you have to know the
theory.

Game 71
R.J.Fischer-P.Keres
Candidates Tournament, Bled/Zagreb/Belgrade 1959
1 e4 c6 2 Nc3 d5 3 Nf3 Bg4 4 h3 Bxf3 5 Qxf3 Nf6

Black’s plans are easy to understand and coherent. Having given up the two bishops he
will now develop swiftly and arrange his pawns according to how White reacts.
Sometimes he goes … e7-e6 and then … c6-c5! in the style of the French, but he may
switch altogether and place his pawns on dark squares. At the time our featured game was
played, a young Bobby Fischer liked the bishop pair and believed he could make
something of this early advantage.
6 d3
6 e5 allows Black a straightforward game after 6 … Nfd7 7 Qg3 (or 7 d4 e6 8 Be2 c5)
7 … e6 8 f4 c5.
6 … e6 7 g3 Bb4 8 Bd2 d4
8 … Qb6, preparing … d5-d4, is also a very good idea; e.g. 9 0-0-0 d4 10 Ne2 Bxd2+
11 Rxd2 Nbd7 with equality.
9 Nb1 Qb6
9 … Bxd2+ 10 Nxd2 e5 11 Bg2 c5 12 0-0 Nc6 13 Qe2 Qe7 is comfortable for Black,
R.J.Fischer-T.Petrosian, Candidates Tournament, Bled/Zagreb/Belgrade 1959.
10 b3 a5 11 a3 Be7
 Keres does not exchange, as he wants to leave Fischer with the problem of getting his
knight on b1 into the game.
12 Bg2 a4 13 b4 Nbd7 14 0-0
White’s hope of an advantage depends exclusively on whether he can arrange f2-f4 at a
good moment and then press on the kingside. Apart from that, his position is nothing to
write home about. In particular, the queenside pieces make a very poor impression.
14 … c5 15 Ra2 0-0 16 bxc5 Bxc5 17 Qe2 e5 18 f4 Rfc8 19 h4 Rc6!
Black counters nicely – the rook on c6 is useful both offensively, in order to attack c2,
and defensively, along the third rank. White’s bishop pair is unimpressive.
20 Bh3 Qc7 21 fxe5 Nxe5 22 Bf4 Bd6 23 h5 Ra5!
 Keres deploys his rooks excellently. This time the rook can come to c5 and attack or
defend on the fourth rank. White has no plan.
24 h6 Ng6 25 Qf3 Rh5! 26 Bg4 Nxf4 27 Bxh5
27 gxf4 Nxg4 28 Qxg4 Rxh6 leaves White’s position wrecked.
27 … N4xh5 28 hxg7 Qd7 29 Kg2 Ng4 30 Nd2 Ne3+ 0-1
Crushing wins such as this did nothing for the reputation of the Two Knights Variation,
both at the time and for years afterwards.

Game 72
E.Bacrot-R.Ponomariov
Biel 2017

1 e4 c6 2 Nf3 d5 3 Nc3 Bg4 4 d4!?

 As far as I recall, this will lead to a variation recommended by Paul Keres. White seeks
a more open position for his bishop pair.
4 … Nf6 5 h3 Bxf3 6 Qxf3 dxe4 7 Qe3
White must protect d4 before he captures on e4. This slight delay gives Black the time
he needs to get to his usual, solid position.
7 … Nbd7 8 Nxe4 Nxe4 9 Qxe4
 9 … g6!?
An attempt to play the line a little differently from the norm. Black seeks an active
location for his dark-squared bishop.
I slightly prefer to put the bishop on e7 or d6, but this is a question of taste. After 9 …
Nf6 10 Qe3 e6 11 c3 Qd5! Black is very solid indeed. Although White has two bishops,
his light-squared bishop is being kept at bay by the black pawn chain. Meanwhile, Black is
active and prepares to castle long. S.Maze-D.Andreikin, Stockholm 2016, continued 12
Qg3 h5!? 13 Be2 h4 14 Qf3 Bd6 15 0-0 0-0-0 16 Qxd5 (more or less an admission that
there is nothing in the position; he could maybe try 16 a4, just probing, but after 16 …
Bc7 17 Qd3 Qd6 18 f4 Nd5 White has been forced to loosen his position with f2-f4 and
now stands worse) 16 … exd5 17 Bd3 Rde8 18 Be3 Nh5 19 Rae1 Nf4 20 Bxf4 Bxf4 (this
is uninspiring stuff; the game hurtles towards a draw) 21 Rxe8+ Rxe8 22 Rd1 Kc7 23 Kf1
Kd6 24 Re1 Rxe1+ 25 Kxe1 Bc1 26 b3 b5 27 Kd1 Bf4 28 b4 g6 29 Ke2 Kc7 30 Bc2 f5 31
a4 a6 ½-½. Uninterrupted boredom you might say and you might be right, but an effortless
draw with Black at this very high level is a satisfactory result.
10 Be2 Bg7 11 0-0 0-0 12 Rd1 Re8 13 c3!
A wall is erected in front of the bishop on g7, but White remains only very slightly
better.
13 … Qc7
I’m not sure why he doesn’t justify his whole opening plan by now playing 13 … e5!,
which seems to equalize immediately: 14 dxe5 (or 14 d5 Nf6 15 Qb4 cxd5 16 Qxb7 Re6!)
14 … Rxe5 15 Qc2 Qe7 is quite equal.
14 d5! c5 15 Bf4 Be5 16 Be3 Rac8 17 a4 Red8 18 a5 Nf6 19 Qf3 Kg7 20 a6 Rb8
 20 … b6 21 Bb5 gives White complete control of the light squares.
21 axb7 Rxb7 22 Ra2 Rdb8 23 Bc1

White is maintaining a small advantage. He has the bishop pair, pressure down the a-file,
and a superior pawn structure. Can he make anything of these trumps?
23 … Ne8 24 Qe3 Bf6 25 Ra6 Qe5 26 Qxe5 Bxe5 27 Ra5 Rxb2!
This type of move separates the top grandmasters from the levels below. In a tough spot
Black seizes his chance for counterplay, sacrificing the exchange to reduce material and
erect a blockade that is difficult to break.
28 Bxb2 Rxb2 29 Bd3 Bxc3 30 Rxc5 Bb4 31 Rc2 Rxc2 32 Bxc2 a5
The opposite-coloured bishops make it easy enough for Black to hold. If White wants
to win he must find a way through on the kingside.
33 f4 h5 34 Kf2 Kf6 35 Kf3 Nd6 36 Ba4 Nf5 37 Bd7 Nh4+ 38 Kf2 Bc5+ 39 Kf1 Nf5
40 Ke2 Bd6 41 Kf3
41 g4 seems to be the best attempt to make progress: 41 … hxg4 42 hxg4 Nh4 43 g5+
Kg7 44 Rd4 Ng2 45 Kf3 Ne1+ 46 Ke4 Ng2 47 f5 and more pawns are being exchanged.
Can White win this? Bacrot prefers the evening off.
41 … Nh4+ 42 Kf2 Bc5+ 43 Kf1 Nf5 44 Ke2 Bd6 ½-½
The line with 7 Qe3 is obviously very sound, but Black doesn’t have to fear the bishop
pair in the positions that arise.

Game 73
A.Firouzja-A.Dreev
Guilan 2016

1 e4 c6 2 Nc3 d5 3 Nf3 Bg4 4 h3 Bxf3 5 Qxf3 e6 6 d3

This is the usual way to play the Two Knights Variation, slowing things down, hoping
gradually to unfold the position for the bishops. Yet the whole process is very slow and
gives Black plenty of time to organize counterplay.

6 … Nd7 7 Bd2
Many other moves have been tried here, including 7 g3, 7 a3, 7 g4, 7 Qg3 and 7 Bd2.
As played, White prepares to castle long and has the hope of commencing a kingside
pawn storm. Dreev’s reaction is excellent. He delays castling until the time is right.
7 … Qb6 8 0-0-0 Bd6 9 g4 Ne7 10 h4 d4
10 … 0-0?! 11 h5 Be5 12 Qg2 Nc5 13 h6 g6 14 f4 gives White the type of position he
wants.
11 Ne2 c5 12 Qh3 0-0-0! 13 f4 Kb8 14 e5
White is trying to get his light-squared bishop going and thus opens the long diagonal,
getting ready for Bg2. He comes to regret conceding the d5-square.
14 … Bc7 15 Ng3 Qa6 16 Kb1 Ba5!
Once the dark-squared bishops are off, the c3-square becomes very tender.
17 Bc1
After 17 Ne4 Bxd2 18 Rxd2 Nd5, White cannot occupy d6 due to 19 Nd6? Nc3+! or 19
Rf2 f6 20 exf6 gxf6.
 17 … Nb6 18 Bg2 Na4 19 Ne2 Nd5
Converging on c3.
20 Bxd5 Rxd5 21 Qf3 Rc8 22 Qe4 c4
With very consistent development, Black has obtained the upper hand. White’s
complete lack of meaningful play on the kingside is noticeable.
23 dxc4 Nc5! 24 Qg2
24 Qxh7 Qxc4 25 Ng3 d3 leaves White lost.
24 … Qxc4 25 Ng3 d3 26 cxd3 Nxd3 27 Ne4? Bd2!
A nice move to power home the attack
28 Nxd2
28 Qxd2 Qxe4 29 Ka1 Nb4 sets up threats which cannot be parried.
28 … Qc2+ 29 Ka1 Rb5! 30 b3 Ra5 31 a4 Nb4 0-1

 Chapter Eight
The King’s Indian Attack

1 e4 c6 2 d3 d5 3 Nd2

At first sight White’s formation looks passive, but the King’s Indian Attack is used by
players who want to get a solid position out of the opening and outplay the opponent
thereafter. Many fine games have been won using precisely this method. The issue for
White is that it works both ways and, because the game is slow, Black has plenty of
opportunity to develop soundly. No immediate pressure is being put on Black at all.

Game 74
G.Sax-A.Martin
Hastings 1983/84

Hastings 1983/84 was my first grandmaster all-play-all and I approached the tournament
resolving to do my best and to play solidly. The Caro seemed a perfect choice, especially
against an attacking genius like Gyula Sax.
1 e4 c6 2 d3 d5 3 Nd2 e5
This is the most straightforward reply, taking the centre.
 4 Ngf3 Bd6 5 g3 Nf6
Black sometimes plays his knight to e7, keeping the option of … f7-f5 available.
6 Bg2 0-0 7 h3!

A move which came as an unwelcome surprise, and I now started wondering what I
was going to do with the bishop on c8.
7 … Re8 8 0-0 a5 9 Re1 Na6
Black seems to be making good moves, but Sax was ready to surprise me once again.
10 d4!
An excellent move, and I now began to understand that my bishop on d6 and the knight
on f6 were less than ideally placed. When you play at the highest level, and Sax was one
of the best players around at that time, you have to concentrate on detail, because they are
on to any imprecision in a flash.

10 … dxe4?
10 … Nxe4 was better, with some interesting lines which took me some time to work
out and then to reject.
1) 11 Nxe4! dxe4 12 Nxe5! (12 Rxe4 Bf5 is equal) 12 … Bxe5 13 dxe5. This was the
position I didn’t like, but White gets a similar, and superior, position in the game.
2) 11 Nxe5 Nxd2 12 Bxd2 Bf5 is okay for Black.
3) 11 dxe5 is met by 11 … Nxd2 12 Bxd2 (or 12 exd6 Nxf3+ 13 Bxf3 Rxe1+ 14 Qxe1
Bxh3) 12 … Bb4! 13 c3 Bf8 and Black is okay.
11 Nxe5! Bxe5 12 dxe5 Rxe5 13 Nxe4 Bf5! 14 Nxf6+! Qxf6 15 Rxe5 Qxe5 16 c3
Sax is clearly better. The potential of his dark-squared bishop is considerable.
16 … Re8
Others do not equalize either; e.g. 16 … Nc7 17 Bf4, or 16 … Qe6 17 Kh2 Nc5 18 Be3
Ne4 19 Qb3; and finally, 16 … h6 17 Be3 Qe6 18 Kh2 Nc7 19 Qb3 also leaves White
better.
17 Be3 Qe6 18 Kh2 Nc7! 19 Qb3 Nd5 20 Re1 Qd7 21 Rd1!




Black is being outplayed. Note White’s safe king, a feature of the King’s Indian Attack.
The only thing I had going for me at this point was the state of White’s clock, as Sax was
using up a lot of time finding all these accurate moves.
21 … Qc8 22 Bxd5 cxd5 23 Qxd5
White wins a pawn, but there might be some hope now that the light squares around the
white king have been weakened.
23 … a4
23 … Bxh3 24 Qxa5 is better for White.
24 Rd4! Bxh3 25 Rxa4 Bf1 26 Rh4 Rd8 27 Qe5 Re8
 28 Qh5??
I guess when you expect to win, you can become a bit casual. Sax allows a tactical
opportunity, which I was lucky to obtain and to take. 28 Qc5! would still leave White a
pawn up.
28 … Qc6!!
The opposite-coloured bishops make this possible.
29 Qxh7+
Or 29 f3 Rxe3, winning.
29 … Kf8 30 Bc5+
After 30 Qh8+ the black king runs to safety: 30 … Ke7 31 Bg5+ Kd7 32 Rd4+ Kc8 33
Rd8+ Kc7!, winning.
30 … Qxc5 31 Kg1
Or 31 Qh8+ Ke7 32 Re4+ Kd6!.
31 … Re1 32 Qh8+ Ke7 33 Rd4 Bh3+ 0-1
34 Kh2 Rh1+! 35 Kxh1 Qc6+ wins.

A fortunate win. You can see White’s game plan from the opening. He sets up a solid
position with a safe king and comes forward slowly, keeping pieces on the board and
maintaining tension. For me this is a strategy which can be used by stronger players. It is
quite useless for a lower-ranked player to try this sort of thing out against somebody better
than them. You will just get outplayed.
Game 75
L.Stein-V.Hort
Interzonal Playoff, Los Angeles 1968

Stein had several games with Vlastimil Hort in this variation and, to my knowledge, this
was the only one he won.
1 e4 c6 2 d3 d5 3 Nd2 g6 4 g3 Bg7 5 Bg2 e5 6 Ngf3 Ne7 7 0-0 0-0 8 c3 Nd7

9 b4!
Here we see Stein’s patent move. There is nothing earth-shattering about 9 b4 but,
without it, White would have very little to shout about.
A harmless alternative is 9 Qc2 b6 10 Re1 Bb7 11 d4 exd4 12 Nxd4 c5 13 Nb5 Ne5 14
f4 a6 15 fxe5 axb5 16 exd5 Bxd5 17 Ne4 Bxe5 18 Nxc5 f6 ½-½, S.Novikov-J.Hawkins,
Vienna 2013. Clearly, a result which suited both players.
9 … b6 10 Bb2 Bb7 11 Re1 Re8 12 Bh3 Qc7 13 exd5
With both players still developing, Stein is the first to break the tension.
13 … cxd5 14 c4 d4 15 Rc1 f5
This is the reason they put the knight on e7 in the first place, and we are now set for a
heavyweight struggle, typical of the KIA. The stronger player will win this complicated
middlegame. We have a Benoni structure here, with White owning a queenside pawn
majority and Black a central pawn majority. Who can make the most of his trump?
 16 Bg2 Bf6 17 c5 b5!

17 … bxc5 may have been playable tactically, but it must help White to open the
queenside, and so Hort prefers to block.
18 Nb3 Bd5 19 a4 a6 20 Na5 Nc6 21 axb5 axb5 22 Nxe5!?
22 Nxc6 Qxc6 23 Qe2 g5 24 h3 h5!? was possibly a bit passive-looking for White, who
uncorks a tactic. Stein was able to pull a rabbit out of his hat at short notice and, while
Hort is okay here, 22 Nxe5 may have come as a surprise. Black must deal with the
unprotected bishop on d5.
22 … Nxb4?
22 … Rxe5 23 Rxe5 Bxg2 was correct, with Black in good shape after 24 Qb3+ (or 24
Re6 Bd5) 24 … Kg7 25 Nxc6 Bxc6.
23 Nxd7 Qxd7 24 c6!



24 … Qf7
24 … Rxe1+ leads to a complicated variation: 25 Qxe1 Qe7 26 Qxe7 Bxe7 27 c7 Rc8
28 Nc6 Bxc6 29 Bxc6 Rxc7 30 Bxb5 Rxc1+ 31 Bxc1. White is a little better here, but a
draw looks likely.
25 Rxe8+ Rxe8 26 Nb7 Be5 27 Bxd5 Qxd5 28 Rc5 Qf7
Not 28 … Qa2? 29 Rxe5! Rxe5 30 c7 Qe6 31 Nd6!! Re1+ 32 Qxe1 Qxe1+ 33 Kg2.
29 Qf3 g5 30 c7 g4 31 Qd1 Bxc7 32 Qd2 Bb6??
He had to play 32 … Be5 to protect his king; e.g. 33 Qxb4 Qxb7 34 Rxb5 Qc7.
33 Rxb5 Nxd3 34 Nd6 Qd7 35 Qg5+
After some head-spinning preliminary tactics, Stein takes advantage of the open black
king.
35 … Kh8 36 Qf6+ Kg8 37 Qg5+ Kh8 38 Nxe8 Qxe8 39 Qxf5 Ne5 40 Rxb6 Nf3+ 41
Kg2 Qa8 42 Qf6+ Kg8 43 Qe6+ Kh8 44 Qc6 1-0

OBSERVATION: Quiet opening lines often lead to very complicated middlegames.

Game 76
L.Stein-G.Barcza
Caracas 1970
Some of the best King’s Indian Attack games were played by Leonid Stein. The players of today stand on his shoulders.

1 Nf3 d5 2 g3 c6 3 Bg2 g6 4 d3 Bg7 5 Nbd2 e5 6 e4

We have conveniently transposed into the Caro-Kann.

6 … Nh6!?
Black has plenty of freedom to develop as he pleases. The idea seems to be … f7-f6
and then … Nf7. Possibly … f7-f5 can be played at once.
7 0-0 0-0 8 b4!
Stein liked b2-b4 in this set-up. White gains space and makes room for his bishop to
come to b2. Queenside and central pressure may follow. Meanwhile, Barcza gets on with
his plan.
8 … f6!? 9 Nb3 dxe4?!
Exchanging queens is no solution to this position and just hands White a slight
advantage on a plate. Black must be combative and play something like 9 … Na6 10 c3
Nf7, when an unclear middlegame is going to result.
10 dxe4 Qxd1 11 Rxd1 b6 12 Bf1 Nf7 13 Be3


White has a small, risk-free advantage, based on the greater activity of his pieces. It’s
not much, but hey, this is the King’s Indian Attack.
13 … Bg4 14 Kg2 Nd7 15 h3 Be6
15 … Bxf3+ 16 Kxf3 Rad8 17 Kg2 may or may not be okay for Black, but it will
certainly involve growing pressure from White and a lot of suffering.


16 Nbd4!!
A quite shocking move and one which I am sure Barcza missed. It is worth
remembering that Barcza was a very strong and solid grandmaster. In what looks like a
quiet position this comes as a bolt from the blue.
16 … exd4 17 Nxd4
Black’s pieces on e6 and d7 are loose.
17 … f5?
In shock, Black plays a poor move. Instead, 17 … Bxh3+! is the best chance; e.g. 18
Kxh3 Ng5+ 19 Kg2 Ne5 (or 19 … Nxe4 20 Nxc6) 20 f4 Ng4 21 Bg1 Nxe4 22 Nxc6 f5 23
Rac1 Rf6 24 b5. This isn’t great for Black, but it is better than the game.
18 Nxe6 Bxa1 19 Nxf8 Nxf8 20 Rxa1 fxe4 21 a4



The smoke clears with White in charge of two powerful bishops.
21 … Ne5 22 Bd4 Nfd7 23 Re1 c5 24 Bxe5 Nxe5 25 Rxe4
Black jettisons a pawn to try to relieve the pressure, but it is no good.
25 … Nd7 26 Re7 Nf6 27 Bc4+ Kh8 28 bxc5 Rc8 29 cxb6! axb6
29 … Rxc4 30 Rxa7 Rb4 31 a5 Nd5 32 b7 wins.
30 Bb3 Rc6 31 Re6 Rxe6 32 Bxe6 Kg7 33 Kf3 Ne8 34 Ke4 Kf6 35 Bc4 Nc7 36 f4 h6
37 g4 g5 38 f5 Ke7 39 Ke5 Ne8 40 Kd5 1-0
Can we accuse Barcza of underestimating Stein in this game? Probably. His play was
too simplistic and he was caught napping by 16 Nbd4. Black must take the KIA very
seriously.

You may have noticed that, while the variation is called the King’s Indian Attack, there
doesn’t seem to be a great deal of attacking going on by White. So I couldn’t resist
showing this recent encounter, where the boot is finally put in on the black king. Please
bear in mind that this is a blitz game, albeit between two strong grandmasters.

Game 77
Z.Andriasian-A.Motylev
European Blitz Championship, Tallinn 2016
1 e4 c6 2 d3 d5 3 Nd2 e5 4 Ngf3 Bd6 5 g3 Nf6 6 Bg2 0-0 7 0-0 Re8 8 Re1 Nbd7 9 b3!?

A move is a move and Andriasian had previous history with 9 b3. To me, it looks like
nothing special.
9 … a5 10 a3 d4!?
I guess I would prefer to block once White has put his bishop on b2. The game
Z.Andriasian-D.Navara, European Rapid Championship, Minsk 2015, saw 10 … Nc5!,
which looks good and White got nothing: 11 Bb2 dxe4 (or 11 … d4) 12 dxe4 b5 13 h3
Qc7 14 Nh4 g6 15 Qe2 Ba6 16 Kh2.
The influence of Carlsen is felt all over modern chess. How many times has he almost
bypassed the opening altogether by playing in a solid, rather vague style, keeping some
tension and then completely outplaying the opposition from move 25 onwards? And isn’t
this exactly what White is trying to do here; just shuffling around waiting for something to
happen? The problem is that not everyone has Carlsen’s talent.
The game continued 16 … Ne6 17 Nhf3?! c5 18 c3 c4 19 b4 Bb7 20 Red1 Bc6 21 Ne1
Nd7 22 Nc2 Nb6 23 Ne3 Na4 24 Ndf1 (White is still shifting wood in a rather lazy
manner and he is now in for a shock) 24 … Nf4! 25 gxf4 exf4 26 Qd2 Rad8 27 Ng4 f3+
28 Kg1 Be7! 29 Qc2 fxg2 30 Ng3 Bh4 31 Bc1 h5 32 Nh6+ Kh7 33 Rxd8 Rxd8 34 Nxh5
gxh5 35 e5+ Rd3 36 Bf4 Be4 37 Re1 Bg6 38 bxa5 Rf3 0-1. Under no pressure throughout,
Black was allowed to show his skills and did so in style.
11 Rf1
 So that when f2-f4 comes, the rook is in the right place.
11 … a4 12 b4 b5 13 Nh4 c5 14 Nf5 Bf8 15 bxc5 Bxc5 16 Nf3 Qc7 17 N3h4 b4 18 f4

At last, White does something to justify the name of the opening.

18 … bxa3 19 g4 Nf8 20 g5 N6d7 21 Nxg7!
Even though this may not be the best move, it is quite enough for a blitz game and is
thematic in a position of this type. White is getting hammered on the queenside and so
seeks an attack to compensate.
21 … Kxg7 22 f5 Ra6 23 Qh5 Kg8 24 Rf3
Computers can defend these positions, but with limited time humans struggle.
24 … Bb4 25 f6 Ng6 26 Nf5 Ndf8??
26 … Nxf6 27 gxf6 Rxf6 is the correct defence and apparently there is nothing for
White after 28 Nh6+ Kg7 29 Rxf6 Kxf6. This may be right, but I know I wouldn’t have
gone for it as Black, especially with no time.
27 Qh6 Ne6 28 Rh3 Ngf8 29 Ng7 Qc3 30 Nxe6 1-0
I feel better now. I have seen a White attack.

We now come to a line for Black which was recommended by grandmaster Ludek
Pachman in his famous Semi-Open Games book. That was a long time ago and the
suggestion has been refuted, but it is definitely worth consideration.

Game 78
P.Schalkwijk-K.Melia
Correspondence 2000

1 e4 c6 2 d3 d5 3 Nd2 dxe4 4 dxe4 e5 5 Ngf3 Bc5?!

If this works, then Black has a free and easy game. Unfortunately, White can take on
e5.
Instead, 5 … Nd7 6 Bc4 Bd6 is respectable. This type of position resembles an Italian
Game more than a Caro-Kann and it is a situation where the move … c7-c6 is
uncommonly useful. Black is already equal: 7 Bb3 Ngf6 8 Nc4 Bc7! 9 0-0 (or 9 Nd6+
Bxd6 10 Qxd6 Nxe4 11 Qb4 Nec5 12 Be3 Nxb3 13 axb3 a5 14 Qd6 Qe7) 9 … 0-0 10 Bg5
a5 11 a4 Qe7 12 Qe2 Nc5 13 Rad1 Re8 was equal in Y.Balashov-A.Polschikov, Taganrog
2011. Here we see yet another good way for Black against the placid opening move order.
6 Nxe5!
Very strong, as long as White knows what to do over the next few moves.
6 … Bxf2+
6 … Qd4 7 Nd3 Bb6 8 c3 does not provide any sort of compensation for Black.
7 Kxf2 Qd4+ 8 Ke1 Qxe5 9 Nc4!

This is the move that refutes the variation. Black is forced into a defensive mess, with
no control over the dark squares.
9 … Qxe4+ 10 Be2! Qe6
10 … Qh4+ 11 g3 Qd8 is a possible defence, although White is still much better after
12 Qxd8+ Kxd8 13 Nd6.
11 Nd6+ Kf8 12 Rf1 f6 13 Nxc8
13 Kf2 Ne7 14 Bc4 Nd5 15 Bf4! also looks very strong.
13 … Qxc8 14 Qd6+ Ke8
14 … Ne7 15 Bh6! is a knockout punch.
15 Qg3 g6 16 Bf4 Ne7 17 Rd1
 The white pieces have flowed out nicely into the attack and there is no defence.
Remember that Black cannot castle!
17 … Nf5
17 … Na6 18 Rd6 Qf5 19 Bxa6 bxa6 20 Qe3 wins for White.
18 Qc3 Nd7 19 Rf2 Kf7 20 Bc4+ Kg7 21 Be6
21 g4 Re8+ 22 Kf1 seems equally devastating.
21 … Re8 22 Re2 Rxe6 23 Rxe6 Ne5 24 Qb3
He might also have played 24 Rxf6! Kxf6 25 Bxe5+! (the queen is needed for light-
squared checks) 25 … Kf7 26 g4! Ne7 27 Qf3+ Kg8 28 Qf6, winning.
24 … Qc7 25 g4 Nh4 26 Bxe5 fxe5 27 Red6 1-0
A nice lightweight game by Schalkwijk.

To close this chapter, here is the most modern game that I can find. White departs from
the normal fianchetto set-up, preferring to put his bishop on e2. What shall we call this, a
reversed Philidor? The impression is the same as with g2-g3. White has little hope of
getting an opening advantage.

Game 79
D.Tsoi-V.Burmakin
Chigorin Memorial, St. Petersburg 2017

1 e4 c6 2 d3 d5 3 Nd2 g6
 Why not?
4 Ngf3 Bg7 5 Be2!? Nd7 6 0-0 Nh6 7 c3 0-0 8 Qc2 e5 9 d4

An early clarification in the centre, which cannot come to anything.
9 … exd4 10 Nxd4 Re8 11 Re1 Nf6 12 exd5 Nxd5 13 N2f3 Nf5
The position is already completely level. I really cannot grasp how White is thinking,
since he is aiming for a position where the stronger player is going to win. Has he looked
at the ratings recently?
14 Rd1 Nxd4 15 Nxd4 Qe7 16 Bf1 Bg4 17 f3 Bd7 18 Qf2 Rad8
Black is now slightly better. The move f2-f3 looks out of place.
19 Bd2 c5 20 Re1 Qd6 21 Nb3 b6 22 Bg5 Bf6 23 Bxf6 Qxf6 24 Nd2 Bf5 25 Rxe8+
Rxe8 26 Re1 Rd8 27 Ne4 Bxe4 28 Rxe4 Ne7 29 Qe3 Nf5 30 Re8+


White has handled the middlegame well and restricted any disadvantage. He really
should draw this position, but the grandmaster is playing on.
30 … Kg7 31 Qe1 Rxe8 32 Qxe8 Qd6 33 Qe2 h5 34 a3 h4 35 b4?!
However unpleasant it may seem, White should sit tight with a move like 35 Qe4.
35 … cxb4! 36 axb4
36 cxb4 weakens the central dark squares.
36 … a5! 37 bxa5 bxa5 38 c4
38 Qc2 a4! 39 Qxa4 Qb6+ sees the Black attack gather speed; e.g. 40 Kh1 Qf2 41 Qb5
Ne3, winning.
38 … a4 39 Qb2+ Kh7 40 Qc3 Qc5+ 0-1
Since … Qf2 followed by … Ne3 will be completely crushing.
We conclude this chapter by observing that the KIA is still perfectly playable for White,
but do not expect good results if you venture it against a higher-ranked opponent. You will
feel that you are banging your head against a brick wall.

 Chapter Nine
Odds and Ends

1 e4 c6

We close the book by looking at some interesting ideas, not really mainstream, but
which Black ignores at his peril. Look out for some sharp chess in this chapter!
By far the most important and popular line herein starts with the unlikely 3 f3, the so-
called ‘Fantasy Variation’. Although White reinforces the centre, the impression is that he
is weakening his kingside. The popularity of the variation is due to the number of quick
White wins in the main line when Black does not know what he is doing. So that’s where
we will start.

Game 80
P.Hair-S.Yee
Asian Seniors Championship, Auckland 2017

1 e4 c6 2 d4 d5 3 f3 dxe4 4 fxe4 e5!



This is a pretty good line for Black, who threatens … Qh4+ but, above all, he has to
show some common sense. Once he has captured on e4 the f-file opens up and f7 becomes
a focal point for attack. You could argue that this position resembles some sort of King’s
Gambit, with the unlikely move … c7-c6 thrown in.
5 Nf3 exd4
To me, 5 … Bg4 or 5 … Be6 would be much less risky, for which see the next game.
6 Bc4!

6 … Bb4+?
Stanley Yee plays the type of move that White is fantasizing about. To encourage White
to play c2-c3, in turn encouraging the queen to come to b3, is folly.
6 … Be6 may be playable, but not the way Black interprets it in I.Solomunovic-
N.Tosic, Serbian Team Championship 2017: 7 Bxe6 fxe6 8 0-0 e5?? (8 … Nf6 had to be
played) 9 Nxe5 Nf6 10 Bg5 Be7 (10 … Nbd7 was a bit better, intending 11 Nxd7 Qxd7 12
Bxf6 gxf6 13 Rxf6 0-0-0, though after 11 Nc4! b5 12 e5 White still has a vigorous attack)
11 Bxf6 Bxf6 12 Qh5+ g6 13 Nxg6 hxg6 14 Qxg6+ Kd7 15 Rxf6 (White wins right out of
the opening; unsurprisingly, the possibility of doing so appeals to many players) 15 …
Kc7 16 Nd2 Rg8 17 Rf7+ Kb6 18 Qf5 a5 19 Nc4+ Ka6 20 a4 Rg5 21 Qf4 Rc5 22 b3 d3
23 Qe3 b6 24 cxd3 Ra7 25 Rxa7+ Kxa7 26 b4 Qh8 27 e5 axb4 28 a5 Nd7 29 axb6+ Kb8
30 Qg3 Nxe5 31 Re1 b3 32 Nxe5 1-0.
It was hard to believe some of the moves in this game, 8 … e5 in particular. But it is
this level of game that is played so often at club level, where you can almost expect a
howler, and that is why 3 f3 continues to be played.
7 c3 Be7
After 7 … dxc3, the reply 8 Bxf7+! seems crushing; e.g. 8 … Ke7 (or 8 … Kxf7 9
Qxd8 cxb2+ 10 Kd1 bxa1Q 11 Ng5+ Kg6 12 Qe8+ Kh6 13 Ne6+ Bd2 14 Bxd2+ g5 15
Bxg5 mate) 9 Qb3!! cxb2+ 10 Qxb4+ Kxf7 11 Bxb2 Nf6 12 Qc4+ Ke8 13 0-0, winning.
8 Qb3 Be6 9 Bxe6 fxe6 10 Nxd4 e5 11 Ne6


There is no good advice to offer Black anymore.
11 … Qd7 12 Nxg7+
12 0-0 is stronger still; e.g. 12 … Nf6 13 Nxg7+ Kd8 14 Rd1, while 12 … Bf6 is met
by 13 Na3! with the idea of Rd1 and Nc4.
12 … Kd8 13 Ne6+ Kc8 14 0-0 Na6 15 Bg5 Nc5 16 Nxc5 Bxc5+ 17 Kh1 Ne7 18 Nd2
Ng6 19 Rf7 Be7 20 Nf3 Re8 21 Rd1
Simple and effective attacking play from Philip Hair.
21 … Qc7 22 Qe6+ Kb8 23 Rd7 Qc8 24 Bxe7 Nxe7 25 Qd6+ 1-0

Needless to say, Black can do much better than the previous game. Let me help you
with a sensible recommendation.

Game 81
G.Popilski-H.Nakamura
Millionaire Open, Las Vegas 2015

1 e4 c6 2 d4 d5 3 f3 dxe4 4 fxe4 e5 5 Nf3 Bg4!




There are two benefits to 5 … Bg4 over 5 … Be6. First, it’s more aggressive; and
second, the bishop on e6 can become a target, whereas on g4 it is less exposed.
Nevertheless, 5 … Be6 is also fine, preventing Bc4 and setting the small trap 6 Nxe5
Qh4+. White usually plays 6 c3 and then 6 … Nd7 7 Bd3 Bd6 8 0-0 Ngf6.
6 c3
You would think White would take the opportunity to play 6 Bc4, but there are no real
difficulties for Black after that; e.g. 6 … Nd7! 7 0-0 (if 7 dxe5 Bxf3 8 Qxf3 Nxe5 9 Qe2
Nxc4 10 Qxc4 Qb6, White is worse already) 7 … Ngf6 8 c3 Bh5 9 Bg5 Bd6 10 Nbd2 0-0
11 Qe1 Qc7 (by strongpointing e5, Black easily maintains the balance) 12 Qh4 Bg6 13
Kh1 Rae8 with an equal game.
6 … Nd7 7 Bd3 Bd6 8 0-0 Ngf6
Note Nakamura’s very tight control over the e5-square. This alone makes it tough for
White to get an attack going. Later, Black hopes to exploit White’s looser pawn structure.
9 a4
9 h3 Bh5 10 Nbd2 0-0 11 Nc4 is a bit more testing, but after 11 … Bc7 12 Bg5 exd4 13
cxd4 (or 13 e5 Bxf3 14 Qxf3 Bxe5) 13 … h6 14 Bh4 Qe8! 15 Qb3 Rb8, Black has the
usual rock-solid position.
9 … 0-0 10 Nbd2 Re8 11 Nc4 Bc7 12 Qc2 exd4 13 cxd4 Bxf3 14 Rxf3 Nb6
Nakamura opens fire at the white central pawns. Are they denying Black squares or are
they merely a target?

15 e5 Nxc4 16 Qxc4
16 exf6 is met by 16 … Qxd4+.
16 … b5!
A more tactical move than it looks. At first sight Black is decoying the white queen
away from the defence of d4, but there is more to it.
17 Qc3
17 Bxh7+!? was interesting. I am sure that Nakamura’s intention was 17 … Nxh7! (not
17 … Kxh7? 18 Qxf7 Qxd4+ 19 Be3 Qd7 20 Rh3+ Qxh3 21 gxh3 Bxe5 22 axb5 cxb5 23
Rxa7, which is clearly better for White; strangely, however, 17 … Kh8 is also good for
Black after 18 Qxf7 Re7!) 18 Qxf7+ Kh8 19 Be3, when White has to pause and now
comes 19 … Rf8! 20 Qh5 Rxf3 and the attack is broken.
17 … Bb6 18 Be3 Nd5 19 Qb3 bxa4 20 Rxa4 Rb8 21 Qc4 Qd7
Black has been able to complete his development and we have a situation where the
players have mutual positional weaknesses, which cancel each other out.
22 Bf2 g6 23 Be4 f5 24 Bc2 Kg7 25 Rfa3 a5 26 Kh1 Bc7 27 Qa2 Nb4 28 Qc4 Nd5 29
Qa2 Nb4 30 Qc4 Nd5 ½-½
If you want a decent, easy-to-learn defence to the Fantasy Variation, then 5 … Bg4 and
the subsequent plan of strongpointing e5 fits the bill.

Game 82
L.Henris-T.Gharamian
Charleroi 2017

Not every Caro player is going to like the type of open game that results from 3 … dxe4
and 4 … e5. One of the reasons White plays this system is to try to drag his opponent into
an unfamiliar situation. Fortunately, there are other good alternatives available.
1 e4 c6 2 d4 d5 3 f3 e6!

A move which still threatens … d5xe4 and … Qh4+ but which will usually give the
game a closed and solid quality, much more in the Caro-Kann style.
4 Nc3
4 Be3!? is not played as often but is an alternative that cannot be ignored. White is
actually sacrificing a pawn, but Black would slip behind in development if he captured: 4
… dxe4 5 Nd2 exf3 6 Ngxf3 Nf6 7 Nc4!? and a curious position has been reached, similar
to the Blackmar-Diemer Gambit. Black is cramped, but he has an extra pawn. High-level
results favour White, as Black is a bit passive, but a good defender could easily handle this
situation.
A better reply is 4 … b6!, when R.Felgaer-A.Dreev, FIDE World Championship,
Tripoli 2004, continued 5 Nd2 c5 6 c3 Qd7 7 Nh3 cxd4 8 Bxd4 Ne7 9 Nf4 Nbc6 (Dreev
has turned the position into a weird French) 10 Bb5 a6 11 exd5 axb5 12 dxc6 Qxc6 13
Ne4 Nf5 14 Nh5 Bb7 15 g4 Nxd4? (15 … 0-0-0! 16 gxf5 e5 looks okay) 16 Qxd4 f6 17 0-
0-0 e5 18 Qd3 b4 19 g5! Ba6 20 Nhxf6+? (the wrong knight; instead 20 Nexf6+! gxf6 21
Qf5! would have set Black serious problems) 20 … gxf6 21 Nxf6+ Kf7 22 Qf5 Qe6 23
Qe4 Be7 24 Rd7 bxc3 25 Rhd1 cxb2+ 26 Kxb2 Bb5 27 Rxe7+ Qxe7 28 Qd5+ Kg7 29
Nh5+ Kg6 30 Ng3 Qa3+ 31 Kb1 Qb4+ 32 Ka1 Rxa2+ 0-1.
 4 … Qb6!
Trying to pin the bishop down to c1.
5 a3!?
5 Nge2 Nf6 6 Qd3 Be7 produces a French-like position, quite closed, where Black is
looking to play … c6-c5 as fast as he possibly can.
5 … Nd7
Another good and aggressive approach was seen in T.Stijve-Y.Pelletier, Swiss
Championship, Graechen 2017: 5 … Nf6! 6 Be3 dxe4! 7 fxe4 Ng4 8 Bc1 e5! 9 Nf3 exd4
10 Nxd4 Bd6 11 Be2 h5 12 Bxg4 Bxg4 13 Qd2 (White has played the opening in a most
unimpressive manner) 13 … Nd7 14 Nf5 Bxh2 15 Nxg7+ Kf8 16 Qf2 Qxf2+ 17 Kxf2 Be5
18 Nf5 Re8 19 Bd2 Nf6 20 Rae1 Nxe4+ 21 Nxe4 Bxf5 22 Ng3 Bd4+ 23 Be3 Rxe3 24
Rxe3 Bg4 25 Rhe1 Rh6 26 Nf1 h4 27 Kg1 Rd6 28 Kh2 Bxe3 29 Nxe3 Be6 30 Re2 Kg7
31 Rf2 Kg6 32 Rf4 Kg5 33 Ra4 Rd2 34 Rxa7 Re2 35 Nf1 Bd5 36 Ra5 Rxg2+ 37 Kh3 Rf2
38 c4 b6 39 Ra8 Be6 mate.
6 Bf4 Ne7
The point of 5 a3 is revealed after 6 … Qxb2 7 Na4.
7 Nge2 Ng6 8 Bg3 Be7 9 Bf2 Qxb2

Perhaps White forgot that Black can capture now?

10 g3 Qb6 11 h4 0-0 12 h5 Nh8 13 Bh3 dxe4 14 fxe4 Qa5 15 h6 g6
A rather humorous position has arisen, with the knight on h8 quite unique! Of course,
Gharamian has plans for the knight to re-emerge. The whole impression is that White’s
play has been dodgy from early on and he is certainly much worse, despite the oddly-
placed knight.
16 Be3 e5 17 d5 Bc5 18 Bd2 Qc7 19 g4 Nb6 20 Ng3 f6 21 Qe2 Nf7
Back from the dead, with revenge in mind!
22 Rf1 Ng5 23 Bg2 Qd7 24 dxc6 bxc6 25 Bxg5 fxg5 26 Bf3 Qd4 27 Qd2 Qxd2+ 28
Kxd2 Nc4+ 0-1

3 … e6 is a very reasonable way to play against the Fantasy Variation. The positions
reached often resemble odd highways and byways of the French Defence and are quite
original. Black can feel confident about 3 … e6 as it leads to systems in the closed Caro-
style.

Game 83
A.Timofeev-I.Khenkin
Capablanca Memorial, Havana 2009

1 e4 c6 2 d4 d5 3 f3 g6!?

This is a popular idea, but I think setting up with … g7-g6 plays into White’s hands. It’s
easy now to continue Be3 and Qd2, when f2-f3 comes in useful and the whole white
position starts to come to life in the style of the Sämisch King’s Indian. If you like a really
messy game, then you can of course still play 3 … g6. It’s just a bit more risky than the
other lines we have discussed.
4 Nc3 Bg7 5 Be3 Qb6 6 Qd2! Qxb2 7 Rb1 Qa3
I didn’t know there was a ‘poisoned pawn’ variation of the Caro-Kann until now.
 8 h4
8 Bd3 Qa5 9 Nge2 is another way to get good compensation for the pawn. In the game
S.Mamedyarov-P.Svidler, World Blitz Championship, Almaty 2008, White developed very
quickly and got a strong attack: 9 … e6 10 h4 h5 11 0-0 Nd7 12 Bg5 a6 (Black argues that
he has no real weaknesses to speak of, but he is in for a shock) 13 Nxd5! Qxd2 14 Nc7+
Kf8 15 Bxd2 Rb8 16 Bb4+ Ne7 17 Bd6 (it has to be pointed out that, given the time limit,
no real blame can be attached to Black; online databases have become infested with these
high-level blitz games and, while some of them are useful, the majority are utter tosh; we
are overwhelmed with chess in the internet age – not necessarily a good thing) 17 … b5 18
f4 Rb7 19 f5 exf5 20 exf5 Kg8 21 Bxe7 Rxc7 22 fxg6 fxg6 23 Bxg6 Rh6 24 Bf7+ Kh7 25
Bg5 Nf6 26 Bb3 Re7 27 Nf4 Ng4 28 c3 c5 29 dxc5 Bxc3 30 Rbd1 Re5 31 Rd6 Kg7 32
Nxh5+ 1-0.
8 … dxe4 9 h5 Nd7 10 fxe4 Ngf6 11 h6 Bf8 12 e5 Ng4 13 Bf4 Nb6 14 Be2 Be6 15
Rh4
While the computer will show a Black advantage, we all know that such random
positions are tough to defend. It’s all about co-ordination and Khenkin really struggles to
get his pieces to work together.
15 … f5 16 Nf3 Nc4 17 Qd3 Qa5 18 Ng5
Perhaps 18 Rxb7! was most accurate, not allowing the light-squared bishop on e6 to
come into the game so easily.
18 … Na3 19 Rxb7 Bc4 20 Qd2 0-0-0!
 A startling move and a very good one. White must retreat.
21 Rb2 Bxe2 22 Nxe2 Qxd2+ 23 Bxd2 Bxh6 24 Bf4 Bxg5 25 Bxg5 Rd7 26 Rh3 Nc4
27 Rb4 Nb6 28 c4 h6 29 e6 Rb7 30 Bf4 a5 31 Rb1 Nxc4?!
Black has been managing the situation well, but here he should prefer 31 … a4!, which
leaves the c-file closed.
32 Rc1 g5?
32 … Nd6 33 Rxc6+ Kd8 34 Bxd6 exd6 35 Nf4 Re8 36 Rxd6+ Ke7 is good for White,
but it would have been better than the game for Khenkin.
33 Bxg5 Nd6 34 Rxc6+ Rc7 35 Rxc7+ Kxc7 36 Bxe7 Ne4 37 d5 Rb8 38 Nc3 1-0
Rather than see the white pawns touch down, Black prefers to minimize the pain.

Summarizing, the Fantasy Variation is a practical and dangerous weapon. If I had to
choose my system of defence, I think I would prefer 3 … e6, but we have seen three good
ways and you may have a completely different opinion.

Game 84
L.Trent-C.Storey
British Championship, Llandudno 2017

1 e4 c6 2 Nc3 d5 3 Qf3!?


I am sure that this was played as a shock weapon. Storey does not play the Caro-Kann
very often and I presume Trent just wanted to catch him on the nose with something
unusual and sharp. White develops quickly and so Black will not be able to refute this
move. Instead, he must try to use the position of the white queen to his advantage. 3 Qf3 is
a trappy line and the graveyards are full of Black players who have been mated on f7. It is
not dangerous if Black demonstrates a bit of care.
3 … e6
White’s chances of gaining an advantage after 3 … dxe4 4 Nxe4 Nd7! are minimal and
this is one of the main reasons the line isn’t played more often. White has committed his
queen a bit too early. 4 … Nf6?! is not as good, as 5 Nxf6+ exf6 6 Bc4 Bd6 7 d4 0-0 8
Ne2 gave White a slight advantage in D.Smerdon-F.Zeller, German Bundesliga 2017.
White’s pieces are aggressively placed and he has the superior pawn structure.
After 4 … Nd7 5 d4 (5 Bc4 Ne5! or 5 Ng5 Ngf6), the game S.Reh-I.Khenkin,
Wiesbaden 2013, continued 5 … Ngf6 6 c3 Nxe4 7 Qxe4 Nf6 8 Qf4 Qd5 9 Nf3 Bf5 and
Black has achieved a comfortable Smyslov-type position with equal chances. Khenkin
went on to win in 53 moves.
4 d4 Qb6!?
Taking the fight to White as soon as possible.
5 Bd3!?
5 Nge2 defends d4 but blocks in the bishop on f1. Black equalizes easily after 5 …
Nf6!, when White already has some problems finding a coherent development plan.
 5 … Qxd4
Storey grabs what is on offer. Meanwhile, White streaks ahead in development.
6 Nge2 Qb6 7 Be3 Bc5

8 0-0?!
8 Qg3! Bxe3 9 fxe3 would make more sense of the early queen adventure and White
seems to have a decent initiative here: 9 … Ne7!? (9 … g6? is bad after 10 exd5 exd5 11
Qe5+, or 10 … cxd5 11 Nxd5!, or 10 … Nf6 11 d6) 10 Qxg7 Rg8 11 Qxh7 Nd7 12 0-0
Ne5 13 exd5. One gathers from these lines that the capture on d4 by Storey was pretty
risky.
8 … Bxe3 9 fxe3 Nh6 10 Qg3
White has played his queen to g3 slightly too late.
10 … 0-0! 11 Rf6
An audacious move, but the game is slipping away already, so he has to try something.
There are no real targets in the black camp and so setting up the idea of Rxh6 is the best
that Trent can find. It is worth pointing out that slower moves such as 11 b3 are met by 11
… f5! and Black is material ahead.
11 … dxe4 12 Bxe4 Nf5 13 Bxf5 exf5
The rook on f6 looks a bit weird. White hastens to get it back into the game.
14 Rd1 Qxb2!
 There is absolutely no reason why Black should not take another pawn.
15 Rfd6 Na6 16 Qe5 Qb4 17 Nf4 Qc5!

In the absence of any White threats at all, Storey takes his time and offers a relieving
queen exchange.
18 Ncd5
More desperation. I don’t think there is any good advice to offer White at this point.
18 … cxd5
Black plays like a machine, taking everything on offer.
19 Nh5 f6 20 Qg3
Or 20 Qe7 Qc7.
20 … Qc7 21 R1xd5 Be6
21 … Qe7 also wins.
22 Rd1 Rae8 23 Nf4 Nc5 24 Qe1 Bf7 25 Qc3 b6 26 Qd4 Qe7 0-1
White’s idea is better than it was made to look in this game and if he had played 8 Qg3,
he could have built some pressure.

Most reliable for Black is 3 … dxe4 4 Nxe4 Nd7, after which White struggles to get
anywhere. This does not detract from the originality of White’s set-up in any way, and so
we can expect to see 3 Qf3 cropping up from time to time when a surprise is needed.

Game 85
S.Mariotti-J.Magem Badals
Andorra Zonal 1987

1 e4 c6 2 d4 d5 3 exd5 cxd5 4 c4 Nf6 5 c5?!

The Gunderam Variation, not quite a Panov Attack. 5 c5 is a very sharp move, which
intends to establish a queenside pawn majority without delay. It is a rare bird at master
level because White is neglecting piece development. Sluggish play by Black can often
leave White well placed after moves such as b2-b4, Bb5, Nf3-e5, a4 etc, and so an
energetic reaction is called for.
5 … Nc6 6 Bb5 e5!
Black has plenty of good replies to the Gunderam Variation and this is one of the best.
White’s centre is immediately placed under attack before he can get his act together.
7 Nc3
7 dxe5 Ne4 leaves White in a mess.
7 … exd4 8 Qxd4 Be7 9 Nge2 0-0
 What has Magem Badals done apart from play healthy developing moves? If White
moves his queen he cedes ground, so he has to give up his b5-bishop, leaving the light
squares weak.
10 Bxc6 bxc6 11 Bg5 h6 12 Bh4 a5!
Trying to strand the pawn on c5.
13 a3 Nd7 14 Bxe7 Qxe7 15 Rc1 Ba6
Tying White up.
16 0-0
16 b4 can be refuted in the simplest manner: 16 … axb4 17 axb4 Bxe2 18 Nxe2 Ra2,
winning for Black.
16 … Qxc5
16 … Bxe2 17 Rfe1 Qxc5 18 Nxe2 gives a few scrambling chances. There is no need to
allow White that.
17 Rfd1
17 Qxc5 Nxc5 18 Rfd1 Rfe8 19 Rc2 Bc4 20 Nd4 Nb3 leaves White a pawn down for
nothing.
17 … Rfe8 18 Na4 Qd6 19 Ng3 Qe5 20 Qd2
20 f4 Qxd4+ 21 Rxd4 Bb5 22 Nc3 Rab8 23 Nf5 may have been a little better than the
game for Mariotti, but Black is still a clear pawn up.
20 … Bb5 21 Re1 Qg5 22 Qc2 Rab8 23 Nc3 Bc4 24 Rxe8+ Rxe8 25 Nce4 Qf4 26
Qc3 Qe5 27 Qd2 Nf6 28 Nc5
 28 Nd6 Qxd6 29 Rxc4 c5 30 Rc1 Qb6 comfortably keeps the extra pawn.
28 … Rb8 29 Na4 Qg5 30 f4 Qg4 31 Nc3 h5 32 Nd1 Bb5 33 Nf2 Qd7 34 Qc2 a4 35
Nf5 d4 36 Qc5

36 … Rd8?
A blemish on what has been a smooth performance so far. He should play 36 … d3! and
I don’t think White will survive; e.g. 37 Rd1 Rd8 38 Ne4 Nd5 (38 … Nxe4? 39 Ne7+) 39
Ned6 Nxf4 40 Qe3 Ne6 with Black now two pawns up.
37 Nxg7?
This looks like mutual time-trouble. Mariotti should have played 37 Ne7+ Kh8 38
Nxc6 Bxc6 39 Qxc6 Qxc6 40 Rxc6 d3 41 Rc1, when he may yet save the game.
37 … Qd5! 38 Nf5 Qxc5 39 Rxc5 d3 40 Ne3 d2 41 Nfd1
41 Ned1 Re8 42 h3 Re1+ 43 Kh2 Nd5 44 Kg3 Be2 leaves White helpless.
41 … Be2 42 Rxc6 Ng4 0-1
For most of that game White was giving pawn odds.

There does not seem any further need to investigate this system, when Black can get
such a good position after 5 … Nc6 6 Bb5 e5!. Yet if you are a gambler or think the
opponent is due for a surprise, there may still be an argument to play 5 c5?!.
Game 86
N.Short-P.Eljanov
Malmö 2017

1 e4 c6 2 f4!?
Quite rare at this point, but 2 f4 cannot be too bad. I suppose it mostly leads to a static
pawn formation in the centre, which not everybody likes. As Nigel Short gets older, he has
tended to play a wide range of opening ideas and he seems quite happy to push the f-pawn
forward with the usual aggressive intent.
2 … d5 3 e5 Bf5!
Black takes up his usual Caro stance and should be able to get to a very solid position.
4 Nf3 e6 5 d4
5 d3!? hasn’t been played too much either; although White blocks in the bishop on f1,
he could well play g2-g3 and Bg2 in the near future.
5 … Nh6
5 … h5 is another idea, but Eljanov does not want to commit himself unnecessarily at
this point in time. Black is looking now to play … c6-c5 or … f7-f6 at the right moment.
6 Nc3 Bg4 7 Bd3?
Surprisingly poor. I think both players are already improvising, which is probably what
Short wanted and something Eljanov was happy to fall in with. 7 h3! is a bit better, but
even there 7 … Bxf3 8 Qxf3 Nf5 9 Be3 Nxe3 10 Qxe3 Qa5! is very comfortable for
Black.
7 … Nf5
The problem with 7 Bd3 becomes clear, as there is no easy way to defend d4.
8 Bxf5
8 Be2 c5 9 0-0 Nc6 10 dxc5 Bxc5+ 11 Kh1 h5 gives Black a very good position indeed.
8 … Bxf5 9 0-0 Be7 10 a4 c5!
Since White has played a2-a4, Black hits back in the centre, frees up his queen and is
delighted to control all the light squares. After ten moves, we can see that Short’s opening
experiment has failed. Black is better!

 11 Be3 c4 12 Ne2 h5 13 Ng3 Bg6 14 Ne1 Qd7 15 Qf3
Nigel must have considered 15 f5!?, the idea being 15 … Bxf5 (or 15 … exf5 16 Qf3)
16 Nxh5, but Black can keep a stable and better position after 16 … Rh7 17 Ng3 Bg6.
There is no progress to be made on the f-file.
15 … h4 16 f5!? hxg3 17 fxg6 gxh2+ 18 Kh1 f5! 19 g4
19 exf6 Bxf6 is a road block, so Short continues to try to batter through with his pawns.
19 … f4! 20 Ng2
20 Bxf4 Nc6 21 c3 0-0-0 looks like a strong attack coming for Black.
20 … fxe3
Basically, Short’s attack is the best that he can come up with in a position that was
getting progressively worse. Maybe the easy way that Black has handled 2 f4 in this game
tells us why the move isn’t played that much.
21 Qf7+ Kd8 22 Nf4 Kc7 23 Nxe6+ Kb6 24 a5+!
White gives Black no time to play … a7-a6, gaining a bolthole for his king.

24 … Ka6??
Putting the king on a6 allows White tactical opportunities. He should have preferred 24
… Kc6, which wins: 24 … Kc6 25 Qxg7 Na6 26 Qf7 e2 27 Rf2 Nc7 28 g7 Nxe6! 29
gxh8Q Rxh8 30 Rxe2 Nxd4 31 Rd2 Rf8 32 Qg6+ Qe6 33 Qxe6+ Nxe6. Black is fully in
control here.
25 b4?
A mistake for a mistake, although Short had only one minute remaining. Instead, 25
Nxg7! turns the tables and now it is White who may be winning: 25 … Nc6 26 Ne6 e2 (26
… Qc8 27 g7 Rg8 28 Qxg8! or 26 … b6 27 axb6+ Kxb6 28 Ra6+! wins) 27 Rfe1 b6 28
g7 Rhg8 29 axb6+ Kxb6 30 Rxe2 a5 31 Rxh2. The knight on e6 is in charge
25 … cxb3 26 Rfb1 Rc8 27 Qf1+ Qb5 28 Qxb5+ Kxb5 29 Rxb3+ Ka6 30 Rxe3
He should surely try 30 Nxg7 Nc6 31 Nf5.
30 … Rc6 31 Nxg7 Rxg6 32 Ne8 Rc6 33 c3 Nd7 34 Ng7 Rac8 35 Nf5 Bg5 36 Rh3
Rxc3 37 Rxc3 Rxc3 38 e6 Nf6 39 e7 Rc1+ 40 Rxc1 Bxc1 41 Nd6 Bf4 42 e8Q Nxe8 43
Nxe8 Kxa5 44 Nf6 Be3 45 Nxd5 Bxd4 46 g5 b5 47 Ne7 Bc3 48 g6 b4 49 Nc6+ Kb5 50
Nxa7+ Kc5 0-1
A very spirited attack by White, conjured up from what was becoming a very bad
strategic day at the office. Could we say that Nigel was unlucky? Not really. 2 f4 will
continue to be played from time to time though and Caro players have to show respect.
Finally, a sideline of the Exchange Variation, which has been relatively uncommon
until recently.
Game 87
V.Kramnik-V.Anand
Grand Chess Tour Blitz, Leuven 2017

1 e4 c6 2 Nf3!? d5 3 exd5 cxd5 4 d4 Bg4!

I think this is the most accurate move. 4 … Nf6 5 Ne5!? is White’s tricky idea. It
appears odd to move the knight twice, but Black must not underestimate this idea. Carlsen
has tried it twice:
1) M.Carlsen-P.Leko, Miskolc (rapid match) 2008, continued 5 … g6 6 Bd3 Bg7 7 0-0
0-0 8 c3 Nc6 9 Qe2 Nd7 10 f4 Nf6 11 Be3 Bf5 12 Nd2 Bxd3 13 Qxd3 e6 14 Rae1 Rc8,
when White has a small lead in development, but that is all. Black’s position is perfectly
solid.
2) 5 … Nc6 should be okay for Black after 6 Bb5 Bd7 7 Nxd7 Qxd7 8 c3 a6 9 Bd3 and
now either 9 … e5 or 9 … e6. Instead, in M.Carlsen-A.Dreev, World Blitz Championship,
Dubai 2014, Black tried 9 … Qg4?! and after 10 Qxg4 Nxg4 Carlsen had just the type of
situation any strong player would like as White – he can play on forever risk-free with the
two bishops. The position probably is equal, but Black should avoid such situations.
Carlsen eventually won.
5 h3 Bxf3 6 Qxf3 e6 7 Bd3 Nf6 8 0-0 Bd6 9 c3
White has obtained the bishop pair, but the position is not really dangerous for Black if
he keeps the queens on the board. Anand’s next move is careful.
9 … h6
Naturally, Black could play something like 9 … 0-0 10 Bg5 Nbd7 11 Nd2 Qb6, but
Anand wants his knight on c6.
10 Nd2 0-0 11 Qe2 Qc7 12 g3 Nc6 13 f4 a6 14 Nf3 b5

Setting up the minority attack to gain counterplay and in this case supporting … Na5-
c4. White might claim he has a good Stonewall and chances of a kingside attack, but
Black has resources.
15 a3 Na5 16 Ne5 Nc4 17 a4 Rab8 18 axb5 axb5 19 g4 Nd7?
As this was a blitz game we must forgive mistakes. 19 … Ra8! would have been pretty
safe.
20 Ra6?
I don’t see what Black does after 20 Nxd7 Qxd7 21 b3, winning a pawn.
20 … Bxe5 21 fxe5 Ra8 22 Rxa8 Rxa8 23 g5 hxg5 24 Bh7+
The mutual error earlier suggests that both players were already short of time and
Kramnik elects to launch a drawing combination. 24 … Kxh7 25 Qh5+ Kg8 26 Qxf7+
Kh7 27 Qh5+ Kg8 28 Qf7+ Kh7 29 Qh5+ Kg8 30 Qf7+ Kh7 31 Qh5+ ½-½
White’s opening idea has merit and is sufficiently different to the normal Exchange
Variation to give Black novel problems. I recommend 4 … Bg4, which will give Black his
usual solid position.
Index of Complete Games
Anderson.J-Hawkins.J, Hastings 2014/15
Andreikin.D-Bocharov.D, Russian Team Championship 2017
Andriasian.Z-Motylev.A, European Blitz Championship, Tallinn 2016
Arizmendi Martinez.J-Houska.J, Llucmajor 2017
Bacrot.E-Ponomariov.R, Biel 2017
Beliavsky.A-Larsen.B, Tilburg 1981
Bentivegna.F-Rombaldoni.D, Torino 2010
Bilek.I-Smyslov.V, Tel Aviv Olympiad 1964
Bruzon Batista.L-Moreno Mateos.J, Calvia 2006
Carstensen.J-Rasmussen.K, Danish Championship, Skorping 2017
De Firmian.N-Morris.M, Chicago Open 1995
Dragun.K-Macieja.B, US Masters, Greensboro 2017
Duboue.P-Battaglini.G, Nimes 2014
Eggleston.D-Short.N, British Championship, Sheffield 2011
Ehlvest.J-Timman.J, Tal Memorial. Riga 1995
Fedoseev.V-Bacallao Alonso.Y, FIDE World Cup (rapid tiebreak), Tbilisi 2017
Firouzja.A-Dreev.A, Guilan 2016
Fischer.R.J-Keres.P, Candidates Tournament, Bled/Zagreb/Belgrade 1959
Fischer.R.J-Petrosian.T, USSR vs. Rest of the World, Belgrade 1970
Gheorghiu.F-Hort.V, Alekhine Memorial, Moscow 1971
Giacomini.P-Adly.A, Menton 2016
Gipslis.A-Seirawan.Y, Saint John 1988
Granda Zuniga.J-Leon.O, Panama City (blitz) 2017
Hair.P-Yee.S, Asian Seniors Championship, Auckland 2017
Hamdouchi.H-Bauer.C, French Championship, Nancy 2013
Harini.S-Amin.B, Dubai 2017
Henris.L-Gharamian.T, Charleroi 2017
Hess.R-Stripunsky.A, US Championship, St. Louis 2013
Hou Yifan-Rapport.R, FIDE Grand Prix, Sharjah 2017
Huerga Leache.M-Larino Nieto.D, Elgoibar 2010
Itkis.B-Van Wely.L, Romanian Team Championship 2000
Ivanov.B-Vladimirov.I, Ruse 2017
Jones.G-Lalic.S, British Championship, Sheffield 2011
Karjakin.S-Mamedyarov.S, FIDE Grand Prix, Zug 2013
Kasparov.G-Bertacco.P, Simultaneous display, Asiag 1991
Kovalev.V-Riazantsev.A, Russian Rapid Team Championship 2017
Kramnik.V-Anand.V, Grand Chess Tour Blitz, Leuven 2017
Kramnik.V-Ivanchuk.V, FIDE World Cup, Tbilisi 2017
Kravtsiv.M-Cheparinov.I, Turkish Team Championship 2017
Lenic.L-Ragger.M, Austrian Team Championship 2017
Lu Shanglei-Mamedyarov.S, World Blitz Championship, Doha 2016
Mariotti.S-Magem Badals.J, Andorra Zonal 1987
Martin.A-Kumaran.D, British Championship, Dundee 1993
Martin.A-Ledger.A, British Championship, Plymouth 1992
Maze.S-Gharamian.T, French Championship, Nimes 2014
Mihok.O-Annaberdiev.M, Budapest 2017
Motylev.A-Anand.V, Russia vs. Rest of the World, Moscow (rapid) 2002
Murugappan.G-Winslow.E, San Francisco 2016
Nepomniachtchi.I-Laznicka.V, European Team Championship, Heraklio 2017
Nevednichy.V-Grant.J, European Team Championship, Heraklio 2017
Nimzowitsch.A-Capablanca.J.R, New York 1927
Nunn.J-Arkell.K, Lloyds Bank Open, London 1987
Oberhofer.A-Dautov.R, German Bundesliga 2017
Onischuk.V-Dimitrov.R, Corsica 2017
Popilski.G-Nakamura.H, Millionaire Open, Las Vegas 2015
Portisch.L-Wade.R, Palma de Mallorca 1966
Quparadze.G-Khamrakulov.D, Ordu 2017
Rapport.J-Houska.J, Isle of Man 2017
Riazantsev.A-Jakovenko.D, FIDE Grand Prix, Sharjah 2017
Rodshtein.M-Arkell.K, Hastings 2014/15
Sambuev.B-Hernandez Gonzalez.W, Capablanca Memorial, Matanzas 2017
Savchenko.B-Maghsoodloo.P, Khomeyn 2017
Sax.G-Martin.A, Hastings 1983/84
Schalkwijk.P-Melia.K, Correspondence 2000
Selivanov.I-Burmakin.V, Chigorin Memorial, St. Petersburg 2016
Shirov.A-Anand.V, Leon (rapid) 2011
Short.N-Eljanov.P, Malmö 2017
Spassky.B-Karpov.A, Candidates Semi-final (Game 6), Leningrad 1974
Stebbings.A-Trois.F, World Seniors Championship, Acqui Terme 2017
Stefansson.H-Granda Zuniga.J, Reykjavik 2015
Stein.L-Barcza.G, Caracas 1970
Stein.L-Hort.V, Interzonal Playoff, Los Angeles 1968
Stukopin.A-Nyzhnyk.I, Kenner 2016
Surtees.M-Houska.J, British Championship, Liverpool 2008
Tal.M-Botvinnik.M, World Championship (Game 8), Moscow 1961
Tal.M-Gurgenidze.B, USSR Championship, Alma-Ata 1968
Timofeev.A-Khenkin.I, Capablanca Memorial, Havana 2009
Torre.E-Korchnoi.V, Buenos Aires Olympiad 1978
Trent.L-Arkell.K, British Championship, Liverpool 2008
Trent.L-Storey.C, British Championship, Llandudno 2017
Tsoi.D-Burmakin.V, Chigorin Memorial, St. Petersburg 2017
Van der Wiel.J-Miles.A, Brussels 1986
Vladimirov.V-Stupak.K, Agzamov Memorial, Tashkent 2017
Vocaturo.D-Prohaszka.P, Benasque 2017
Vykouk.J-Simacek.P, Ceske Budejovice 2017
Zherebukh.Y-Nakamura.H, US Championship, St. Louis 2017
Zhigalko.A-Hovhannisyan.R, Asrian Memorial, Yerevan 2016