On The MARKET

Trend-following winners
are not lucky monkeys
Trend following gets a bad rap in some circles, but there’s no denying
the profits some trend traders have reaped over the years.

BY MICHAEL COVEL

F
amed Stanford University psychologist Leon Festinger once • John W. Henry, who bought the Red Sox through trend fol-
said, “A man with a conviction is a hard man to change. lowing.
Tell him you disagree and he turns away. Show him facts or • Bruce Kovner, who is worth more than $4 billion.
figures and he questions your sources. Appeal to logic and • Bill Dunn, who made $80 million in 2008.
he fails to see your point.” • Michael Marcus, who turned an initial $30,000 into $80
Although trend following has been one of the most successful million.
trading strategies for decades, some critics downplay the massive • David Harding, who is worth more than $690 million.
profits accumulated by trend followers, arguing there are just a • Ed Seykota, who turned $5,000 into $15 million in 12 years.
few chance winners — “lucky monkeys,” they claim. • Kenneth Tropin, who made $120 million in 2008.
Not true. Large numbers of trend followers have found a way to
outpace market averages. They have done so with hard work and Most of these traders did not come from privileged back-
the ability to stick with a trading plan — usually for a very long grounds; many did not learn trend trading in college. They came
time. However, some detractors seem happy to snub their nose at from different disciplines, worked disparate jobs, and saved a
success, perhaps even until they have wasted a lifetime on sub-par nest egg to begin trading with trend-following rules. They are
trading strategies. Others, however, choose to test, accept proof, proof that anyone can rise to the top if ambitious enough.
and build toward a profitable life. Who are some inspirational win- Some people don’t buy it, however. They gripe, “I can’t be a
ners? The list of successful trend-following traders includes: trend follower because you need a huge portfolio, which requires
lots of capital.” First, part of that argument is correct. You can’t
The purpose of trading is to make trade if you are broke. However, trading as a trend follower
doesn’t necessarily mean massive capital. Micro and mini futures
money, not to be right — a distinction contracts, ETFs, or LEAPs give just about everyone the opportu-
nity to use trend-following techniques.
that separates trend-following from
Then there’s the argument that noting the achievements of
other trading philosophies. continued on p. 18

16 www.activetradermag.com • ACTIVE TRADER

 On The Market

successful trend followers ignores the reality of the losing traders:

TABLE 1: EXAMPLE SYSTEM A
“This trend-following firm failed, which means the winners are
Average winning trade $400
just lucky survivors.” The problem with this analysis is that it only
looks at the end. After-the-fact statistical analysis is not a substitute Average losing trade $550
for direct observation. There are usually many reasons for failure, Ratio 0.73
and if you don’t look at the exact cause of failure — assuming Winning percentage 65%
some traders were just the unlucky losers makes little sense.
How does one go about ensuring success? Interestingly, the representing overall results. Of course, there will be runs where
performance data of trend-following traders offers fantastic nearly every trade is profitable, and there will be runs where
insights on how to become a more successful trader. One of the nearly every trade is a loser. This is the nature of trading. You
keys to trading success is learning how to extrapolate from a cannot expect 6.5 winners for every 10 trades taken. Trading is
professional trader’s track record. random and there will be deviations from the mean.
Let’s start with a cold, hard truth: All traders suffer drawdowns Turning back to our example system, if the winning percentage
at some time. No matter the trading style or duration, all traders drops to 50 percent you should expect losses:
have times when they lose. Long periods of losses, accumulated
into drawdowns, are often the trigger that forces traders out. Expectation = (400 * 0.50) - (550 * 0.50)
However, analyzing a system’s “breakeven percent,” which is = 200 - 275
the minimum percentage of profitable trades needed to produce = -$75
profits for a given average winning trade/losing trade ratio, can
show how wrong you can be before actually losing a substantial Solving for the breakeven percent shows the system breaks
amount of money. For example, Table 1 shows statistics for a even at 58 percent; below this threshold the system will generate
hypothetical trading system. losses.
At first glance, the 65-percent winning statistic looks promis- Now consider the hypothetical trading system in Table 2. At
ing. However, if you look closer at the data you will find pos- first glance the idea of being right only 45 percent of the time
sible issues. The system has a positive mathematical expectation might be difficult to handle. After all, who wants to be wrong 55
of $67.50, meaning it will generate profits at these levels at 65 percent of the time? However, the system’s mathematical expecta-
percent accuracy. tion is larger than System A’s:

Expectation = (avg. win trade * win pct.) - (avg. losing Expectation = (650 * 0.45) - (325 * 0.55)
trade * (1-win pct.)) = 292.50 – 178.75
= (400 * 0.65) - (550 * (1-.65)) = $113.75
= (400 * 0.65) - (550 * 0.35)
= 260 - 192.50 Solving for the breakeven percent shows the system must have
= $67.50 a winning percentage of 33.34 percent to break even. The system
allows you to be wrong 66.65 percent and still breakeven. It is
The winning percentage here is just an aggregate statistic easier to produce profits when a system allows one to be wrong

18 www.activetradermag.com • ACTIVE TRADER

 FIGURE 1: LOWER WINNING PERCENTAGE, BETTER RESULTS

more than half the time.

Using a randomly distributed perfor-
mance result to project the two systems’
equity curves highlights their differ-
ences. System A has a win/loss ratio of
0.73, while System B has a win/loss ratio
of 2.00. The effects of a larger win/loss
ratio are obvious in Figure 1. System B
produces an approximately $11,300 gain
while System A produces an approximately
$6,800 gain.
We all want to be right and have that
“magic method that produces big winners.
That’s human nature. However, the pur-
pose of trading is to make money, not to
be right. The ability to make this distinc-
tion separates the trend-following mind-set
from other types of trading philosophies
and techniques.
Using our breakeven insights begins
the process of evaluating why trend fol-
lowing wins and why it seems to have
greater longevity than other trading styles.
Consider performance data from the six
trend-following programs in Table 3 (Jan.
1, 2006 to Sept. 2011). The higher the
average win/loss ratio on a monthly basis,
the lower the breakeven percent. They
programs can afford to be wrong more
than 50 percent of the time and still earn
profits.
The same lesson applies to individual
traders looking to go down the trend-
System B has a much lower winning percentage than System A, but
following path: Larger win/loss ratios have because its average winning trade/losing trade is so much higher than
one of two sources — small losses or out- System A’s — typical of a trend-following strategy — it is much more
sized gains. Why? Trend followers tend to profitable.
continued on p. 20

19
 On The Market

have large and unpredictable months because they are constantly

TABLE 2: EXAMPLE SYSTEM B
in position to profit whenever markets move substantially higher
Average winning trade $650
or lower. The hunt for those “black swans” — outlier events —
begins with answering a handful of key questions. Average losing trade $325
Ratio 2.0
The questions Winning percentage 45%
What is the foundation of trend-following success? Profitable
trend-following traders ask five critical questions before putting
capital at risk: During these very uncertain times — times so perfectly suited
for a black swan to swim in and surprise — “tail risk” can hit.
1. What market do you buy or sell at any time? Tail risk, or the possibility that an investment will move beyond
2. How much of a market do you buy or sell at any time? three standard deviations, should be the only discussion today.
3. When do you buy or sell a market? However, this discussion is usually broached without mentioning
4. When do you get out of a losing position? the only true tail-risk hedge: trend following.
5. When do you get out of a winning position? Consider analysis from noted market pro James Montier from
his report “A Value Investor’s Perspective on Tail Risk Protection:
Stated another way: An Ode to the Joy of Cash” (GMO, June 2011): “When consider-
ing tail risk protection, investors must start by defining the tail
1. What is the state of the market? risk they are seeking to protect themselves against . . . The hard-
2. What is the volatility of the market? est element of tail risk protection is likely to be timing. It is clear
3. What is the equity being traded? that a permanent allocation is likely to do more harm than good
4. What is the system or trading orientation? in many situations. When it comes to timing tail risk protection,
5. What is the risk aversion of the trader or investor? continued on p. 21

TABLE 3: TREND-FOLLOWING PROGRAM BREAKEVEN PERCENT

Average win Average loss Average win/ Monthly breakeven
Program
(monthly) (monthly) loss ratio percentage
Dunn 8.48% -7.04% 1.20 45.36%
EMC Classic 4.12% -3.61% 1.14 46.70%
Fall River Global HL 4.60% -3.34% 1.38 42.06%
Hawksbill 8.50% -5.28% 1.61 38.31%
Saxon Aggressive 4.08% -2.80% 1.46 40.70%
Winton 3.05% -2.13% 1.43 41.12%
These trend-following programs illustrate the higher the average winning trade/losing trade ratio, the lower the
winning percentage the system requires to break even.

20 www.activetradermag.com • ACTIVE TRADER

 Related Reading

Books by Michael Covel:

Trend Commandments: Trading for Exceptional

a long-term value-based approach and an emphasis on absolute Returns (FT Press, 2011).
standards of value, coupled with a broad mandate (a wide oppor-
tunity set, or, investment flexibility, if you prefer) seems to offer The Little Book of Trading: Trend Following
the best hope.” Strategy for Big Winnings (Wiley, 2011).
Value investing is the solution to tail risk? No. That is a tired
argument considering buy-and-hold/value philosophies have The Complete TurtleTrader (HarperCollins, 2009).
languished for 13 years. News flash: Buy-and-hold forever is not
a strategy. Further, Montier argues that timing is the hardest part? Trend Following (FT Press, 2009).
Recall the five questions — timing is but one part of an overall
trading plan.
Remember, trend followers have no idea if there is a current
bubble in stocks, bonds, commodities, or any other market. They
don’t know when the next tail risk event will appear, but they
are ready to act, and prosper no matter when the event or trend
unfolds. On the flip side, it is truly amazing how some people
try to convince others that they know, and offer value investing as
the solution. Has there ever been a time when the majority knew
it was in the middle of a bubble and acted accordingly? Bubbles
are never clear until the dust settles and they are popped. That’s
why trend following works well — it doesn’t require advance
knowledge of a future surprise event to profit.
In short, a proven path to profits is having a trend-following
mind-set ingrained in your psyche before a big dip or rise in the
markets, having predefined entry and exits rules to your trade,
and having that rigid process in place. Don’t forget for a second
the ugly alternatives. On average, according to the Wall Street
Journal, 95 percent of Wall Street’s recommendations are to buy.
No sell strategy? That equals trouble — big trouble.
In the end, you have to decide what you want. Do you want
the exciting feeling of listening to some guru stock picker at-
tempting to predict what will happen tomorrow, or do you want
to control how much you can lose of your limited capital and
ride unpredictable trends for possible lifelong profits? ◆
For information on the author, see p. 8.

21