Journal of Strength and Conditioning Research, 2006, 20(3), 492–499

䉷 2006 National Strength & Conditioning Association

DYNAMIC VS. STATIC-STRETCHING WARM UP: THE

EFFECT ON POWER AND AGILITY PERFORMANCE
DANNY J. MCMILLIAN,1 JOSEF H. MOORE,2 BRIAN S. HATLER,3 AND DEAN C. TAYLOR3
U.S. Army MEDDAC, Heidelberg, Germany; 2U.S. Army-Baylor University Doctoral Physical Therapy Program,
1

U.S. Army Medical Department Center and School, Fort Sam Houston, Texas 78234; 3Keller Army Community
Hospital, West Point, New York 10996.

ABSTRACT. McMillian, D.J., J.H. Moore, B.S. Hatler, and D.C. demands of their sport (12). Although many variations on
Taylor. Dynamic vs. static-stretching warm up: The effect on the DWU theme exist, most feature progressive, contin-
power and agility performance. J. Strength Cond. Res. 20(3):492– uous movement. Calisthenics such as squatting and lung-
499. 2006.—The purpose of this study was to compare the effect ing movements often are paired with running drills that
of a dynamic warm up (DWU) with a static-stretching warm up
(SWU) on selected measures of power and agility. Thirty cadets
include forward, lateral, and change-of-direction move-
at the United States Military Academy completed the study (14 ment. Investigators have shown DWU to improve knee
women and 16 men, ages 18–24 years). On 3 consecutive days, joint position sense, to increase oxygen uptake, to lower
subjects performed 1 of the 2 warm up routines (DWU or SWU) lactate concentration and raise blood pH, to improve ef-
or performed no warm up (NWU). The 3 warm up protocols last- ficiency of thermoregulation, and to improve performance
ed 10 minutes each and were counterbalanced to avoid carryover for bicycle sprints and vertical jumps (1, 6, 7, 15, 20, 37).
effects. After 1–2 minutes of recovery, subjects performed 3 tests Recently the United States Army Physical Fitness
of power or agility. The order of the performance tests (T-shuttle School (APFS) developed a DWU for individuals and mil-
run, underhand medicine ball throw for distance, and 5-step itary units. The stated objectives are to increase body
jump) also was counterbalanced. Repeated measures analysis of
temperature and heart rate, pliability of joints and mus-
variance revealed better performance scores after the DWU for
all 3 performance tests (p ⬍ 0.01), relative to the SWU and cles, and responsiveness of nerves and muscles in prep-
NWU. There were no significant differences between the SWU aration for physical readiness training activities. This
and NWU for the medicine ball throw and the T-shuttle run, but DWU was used before each exercise session as part of an
the SWU was associated with better scores on the 5-step jump intervention to decrease injuries and to improve physical
(p ⬍ 0.01). Because the results of this study indicate a relative performance among soldiers in a basic training battalion.
performance enhancement with the DWU, the utility of warm Static stretching, a prominent feature of the warm up for
up routines that use static stretching as a stand-alone activity generations of soldiers, was not included. Although mul-
should be reassessed. tiple interventions confounded the effect of the DWU, in-
KEY WORDS. flexibility, performance testing, conditioning, cal- jury rates over the 9-week training period were signifi-
isthenics cantly decreased compared with both a control battalion
and historic trends. Performance on physical fitness test-
ing generally was improved (19).
INTRODUCTION Given the ubiquity of static stretching in warm up ac-
re-exercise warm up routines are common prac- tivities, the purpose of this study was to compare the ef-

P tice, despite limited scientific evidence support-

ing one protocol over another. For this reason,
warm up protocols tend to reflect the experi-
ence of individual coaches, trainers, and ath-
letes. Traditionally, static-stretching exercises have been
fect of a DWU (based on the APFS model) with that of a
static-stretching warm up (SWU) or no warm up (NWU)
on selected measures of power and agility. Dependent
variables to assess power and agility were chosen, be-
cause these attributes are common requirements for a va-
a prominent feature of warm up routines (6, 34, 37). Sup- riety of sports. The DWU protocol in this study closely
port for a more dynamic warm up (DWU) has grown in mimicked the power and agility requirements of many
recent years, because several investigations have shown sports, so we hypothesized that it would result in a per-
the potential for acute, static stretching to degrade per- formance enhancement relative to the SWU or NWU.
formance on vertical jumps, short sprints, tasks requiring
maximal voluntary contractions, muscle strength-endur- METHODS
ance performance, balance challenges, and reaction time
Experimental Approach to the Problem
(2, 3, 6, 10, 11, 20, 21, 23, 27, 37). Additionally, several
studies now indicate that pre-exercise static stretching All subjects attended a 2-part orientation session that in-
does not offer the presumed benefit of injury risk reduc- cluded (a) instruction for active participation in both the
tion (5, 14, 17, 26, 30). DWU and SWU and (b) practice of the 3 performance
Smith (32) indicated that the general purpose of a pre- measures (i.e., the T-drill, 5-step jump, and the medicine
exercise warm up is to increase muscle and tendon sup- ball throw for distance). The independent variable was
pleness, to stimulate blood flow to the periphery, to in- the type of warm up used before performance testing. The
crease body temperature, and to enhance free, coordinat- dependent variables were the scores on the 3 performance
ed movement. Professionals in the strength and condi- measures. During the orientation, subjects were given
tioning community have increasingly touted various feedback to enhance proper execution of both the warm
DWUs as the best way to prepare athletes for the physical up techniques and the performance measures. To ensure

492
 EFFECT OF WARM UP PARAMETERS 493

that subjects had mastered the techniques for the perfor- were aggregate lower body power and stability. Wiklan-
mance measures, they repeated each of the 3 events until der et al. have shown the 5-step jump to be a reliable
their scores no longer improved. Rest between trials of measure that correlates well with the vertical jump, long
the T-drill lasted approximately 2 minutes. Rest between jump, and isokinetic leg strength (35).
trials of the 5-step jump and medicine ball throw for dis- The medicine ball throw for distance was chosen as a
tance was at the subject’s discretion, but generally of 30– measure of total-body power. Stockbrugger et al. have
60 seconds’ duration. Subjects were encouraged to take as shown this test to be a valid and reliable test for assessing
long as necessary to recover from the previous effort. explosive power for an analogous total-body movement
pattern and general athletic ability (33).
Subjects The T-drill was chosen primarily as a measure of agil-
Thirty cadets at the United States Military Academy ity. For this test, the component tasks of (a) forward,
(USMA) volunteered for and completed the study. Sub- backward, and lateral running; (b) stopping and changing
jects were recruited from USMA club sports. Cadets were direction; and (c) reaching with an upper extremity while
eligible for the study if they were fit for full military duty lowering the center of gravity are all representative of
without restrictions. All subjects completing the study commonly encountered tasks in sports. Pauole et al. have
were members of rugby, lacrosse, or strength and condi- shown this test to be a valid and reliable measure of agil-
tioning teams. Members of the rugby and lacrosse teams ity, leg power, and leg speed in college-age men and wom-
were competing weekly. In addition, all cadets have rou- en (25). To emphasize lateral movement, the forward- and
tine physical requirements. For these reasons, the sub- backward-run portions of the T-drill were set at 5 m rath-
jects were screened by the primary investigator before the er than the 10-yd distance described by Pauole.
study to establish eligibility and before each training or Data collection began the day after the orientation
testing session to ensure continued eligibility. Exclusion and ran for 3 consecutive days. Subjects performed 1
criteria were: (a) acute impairment of the spine or lower warm up protocol (DWU, SWU, or NWU) before data col-
extremities, vestibular dysfunction, or balance disorder, lection each day. Subjects were instructed to avoid exer-
(b) history of surgery in either lower extremity, and (c) cise or vigorous physical exertion the morning of testing.
history of a neurological disorder affecting the upper or All tests were conducted at 6 AM at the same test site
lower extremities. All subjects gave written, informed each day.
consent prior to participation in the study. The mean ⫾ After completing 1 of the warm up conditions (or 10
standard deviation (SD) for age, height, and weight for minutes of rest for the NWU group), subjects proceeded
the 16 men were 20.2 ⫾ 1.2 years, 182.4 ⫾ 6.6 cm, and to the performance testing stations. The time between
88.8 ⫾ 9.0 kg, respectively. The mean ⫾ SD for age, finishing the warm up and beginning the performance
height, and weight for the 14 women were 20.4 ⫾ 1.5 testing was approximately 2 minutes. The order of testing
years, 167.1 ⫾ 7.9 cm, and 64.0 ⫾ 7.8 kg, respectively. was counterbalanced to avoid carryover effects. A physi-
All subjects gave written informed consent prior to par- cal therapist or physical therapy assistant who was un-
ticipation. The study was approved by the Human Sub- aware of the subject’s group assignment scored each per-
jects Research Review Board of Keller Army Community formance test. None of the investigators participated in
Hospital, West Point, NY. data collection. The primary investigator then compiled
all data for analysis.
Warm Up Protocols Attempts were made to control potentially confound-
Subjects executed the warm up sessions in small groups ing variables. For example, (a) testing occurred at 6 AM
with the primary investigator leading the DWU (Table 1) each day, with subjects advised not to eat or drink any-
and an associate investigator (BH) leading the SWU (Ta- thing other than water before testing; (b) subjects were
ble 2). The order in which the subjects performed the 3 queried for injuries, illness, or excessive fatigue each day;
warm up conditions was counterbalanced to avoid poten- (c) subjects were reminded of the importance of maximal
tial biasing effects associated with test sequence. Each effort each day before testing; (d) graders were either
warm up session lasted 10 minutes. Subjects scheduled physical therapists or physical therapy assistants with at
for NWU rested in an area adjacent to the testing site. least 1 year of experience collecting performance mea-
surement data for another study; and (e) graders received
Performance Testing a standardized orientation to the measurements required
Most recent investigations of pre-exercise stretching have for the study.
used vertical jump tests as the measure of performance. Ten subjects were removed from the study after com-
The present investigation used other performance mea- pleting the orientation: 2 subjects for excessive fatigue
sures (Table 3) in order to evaluate agility as well as a from cadet physical requirements the previous day, 2 sub-
broader spectrum of tasks requiring power. Care was tak- jects from injuries related to military training, and 6 sub-
en to avoid tasks that would induce fatigue, because fa- jects for missed testing sessions. No subjects were injured
tigue has been shown to hinder local muscular perfor- during the performance of either of the warm up condi-
mance, especially for tasks that involve the stretch–short- tions or performance testing.
ening cycle.(18)
The 5-step jump was chosen as a measure of function- Statistical Analyses
al leg power. Single-leg hop tasks are used more com- Pre hoc power analysis was used to establish the appro-
monly as functional tests; however, they are most often priate sample size, based on the following parameters:
used to assess symmetry, and therefore normalcy, of the effect size ⫽ 0.27 (based on previously reported data on
lower extremities following unilateral lesions, such as an- the 5-step jump [35]), 3 degrees of freedom, power ⫽ 0.80,
terior cruciate ligament deficiency or reconstruction (9). and alpha ⫽ 0.05. Repeated measures (2 [gender] ⫻ 3
For our purposes, symmetry was less important than [warm up protocol]) analysis of variance (ANOVA) was
 494
TABLE 1. The dynamic warm up.
Exercises Execution
Calisthenics: Perform 10 repetitions of each exercise at a slow to moderate cadence unless otherwise indicated. All component movements are required for each repetition.
Bend and reach Reach high overhead. Squat and reaching between the legs, allowing the back to flex, but keeping the heels down. Return to the starting position.
MCMILLIAN

Perform at a slow cadence.

Rear lunge and Start with hands on hips. Lunge to the rear while simultaneously reaching overhead. Return to the starting position in one motion. Repeat with the
ET AL.

reach opposite leg. Keep most of the weight on the front leg. Lunge progressively further and deeper with each repetition. Keep the abdominals tight to
maintain a stable trunk. Perform at a slow cadence.
Turn and reach Stand with arms extended to the side at shoulder level with the palms up. Turn to the left and pause while keeping the pelvis facing forward. The
arms should now be directed forward and rearward. Return to the starting position, then repeat to the other side. Keep the abdominals tight
throughout. Keep the head directed forward throughout. Perform at a slow cadence.
Squat Start with hands on hips. Squat until the thighs are parallel to the floor (or to your tolerance). Keep the heels on the floor. The arms should be
raised to shoulder level for counterbalance.
Rower Start in the supine position with arms overhead, head a few inches off the ground with the chin slightly tucked. In one motion, raise to a seated
position, bend the knees to bring the feet flat, and bring the arms parallel to the ground.
Power jump Start with the arms high overhead, with the feet, knees and hips aligned vertically. Squat and reach toward the ground with the arms outside the
legs, keeping the back straight. Jump and reach overhead, landing in the squat position described above. Return to the starting position.
Prone row From the prone position with the arms overhead and several inches off of the ground, begin by raising the chest slightly and bringing the hands
back to shoulder level in a rowing motion. Maintain abdominal muscle tension throughout the exercise. The hands and elbows remain parallel to
the ground at all times. Maintain the neck in a neutral position.
Push-up At the starting position, the hands are directly under the shoulders or slightly wider. Elbows are straight, but not locked. The abdominals are
contracted to maintain the trunk in line with the thighs. Do not lower the trunk past the point at which the upper arms are parallel to the
ground. Perform at a moderate to fast cadence.
Windmill From a relatively wide stance with the arms extended sideways and palms down, squat, bend forward and rotate the trunk to the left in order to
reach the right hand to the left foot. Return to the starting position, then repeat to the opposite side. Keep the arms directed in opposite direc-
tions. Avoid excessive flexion of the spine.
Diagonal lunge Start with the arms high overhead. Lunge diagonally forward to the left while simultaneously lowering the hands to the lower leg. Return to the
and reach starting position in one motion. Repeat to the right. Keep the foot of the forward leg directed to the front, rather than in the direction of the
lunge. Keep the trunk straight and the head up. Do not allow the knee of the forward leg to go beyond the toes or lateral to the foot.
Movement drills: Perform each exercise over a 20- to 25-m segment. Pause for 10–15 seconds of rest and return to the start point. This completes 1 repetition. Perform 1
repetition of each exercise. Maintain a slow to moderate pace unless otherwise indicated.
Verticals Run forward on the balls of the feet, raising the knees to waist level and maintaining a tall, upright stance. Use strong arm action to support the
movement. Hands should move from waist to chin level with an approximately 90⬚ bend in the elbows throughout. There should be no backswing
of the legs with this drill.
Laterals Stand perpendicular to the direction of movement, in a slight crouch with the back straight. Step to the side by rising slightly and bringing the
trailing leg to the lead leg. Quickly hop to the side and land back in the crouch with the knees shoulder-width apart. Face the same direction for
the down and back segments.
Crossovers Stand perpendicular to the direction of movement, in a slight crouch with the back straight. Cross the trailing leg to the front of the lead leg and
step in the direction of travel to return to the starting position. Then cross the trailing leg to the rear of the lead leg and step in the direction of
travel to return to the starting position. Repeat this sequence to the 25-yd stopping point. Face the same direction for the down and back seg-
ments. Let the arms swing naturally side to side to support balance. Allow the hips to swivel naturally.
Skip Step and then hop, landing on the same leg, followed by the same action with the opposite leg. Use strong arm action to support the movement.
Hands should move from waist to chin level with an approximately 90⬚ bend in the elbows throughout. When the right leg is forward, the left arm
swings forward and the right arm is to the rear. When the left leg is forward, the right arm swings forward and the left arm is to the rear.
Shuttle sprint Run at a moderate pace to the 25-yd line. When nearing the line, slow the movement, make a quarter-turn clockwise, plant the left foot parallel to
the line, and squat or bend in order to touch the ground at the line. Run back to the starting line, turning counterclockwise to touch the ground
with the right hand. Run back to and through the 25-yd line, gradually accelerating to near maximum speed.
TABLE 2. The static-stretching warm up.*
Stretches Execution
Overhead arm Raise the right arm overhead and place the right hand behind the head. Grasp below the right elbow with the left hand and pull to the left, leaning
pull the body to the left. Repeat on the opposite side.
Turn and reach Stand with arms extended to the side at shoulder level with the palms up. Rotate the trunk to the left while keeping the hips directed forward to
bring the arms in line from front to rear. Keep the hips set and abdominal muscles tight throughout to prevent pelvic rotation. The head and eyes
remain directed forward. Hold for only 10–15 seconds to avoid shoulder fatigue. Repeat in the opposite direction.
Rear lunge and Step rearward with the left foot and reach overhead with both arms. This is the same position as the first movement of the rear lunge and reach in
reach the dynamic warm up. Repeat on the opposite side. Maintain straightness of the back by keeping the abdominal muscles tight throughout the
motion. Reach fully overhead with both arms. Arms should be shoulder width apart with palms facing each other with the fingers and thumb
extended and joined. After the foot touches down on counts 1 and 3, allow the body to continue to lower. This promotes a better opening of the hip
and trunk.
Hamstring Take a step forward with the left leg and reach toward the left foot by bending at the waist. Both knees are slightly bent and the arms are straight
stretch on either side of the forward leg. The trunk remains straight with the head in a neutral position. Repeat on the opposite side.
Calf stretch Step forward 8–10 in. with the left foot and place the heel on the ground with the toes up. Bend forward and grasp the sides of the left foot with
both hands. Gradually straighten the knee of the left foot and pull the ball of the foot back toward the shin. Attempt to keep the trunk straight
with the head in a neutral position. Repeat on the opposite side.
Quadriceps Lie on the left side. Grasp the right ankle or foot with the right hand. Pull the right heel toward the buttocks and the thigh rearward. The right
stretch thigh may be further extended with pressure from the left foot. Repeat on the opposite side. Do not pull the heel forcefully to the buttock, espe-
cially if there is discomfort in the knee joint. In this case, achieve a beneficial stretch by allowing the knee to straighten slightly and pull the
thigh further to the rear.
Posterior hip From the supine position, cross the right ankle over the left thigh. Grasp the right knee with both hands and pull it towards the left shoulder while
stretch raising the left knee toward the chest Repeat on the opposite side.
Trunk flexion/ex- Part 1: Move to the quadruped position, then continue rearward to sit back onto the legs while keeping the arms extended to the front. The head
tension stretch remains passively flexed.
Part 2: Move to the prone position. Perform the prone press up from either the forearm or hands, depending on individual flexibility and comfort.
The thighs and pelvis rest on the ground. Relax the back and abdominal muscles while bearing the bodyweight through the straight arms. Toes
point to the rear.
EFFECT

* Subjects performed 1 repetition of each stretch to each side. Stretches were held 20–30 seconds unless otherwise indicated.
OF
WARM UP PARAMETERS 495
 496 MCMILLIAN ET AL.

ment was recorded to the nearest inch (based on preexisting floor markings) and converted to meters. The average of 2 trials was used for statisti-
steps 5 m to the left cone and touches that cone. The subject side steps 10 m to the far right cone and touches that cone, then side steps 5 m back
to the right leg, back to the left leg, back to the right leg and finally stopping with a 2-leg landing. The distance from the starting line to the back

the base of the T. The grader gives the signal to go and starts the stopwatch. The subject runs to the middle cone and touches the cone, then side
TABLE 4. Performance on each dependent variable based on

From a parallel stance behind the starting line, subjects maximally throw a 9-lb medicine ball using an underhanded toss. Subjects are encouraged
to use countermovement as long as the feet remain parallel on the ground until the ball is released. The feet may leave the ground and cross the
to the nearest inch (based on preexisting floor markings) and converted to meters. The average of 2 trials was used for statistical analysis. Trials

passes the base of the T. Measurement was recorded to 0.01 of a second. The average of 2 trials was used for statistical analysis. Trials in which

starting line as the ball is released. The grader will measure the distance from the starting line to the point where the ball first lands. Measure-
From a parallel stance behind the starting line, subjects maximally jump from both legs to land on the left leg. Without stopping, maximally jump

of the most rearward heel is recorded. The jump must be repeated if the subject falls backward on the final landing. Measurement was recorded

Three cones are set 5 m apart on a straight line. A fourth cone is placed 5 m from the middle cone to form a ‘‘T.’’ The subject starts at the cone at
warm up conditions (N ⫽ 30). Data are mean ⫾ SD.*

to the middle cone and touches that cone. The subject runs 5 m backwards past the base of the T. The grader stops the watch when the subject
Medicine ball
throw for 5-step jump
T-drill (s) distance (m) (m)
Control (NWU) 9.77 ⫾ 0.82 9.47 ⫾ 2.89 9.51 ⫾ 1.14‡
SWU 9.69 ⫾ 0.85 9.34 ⫾ 2.87 9.78 ⫾ 1.172‡
DWU 9.56 ⫾ 0.79† 9.79 ⫾ 3.01† 10.06 ⫾ 1.23‡
* NWU ⫽ no warm up; SWU ⫽ static-stretching warm up;
DWU ⫽ dynamic warm up.
† Denotes significant difference from the other 2 warm up con-
ditions ( p ⬍ 0.01).
‡ Denotes significant difference between all 3 warm up con-
ditions ( p ⬍ 0.01).

used to evaluate the effect of warm up conditions on the

3 performance measures. Tukey’s honestly significant dif-
ference (HSD) was used for post hoc analysis. Statistical
significance was set at p ⱕ 0.05.
cal analysis. Trials in which either graders or subjects noted execution errors were not recorded.

RESULTS
Descriptive statistics representing the performance on
each dependent variable based on warm up conditions are
presented in Table 4. Repeated measures ANOVA re-
vealed neither a significant main effect nor interaction for
Description

gender; therefore, data were collapsed for post hoc test-

in which either graders or subjects noted execution errors were not recorded.

ing. The main effect for warm up protocol was significant.

Pair-wise comparisons using Tukey’s HSD revealed that
subjects scored better after the DWU than after the NWU
either graders or subjects noted execution errors were not recorded.

or SWU on all 3 performance tests (p ⬍ 0.01). There were

no significant differences between the SWU and NWU for
the medicine ball throw and the T-drill; however, subjects
scored better after the SWU than after the NWU on the
5-step jump (p ⬍ 0.01).
DISCUSSION
The purpose of this study was to compare the effects of
* The order of testing was counterbalanced over the 3 days of testing.

DWU, SWU, and NWU on selected measures of power

and agility. Results indicate that the DWU conferred a
modest performance enhancement for all 3 measures of
power and agility relative to the SWU and NWU. These
results are consistent with Bishop’s review of the litera-
ture, indicating that an active warm up of moderate in-
TABLE 3. Performance tests and their description.*

tensity is likely to significantly improve short-term per-

formance on a range of tasks as long as fatigue is not
induced (4). Although static-stretching warm up exercises
have been shown to decrease power and one-repetition
maximum strength tasks (21, 37), our results show no
significant difference between the SWU and NWU for the
T-drill and the medicine ball throw for distance. The
SWU was a significant improvement over NWU for the
5-step jump.
In a review of the warm up literature, Bishop cites
several reasons why an active warm up such as the DWU
used in this study might improve short-term performance
(4). Most factors are related to temperature and include
decreased stiffness of the muscles and joints; increased
Medicine ball un-
derhand throw

transmission rate of nerve impulses; changes in the force-

for distance

velocity relationship; and increased glycogenolysis, gly-

5-step jump
Test

colysis, and high-energy phosphate degradation. In ad-

dition to these temperature-related changes, 2 neuromus-
T-drill

cular phenomena possibly activated by the DWU could

potentially enhance power and agility performance. Post-
activation potentiation (PAP; an increase in muscle
 EFFECT OF WARM UP PARAMETERS 497

twitch force and rate of force development following a con- cific nature of flexibility, suggesting that flexibility train-
ditioning contractile activity) could theoretically improve ing might enhance performance in sports that rely on ex-
power and agility performance, though the optimal pa- tremes of motion for movement. Conversely, decreased
rameters to exploit PAP are unknown (29). Similarly, po- flexibility might actually increase economy of movement
stcontraction sensory discharge (increased neural activity in sports such as distance running, where only the mid-
measured in the dorsal roots following contraction) might portion of the range of motion is used (14). The evidence
enable a more rapid and forceful response to perturba- suggests that flexibility training should be applied, based
tions of muscle length (8). Active warm up also may de- on individual needs and the physical demands of the ac-
crease muscle stiffness by breaking the stable bonds be- tivity.
tween actin and myosin filaments, though stretching like- Although the current investigation examined only the
ly has the same effect (4, 34). effect of warm up parameters on performance, injury pre-
Although none of the physiological factors mentioned vention is cited routinely as a reason for pre-exercise
above were measured directly, we believe that the de- warm up. As reported by Shrier (30), the recent epide-
mands of the DWU used in this study are generally con- miological evidence suggests typical pre-exercise muscle
sistent with the recommendations of Bishop (4). For en- stretching protocols do not produce meaningful reduc-
hancement of short-term performance (10 seconds or tions in risk of exercise-related injury. Conversely, basic
less), evidence suggests a warm up of 5–10 minutes, per- science supports the notion that an active warm up might
formed at 40–60% of V̇O2max, followed by 5 minutes of protect against muscle strain injury, though clinical re-
recovery (4). Although the recovery interval used in the search is equivocal on this point (14). Theoretically, warm
present study was less than Bishop’s recommendation, fa- up activities that enhance neural activation will better
tigue did not appear to be significant in our athletic sub- prepare muscles to absorb loads that might otherwise be
jects. transmitted to other structures such as ligaments, ten-
In contrast to the benefits of an active warm up men- dons, and the muscle cytoskeleton. This concept is sup-
tioned above, there are at least 2 theories why pre-exer- ported by research showing that muscles under active
cise stretching might decrease subsequent performance contraction absorb significantly more energy than mus-
relative to a more dynamic warm up. First, several re- cles at rest (13). Recently, Olsen et al. were the first to
searchers have cited reduced neural activation as a use a large, randomized, controlled study to show reduced
means by which repeated stretches reduce the number of rates of injury in a group performing a dynamic, func-
motor units available for contraction (3, 11, 21). If the tional warm up (24).
SWU reduced neural activation relative to the DWU, per- The following factors should be considered when in-
formance of power and agility tasks, such as those used terpreting the results of the present investigation. First,
in this study, might be diminished. Because neural acti- due to study design and restriction on the availability of
vation was not measured, its effect on the performance subjects, only 3 repeated measures (one each following
measures used in this study is purely speculative. DWU, SWU, and NWU) were conducted. Therefore, the
In addition, other investigators have suggested that combined effect of dynamic and static stretching warm up
increased compliance (i.e., the length change that occurs components was not tested. Few studies have examined
when a force is applied) in the tendon results in a brief the effect of pre-exercise stretching combined with a dy-
moment when muscle force is taking up slack within the namic component. Church et al. compared the effect of a
tendon, rather than contributing to gross movement (14, general warm up consisting of a 10-minute circuit of body-
21). Potentially, such an effect could hinder power and weight exercises with the same warm up paired with ei-
performance. However, some studies have shown in- ther static-stretching or PNF stretches (6). Vertical jump
creased joint range of motion without changes in the com- performance was limited only by the PNF stretch warm
pliance of the musculotendinous unit (16, 22), suggesting up. The investigators theorized that the increased inten-
that greater stretch tolerance might account for the in- sity of the PNF stretching might induce autogenic inhi-
creased range of motion. Taylor et al. have shown that bition and, therefore, might limit vertical jump perfor-
stretching and isometric contractions both result in sub- mance.
sequent relaxation of the muscle–tendon unit (34). This Rosenbaum et al. found that decreased force and rate
concept is supported by a clinical study in which 3 differ- of force development related to stretching was returned
ent warm up conditions (i.e., body-weight circuit exercis- to normal after 10 minutes of running (28). This suggests
es, static stretching, and proprioceptive neuromuscular that pre-exercise stretching may not hinder power per-
facilitation stretching [PNF]) each resulted in equivalent formance if followed by dynamic movements that mimic
increases in hamstring flexibility (6). This suggests that the tasks that follow. Warm up protocols that combine
a dynamic warm up might increase flexibility from the dynamic and static-stretching exercises would add com-
resting state without the potential compromise of neural parative value and are encouraged for future investiga-
activation associated with an isolated, static-stretching tions. Still, for teams and individuals that are under time
warm up. constraints for warm up, the current body of evidence
It is important to distinguish between pre-exercise suggests that static stretching might be unnecessary.
stretching and flexibility training in general. The perfor- Another limiting factor of this study is that physiolog-
mance-related issues from pre-exercise stretching men- ical parameters of the warm up protocols were not estab-
tioned above, especially reduced neural activation, might lished. Controlling for factors such as muscle temperature
not apply to stretching exercises performed at other and oxygen utilization would have allowed for greater
times. In fact, investigations have noted improved perfor- precision when describing warm up parameters. Caution
mance correlated to regular stretching (31) and increased should be used when generalizing the results of this study
flexibility (36). Gleim et al., in a review of the literature to other populations. Our subjects were young athletes
on flexibility and sports performance, noted the sport-spe- accustomed to vigorous athletic and military training;
498 MCMILLIAN ET AL.

older or less-athletic populations might respond differ- 14. GLEIM, G.W., AND M.P. MCHUGH. Flexibility and its effects on
ently to the warm up protocols used in this study. sports injury and performance. Sports Med. 24:289–299. 1997.
Though evidence from previous investigations allows 15. GRAY, S.C., G. DEVITO, AND M.A. NIMMO. Effect of active warm
us to make general recommendations for the specificity, up on metabolism prior to and during intense active exercise.
Med. Sci. Sports Exerc. 34:2091–2096. 2002.
duration, intensity, and recovery interval of the warm up
16. HALBERTSMA, J.P., A.I. VAN BOLHUIS, AND L.N. GOEKEN. Sport
(4), questions remain as to the optimal parameters for
stretching: Effect on passive muscle stiffness of short ham-
these factors. Future clinical research should continue to strings. Arch. Phys. Med. Rehabil. 77:688–692. 1996.
investigate not only the optimal warm up parameters for 17. HERBERT, R.D., AND M. GABRIEL. Effects of stretching before
duration, intensity, and recovery interval, but also the and after exercising on muscle soreness and risk of injury: Sys-
interplay of dynamic and static stretching components, tematic review. BMJ. 325:468. 2002.
sports specificity, environmental conditions, and psycho- 18. HORITA, T., P.V. KOMI, C. NICOL, AND H. KYROLAINEN. Stretch
logical factors. In addition, more investigations are need- shortening cycle fatigue: Interactions among joint stiffness, re-
ed to establish the optimal warm up conditions for injury flex, and muscle mechanical performance in the drop jump
control. [published correction appears in Eur. J. Appl. Physiol. 74:575.
1996.]. Eur. J. Appl. Physiol. Occup. Physiol. 73:393–403. 1996.
PRACTICAL APPLICATIONS 19. KNAPIK, J.J., K.G. HAURET, S. ARNOLD, M. CANHAM-CHERVAK,
A. J. MANSFIELD, E.L. HOEDEBECKE, AND D. MCMILLIAN. In-
For tasks requiring power and agility, the results suggest jury and fitness outcomes during implementation of physical
that a dynamic warm up might offer performance benefits readiness training. Int. J. Sports Med. 24:372–381. 2003.
not found with static stretching or no warm up. It is likely 20. KNUDSON, D., K. BENNETT, R. CORN, D. LEICK, AND C. SMITH.
that a DWU similar to that used in this study will achieve Acute effects of stretching are not evident in the kinematics of
general warm up goals without invoking the mechanical the vertical jump. J. Strength Cond. Res. 15:98–101. 2001.
and neural activation drawbacks associated with acute, 21. KOKKONEN, J., A.G. NELSON, AND A. CORNWELL. Acute muscle
static-stretching. For tasks demanding a high degree of stretching inhibits maximal strength performance. Res. Q. Ex-
flexibility, power, and agility, warm up activities should erc. Sport. 69:411–415. 1998.
be sequenced so that static-stretching (if it is deemed nec- 22. MAGNUSSON, S.P., P. AAGARD, E. SIMONSEN, AND F. BOJSEN-
MOLLER. A biomechanical evaluation of cyclic and static stretch
essary) is followed by dynamic, progressive movements
in human skeletal muscle. Int. J. Sports Med. 19:310–316.
that mimic the goal activity without inducing fatigue. 1998.
23. NELSON, A.G., J. KOKKONEN, AND D.A. ARNALL. Acute muscle
REFERENCES stretching inhibits muscle strength endurance performance. J.
1. BARTLETT, M.J., AND P.J. WARREN. Effect of warming up on Strength Cond. Res. 19:338–343. 2005.
knee proprioception before sporting activity. Br. J. Sports Med. 24. OLSEN, O.E., G. MYKLEBUST, L. ENGEBRETSEN, I. HOLME, AND
36:132–134. 2002. R. BAHR. Exercises to prevent lower limb injuries in youth
2. BEHM, D.G., A. BAMBURY, F. CAHILL, AND K. POWER. Effect of sports: Cluster randomized controlled trial. BMJ. 330:449.
acute static stretching on force, balance, reaction time, and 2005.
movement time. Med. Sci. Sports Exerc. 36:1397–1402. 2004. 25. PAUOLE, K., K. MADOLE, J. GARHAMMER, M. LACOURSE, AND R.
3. BEHM, D.G., D.C. BUTTON, AND J.C. BUTT. Factors affecting ROZENEK. Reliability and validity of the T-test as a measure of
force loss with prolonged stretching. Can. J. Appl. Physiol. 26: agility, leg power, and leg speed in college-aged men and wom-
261–272. 2001. en. J. Strength Cond. Res. 14:443–450. 2000.
4. BISHOP, D. Warm up II: Performance changes following active 26. POPE, R.P., R.D. HERBERT, J.D. KIRWAN, AND B.J. GRAHAM. A
warm up and how to structure the warm up. Sports Med. 33: randomized trial of pre-exercise stretching for prevention of
483–498. 2003. lower-limb injury. Med. Sci. Sports Exerc. 32:271–277. 2000.
5. BLACK, J., AND S. JONES. Passive stretching does not protect 27. POWER, K., D. BEHM, F. CAHILL, M. CARROLL, AND W. YOUNG.
against acute contraction-induced injury in mouse EDL mus- An acute bout of static stretching: Effects on force and jumping
cle. J. Muscle Res. Cell Motil. 22:301–310. 2001. performance. Med. Sci. Sports Exerc. 36:1389–1396. 2004.
6. CHURCH, J.B., M.S. WIGGINS, F.M. MOODE, AND R. CRIST. Ef- 28. ROSENBAUM, D., AND E.M. HENNIG. The influence of stretching
fect of warm up and flexibility treatments on vertical jump per- and warm up exercises on Achilles tendon reflex activity. J.
formance. J. Strength Cond. Res. 15:332–336. 2001. Sports Sci. 13:481–490. 1995.
7. CHWALBINSKA-MONETA, J., AND O. HANNINEN. Effect of active 29. SALE, D.G. Postactivation potentiation: Role in human perfor-
warming-up on thermoregulatory, circulatory, and metabolic mance. Exerc. Sport Sci. Rev. 30:138–143. 2002.
responses to incremental exercise in endurance-trained ath-
30. SHRIER, I. Stretching before exercise does not reduce the risk
letes. Int. J. Sports Med. 10:25–29. 1989.
of local muscle injury: A critical review of the clinical and basic
8. ENOKA, R.M. Neuromechanics of Human Movement. Cham-
science literature. Clin. J. Sport Med. 9:221–227. 1999.
paign, IL: Human Kinetics, 2002.
31. SHRIER, I. Does stretching improve performance? A systematic
9. FITZGERALD, G.K., S.M. LEPHART, J.H. HWANG, AND R.S. WAIN-
and critical review of the literature. Clin. J. Sport Med. 14:267–
NER. Hop tests as predictors of dynamic knee stability. J. Or-
273. 2004.
thop. Sports Phys. Ther. 31:588–597. 2001.
10. FLETCHER, I.M., AND B. JONES. The effect of different warm up 32. SMITH, C.A. The warm up procedure: To stretch or not to
stretch protocols on 20 meter sprint performance in trained stretch. A brief review. J. Orthop. Sports Phys. Ther. 19:12–17.
rugby union players. J. Strength Cond. Res. 18:885–888. 2004. 1994.
11. FOWLES, J.R., D.G. SALE, AND J.D. MACDOUGALL. Reduced 33. STOCKBRUGGER, B.A., AND R.G. HAENNEL. Validity and reli-
strength after passive stretch of the human plantar flexors. J. ability of a medicine ball explosive power test. J. Strength
Appl. Physiol. 89:1179–1188. 2000. Cond. Res. 15:431–438. 2001.
12. GAMBETTA, V. Building the Complete Athlete. Sarasota, FL: Op- 34. TAYLOR, D.C., D.E. BROOKS, AND J.B. RYAN. Viscoelastic char-
timum Sports Training Inc., 1997. acteristics of muscle: Passive stretching versus muscular con-
13. GARRETT, W.E. JR, M.R. SAFRAN, A.V. SEABER, R.R. GLISSON, tractions. Med. Sci. Sports Exerc. 29:1619–1624. 1997.
AND B.M. RIBBECK. Biomechanical comparison of stimulated 35. WIKLANDER, J., AND J. LYSHOLM. Simple tests for surveying
and nonstimulated skeletal muscle pulled to failure. Am. J. muscle strength and muscle stiffness in sportsmen. Int. J.
Sports Med. 15:448–454. 1987. Sports Med. 8:50–54. 1987.
 EFFECT OF WARM UP PARAMETERS 499

36. WILSON, G.J., B.C. ELLIOTT, AND G.A. WOOD. Stretch-shorten collection; Lieutenant Colonel Paul Stoneman for review of the
cycle performance enhancement through flexibility training. manuscript; and reference librarian Joan Stehn.
Med. Sci. Sports Exerc. 24:116–123. 1992. This study was conducted at the United States Military
37. YOUNG, W.B., AND D.G. BEHM. Effects of running, static Academy, West Point, by the U.S. Army-Baylor University Post-
stretching and practice jumps on explosive force production Professional Sports Medicine-Physical Therapy Doctoral
and jumping performance. J. Sports Med. Phys. Fitness. 43:21– Program.
27. 2003. Disclaimer: The opinions and assertions contained herein are
the private views of the authors and are not to be construed as
Acknowledgments official or as reflecting the views of the Department of the Army
or Department of Defense.
The authors would like to thank Lieutenant Commander Mike
Rosenthal, Captain Marc Weishaar, Captain Erin Slivka, and Address correspondence to: Danny J. McMillian,
Staff Sergeant William Woodside for their assistance with data danny.mcmillian@us.army.mil.