Stretching, performance and injury prevention

Pre-exercise stretching – is it time to tear up the old rule book?

Pre-exercise stretching – is it time to tear up the old rule book?

In accordance with traditional guidelines(1), many individuals stretch before exercise(2,3). But, there’s increasing evidence that pre-exercise stretching fails to prevent injury and is detrimental to sports performance. Gary O’Donovan reviews the literature.

 What are the current guidelines on stretching before exercise?

The American College of Sports Medicine (ACSM) has traditionally recommended that exercise sessions contain a warm-up of around ten minutes’ duration, during which light activities precede static stretches of the major muscle groups(4,5). However, the ACSM now recognises that there is increasing evidence that stretching is detrimental to performance. Indeed, its latest guidelines suggest that the pre-exercise warm-up need only consist of cardiovascular exercise(6).

Does pre-exercise stretching reduce the risk of injury?

In order to investigate cause-and-effect relationships, scientists have conducted randomised, controlled trials. In one such study, researchers investigated the effect of pre-exercise stretching on the risk of exercise-related injury in 1,538 male army recruits(7).

Nineteen platoons were randomly allocated to the stretch group and 20 platoons to the control group. Throughout the 12 weeks of training, both groups warmed up for approximately five minutes before marching, running, swimming or wrestling. The stretch group also followed a typical stretching routine during every warm-up (one 20-second stretch of the gastrocnemius, soleus, hamstring, quadriceps, hip adductor and hip flexor muscle groups). The control group did not stretch before exercise. The lower-limb injury rate among 735 men who stretched (158 injuries, or 21.5%) was not significantly different to that of the control group (175 injuries, or 21.8%). In an earlier study of 1,093 servicemen, the same group also found that pre-exercise stretching did not protect from injury(8).

In another randomised, controlled trial, Dutch scientists found that warming up and stretching did not reduce the risk of injury in 421 recreational runners(9). During the 16-week study, there were 5.5 injuries per 1,000 hours of running in those who stretched before exercise, and 4.9 injuries per 1,000 hours of running in those who did not stretch before exercise.

Bucking the trend, a study carried out three years ago reported that injury rates were lower among 518 army recruits who performed static stretching before exercise than among 383 army recruits who did not perform static stretching before exercise(10). However, a closer look at this study reveals that the so-called ‘control group’ performed five to ten minutes of ballistic stretching before each physical training session.

 The results of another randomised controlled trial must also be interpreted with caution. In this study, researchers reported 75% fewer injuries among soccer players who took part in an injury-prevention programme(11). However, it is impossible to distinguish the effect of pre-exercise stretching from the co-interventions, which included leg guards, ankle taping, corrective training and the support of doctors and physiotherapists.

Collectively, these randomised, controlled trials suggest that pre-exercise stretching does not reduce the risk of injury.

It has been suggested that stretching might prevent injuries in sports involving jumping and bouncing, such as soccer and basketball(12). This would seem to be the case if ‘stretchy’ muscle were better able to absorb energy. However, it has in fact shown that less force is required to rupture ‘stretchy’ muscle than a ‘stiff’ muscle(13).

Does pre-exercise stretching improve performance?

To investigate the effect of static stretching on subsequent leg strength, a team of researchers from the University of Hawaii tested the one-repetition maximum (1RM) lift of 30 physical education students following stretching of the hip, thigh and calf muscles(14).

Each volunteer performed a 1RM prone (lying face-down) knee flexion and a 1RM seated knee extension on two successive days following either ten minutes of quiet rest or 20 minutes of stretching. The stretching protocol consisted of three 15-second conventional stretches of five muscle groups involved in knee flexion and knee extension followed by three 15-second assisted stretches of the same groups. Nautilus machines were used to measure knee-flexion strength and knee-extension strength. Each subject performed one progressively heavier lift every 60 seconds until failure.

In a later study, scientists at the University of Louisiana used an identical protocol to determine the effects of ballistic stretching on leg strength in 22 physical education students(15). This time, the stretching procedure included three 15-second unassisted ballistic stretches and three 15-second assisted ballistic stretches of five muscle groups involved in knee flexion and knee extension. Unassisted ballistic stretching required the subjects to bob up and down about once per second after feeling the stretch. During assisted ballistic stretching, an experimenter moved the participant’s joint back and forth through approximately two to five degrees after feeling the stretch. Ballistic stretching reduced knee-flexion 1RM by 7.5% and knee-extension 1RM by 5.6%.

In 2000, Canadian researchers published a study showing that calf muscle strength is reduced for an hour after static stretching(16). To measure maximal calf-muscle strength, 10 subjects were each secured in a leg-holding device designed to position the knee and hip angles at 90 degrees whilst a foot was strapped to a metal plate that rotated about the same axis of rotation as the individual’s ankle. Force applied to the footplate was measured by strain gauges. The footplate could also push against the foot in order to passively stretch the soleus muscle. The experimental procedure for each subject was:

1. Baseline maximum voluntary contraction;
2. Ten-minute rest;
3. Thirteen 135-second maximal passive stretches over 33 minutes or, on a separate occasion, non-stretch control;
4. Maximum voluntary contractions immediately after (post) and at 5-, 15-, 30-, 45- and 60-minute intervals.

In this case study, muscle strength was 28% lower immediately after stretching. Compared with the baseline value, maximum voluntary contractions (MVC) were also lower 5, 15, 30, 45 and 60 minutes after stretching. Another study, carried out in 2004, has shown that quadriceps strength is reduced for 120 minutes after static stretching(17).

After finding that quadriceps strength was reduced by 12% after static stretching, Canadian scientists concluded that ‘it would be difficult to imagine that overall athletic performance would be consistently enhanced if, following acute bouts of prolonged stretching, force output was diminished’(18). Indeed, the same group later demonstrated that running and practice jumps had a positive influence on jumping performance, whilst static stretching of the knee extensors had a negative influence on jumping performance(19).

More recently, the Louisiana group (see above) reported that stretching decreases subsequent muscle strength-endurance(20). On average, 22 physical education students were able to perform 14.5 knee-flexion lifts at 60% of body weight without prior stretching, and only 11 knee-flexion lifts at 60% of body weight after passive stretching of the hip, thigh and calf muscle groups. Successful strength trainingrequires progressive overload. For example, a lifter might add five kilos to his or her next training session once he or she has successfully pressed 100 kilos for four reps. These results suggest that stretching prior to lifting could induce a ‘false negative’ experience that would delay the progression to a higher load.

Stretching is not just detrimental to performance in the contrived setting of the laboratory; it’s also detrimental on the track. A study carried out last year found that 20-metre sprint times were significantly increased in collegiate athletes following four 30-second static stretches of the hamstring, quadriceps and calf muscles(21). Twenty-metre sprint times were also increased in rugby union players following 20-second static stretches of the gluteals, hamstrings, quadriceps, adductors, hip flexors, gastrocnemii, and solei(22).

Does flexibility training improve performance?

No large studies have investigated the effect of flexibility training on performance. However, cross-sectional data suggest that runners are less flexible than their sedentary counterparts, demonstrating tight hamstring and soleus muscles(23). Given the elastic properties of muscles and tendons, such tightness is likely to improve performance. Indeed, it has been shown that the least flexible runners are also the most economical(24).

Why is pre-exercise stretching detrimental to performance?
Two mechanisms may explain why pre-exercise stretching is detrimental to performance. Firstly, stretching damages the contractile proteins in skeletal muscle. Secondly, stretching reduces one’s ability to recruit skeletal muscle.

Skeletal muscle contains thick filaments and thin filaments that are connected by cross-bridges. When a nerve signal reaches the muscles, the thin filaments slide over the thick filaments. However, movement cannot occur if the cross-bridges between the filaments are broken. Indeed, animal studies have shown that force production is reduced when muscle filaments are stretched beyond overlap(25). Animal studies have also shown that cross-bridges are broken when muscle is stretched only 20% beyond its resting length(26). In humans, there is evidence of muscle damage 24 hours after a bout of stretching, which has led scientists to conclude that stretching causes delayed onset muscle soreness(27).

The nerve signals that initiate muscle contraction are electrical in nature. Thus, electrodes can be used to monitor muscle activity. In humans, such studies have shown that muscle activity and force production are reduced after stretching(16,18,28,29). These findings suggest that stretching produces some kind of neural inhibition that is detrimental to performance. This hypothesis is supported by a study showing that balance and reaction time are also impaired after static stretching(30).


What does this mean for sportsmen and women? These large, well-conducted, randomised, controlled trials provide compelling evidence that pre-exercise stretching does not reduce the risk of injury. The available evidence also suggests that pre-exercise stretching is detrimental to maximal strength, strength-endurance, jumping height and sprinting time. Given this evidence, it seems only reasonable to conclude that stretching should not be performed before exercise. Instead, warm-ups should consist of aerobic activities that are likely to prevent injury and improve performance(31,32).

Dr Gary O’Donovan is a research fellow at Brunel University and an exercise physiologist accredited with the British Association of Sport and Exercise Sciences

1. ACSM’s Resource Manual for Guidelines for Exercise Testing and Prescription, 4th ed
2. Am J Sports Med 1986; 14:151-5
3. J Am Podiatry Assoc 1983; 73:160-4
4. ACSM’s Guidelines for Exercise Testing and Prescription, 6th ed 2000
5. ACSM’s Resource Manual for Guidelines for Exercise Testing and Prescription, 4th ed
6. ACSM’s Resource Manual for Guidelines for Exercise Testing and Prescription, 5th ed 2006:362-364
7. Med Sci Sports Exerc 2000; 32:271-7
8. Aust J Physiother 1998; 44:165-172
9. Am J Sports Med 1993; 21:711-9
10. Mil Med 2003; 168:442-6
11. Am J Sports Med 1983; 11:116-20
12. Sports Med 2004; 34:443-9
13. Am J Sports Med 1993; 21:517-22
14. Res Q Exerc Sport 1998; 69:411-5
15. Res Q Exerc Sport 2001; 72:415-9
16. J Appl Physiol 2000; 89:1179-88
17. Med Sci Sports Exerc 2004; 36:1389-96
18. Can J Appl Physiol 2001; 26:261-72
19. J Sports Med Phys Fitness 2003; 43:21-7
20. Strength Cond Res 2005; 19:338-43
21. Sports Sci 2005; 23:449-54
22. J Strength Cond Res 2004; 18:885-8
23. J Orthop Sports Phys Ther 1993; 17:102-7
24. Int J Sports Med 2002; 23:40-3
25. J Muscle Res Cell Motil 1988; 9:491-8
26. Am J Physiol 1996; 271:C1438-46
27. Res Q Exerc Sport 1993; 64:103-7
28. Eur J Appl Physiol 2005; 93:530-9
29. J Strength Cond Res 2003; 17:484-8
30. Med Sci Sports Exerc 2004; 36:1397-402
31. Exercise & Sports Cardiology (2001) New York: McGraw-Hill
32. McArdle WD, Katch FI, Katch VL (1996) Exercise Physiology: Energy, Nutrition and Human Performance, Williams and Wilkins