Stretching has traditionally been a part of the preexercise and competition warm-up to increase joint range of motion (1), which potentially reduce injury risk (21,28) and improve performance (36). In recent years, however, the practice of static stretching before exercise has been questioned. A review conducted by Shrier (30) concluded that it is unlikely that preactivity static stretching prevents injury. There is also a great deal of evidence indicating that preexercise static stretching has a negative impact on strength and power performances (7,9,16,22,25,26,37).
Despite the vast amount of research that has examined the effects associated with static stretching of the agonist, we are unaware of the published research investigating the effects of stretching the antagonist musculature on subsequent expression of strength and power. Concurrent neural adaptations to both the agonist and antagonist muscles are important to facilitate greater torque and power output (5,17). The net external force applied during a movement is proportional to the force produced by the agonist and inversely proportional to the force produced by the antagonist muscles (2,14). Therefore, inhibiting the force produced by the antagonist muscles, via static stretching, may allow the agonist muscles to apply greater external force and power output.
Gains in strength might be accompanied with an increase in neural activity of the agonist and neurological inhibition of the antagonist (5,17). It has already been postulated that stretching the agonist before a given movement may decrease the agonist muscle strength and power perhaps, through decreased neural drive (9,16,33). Conversely, stretching the antagonist may result in its inhibition and reciprocally facilitate increased activity of the agonist, with subsequent improvements in strength and power–related performance. Therefore, the purpose of this study was to investigate the effects of static stretching of the antagonist muscles on peak torque of the knee extensors recorded at 60°.s−1 (SlowKE) and 300°.s−1 (FastKE), and vertical jump height (VJheight) and power (VJpower). This study also sought to determine whether antagonist stretching would affect the neural activity in the agonist and antagonist musculature.
Experimental Approach to the Problem
To determine if stretching the antagonist musculature affects performance, subjects were tested for peak knee extension torque at 60°.s−1 and 300°.s−1 and for VJheight. All tests were performed with and without preceding the antagonist stretching for each subject. The study used a within-group design, and the treatment was provided in a randomized order. Subjects underwent the SlowKE and FastKE on the same day with or without the stretching treatment; they received the opposite treatment 1–3 days later. The SlowKE and FastKE were performed in a randomized order on the first testing day. The opposite order was repeated with the opposite treatment on the second testing day. The VJheight was tested by itself on 2 separate days with or without stretching treatment with 1–3 days between each session.
Of the 18 people who volunteered for the study, 2 dropped out because of scheduling issues. The remaining 16 men (22.5 ± 4.9 years) finished all stretching and nonstretching trials for SlowKE, FastKE, and VJheight. The average height and mass of the study sample was 180.3 ± 10.1 cm and 84.9 ± 19.5 kg, respectively. Subjects had engaged in resistance training a minimum of 2 times a week, for the past 6 months. All subjects were free from musculoskeletal, cardiovascular, and metabolic disorders at the time of the study. Subjects gave written informed consent. The study was approved by the Institutional Review Board of Utah State University.
All stretches for the SlowKE, FastKE, and VJheight, were held for 30 seconds and repeated 3 times with 20-second rest between the stretches. Previous research has recommended holding static stretches for a duration of 30 seconds (1,6). A 90-second rest period was provided between stretching and knee extension and VJ tests.
Stretches before the knee extension tests emphasized the hamstring group. With a subject supine on a training table, the investigator stabilized the opposing limb and put one hand on the subject's heel and the other hand just above the knee. The investigator then pushed the subject's heel and took him into knee extension and hip flexion (see Figure 1).
The stretching treatment before the vertical jump emphasized the stretching of the hip flexors and dorsiflexors. To stretch the hip flexors, the subject was positioned in a half-kneel position. For comfort, a foam pad was placed under the knee of the kneeling limb. The subject was instructed to keep an erect torso. The hip that was posterior was then extended by contracting the gluteals. The subject was then instructed to internally rotate the leg or turn his foot out (see Figure 2). Internal rotation of the hip joint effectively stretches the hip flexors (iliopsoas group) because of their insertion point on the lesser trochanter of the femur. Because the knee was placed only in partial flexion, a greater emphasis of stretch was placed on the single-joint hip flexors (iliopsoas group) vs. the 2-joint hip flexors that also act at the knee joint. For example, the rectus femoris, which has a dual role in hip flexion and knee extension, was lengthened more at the origin (anterior inferior iliac spine and acetabulum) than at the insertion (tibial tuberosity via the quadriceps tendon).
To stretch the dorsiflexors, subjects began in a supine position on a training table with their feet hanging freely off the edge of the table. An investigator put the foot into plantar flexion by pulling on the toes and pushing on the heel (see Figure 3). The dorsiflexors were stretched to a point of mild discomfort. The dorsiflexors were stretched first followed by the hip flexors.
Knee extensors peak torque was measured on a Biodex isokinetic dynamometer (Biodex, Shirley, NY, USA). Calibration was performed before testing. Subjects were tested in a seated position with straps placed over their waist and distal thigh for stabilization. The tibial pad was placed and secured approximately 2 finger widths proximal to the medial malleolus on the dominant leg. The axis of rotation of the dynamometer was aligned with the medial epicondyle of the knee. Concentric peak torque of the knee extensors was recorded at 60°.s−1 and 300°.s−1. Each testing velocity was performed in a randomized order with 10-minute rest between maximal tests. For the stretching treatment, the stretching protocol was repeated before each maximal attempt at each testing velocity. Five maximal attempts were made, and the highest value was used for data analysis. A similar 5 repetition isokinetic protocol has been shown to be reliable (r = 0.95) for 60°.s−1 and 300°.s−1 (15).
Electromyography (EMG) was collected from the vastus lateralis and the long head of the biceps femoris muscles during knee extension tests using Biopac Systems instrumentation (MP 150; Goleta, CA, USA). Positioning and placement of the electrodes was determined using procedures described by Herda et al. (20). Before applying the EMG electrodes, the skin at the placement sight was shaved, rubbed with alcohol, and slightly abraded to ensure good surface contact and to reduce skin resistance. Bipolar surface electrodes (2.5 cm interelectrode distance) were placed at the approximate center of each muscle belly. A ground electrode was applied to the tibial tuberosity. For the biceps femoris, the electrodes were placed at 50% of the distance from the ischial tuberosity to the lateral epicondyle of the tibia. For the vastus lateralis, the electrodes were placed at 50% of the distance between the greater trochanter and the lateral epicondyle of the femur. The positions of the electrodes were marked with a small ink mark on the skin. The precise distance was also recorded and used for electrode placement in all conditions. The electrodes were placed before commencing the stretching treatments.
The raw EMG signals were preamplified 100 times at the electrode site, then further amplified for a total gain of 5,000 with a bandwidth of 10–500 Hz. A low-pass filter was used with a cutoff frequency of 250 Hz. A high-pass filter was used with a cutoff frequency of 25 Hz. The EMG signal was smoothed by integration. As recommended by Basmajian and DeLuca (3), the root mean square was calculated for the signal after it was filtered and smoothed.
Normalization of EMG signal characteristics is typically performed when comparisons are to be made, which involve reapplication of electrodes or between individuals (4,13). In this study, normalization was done by taking the EMG signal from each treatment and comparing it with a reference contraction of the same muscle with the same electrode placements. The reference was a maximal voluntary contraction (MVC) with the dynamometer set at 45° of knee flexion for terminal point. Subjects were tested in a seated position described previously for torque testing. Velocity was set at 30°.s−1 and subjects performed 3 maximal knee extensions with a 5-second isometric hold at the terminal range of motion. Subjects were instructed to exert maximal effort against the tibial pad. The MVC was performed, and then subjects rested 10 minutes before performing stretching and randomized isokinetic testing. The peak EMG voltage for the vastus lateralis and biceps femoris for SlowKE stretch, SlowKE nonstretch, FastKE stretch, and FastKE nonstretch was divided by the peak EMG voltage for the MVC. The EMG results are represented as a percentage of MVC. A summary of the time line for isokinetic knee extension and EMG testing is shown in Figure 4.
Vertical Jump Testing
The vertical jump test was performed using a Vertec device (Sports Imports, Columbus, OH, USA) using a protocol described by Harman et al. (18). The investigator adjusted a vertical column with vanes low enough that the subject could register a standing reach measurement. The subject then stood so that the dominant hand reached straight upward and directly below the center of the vanes. The highest vane that could be pushed forward while standing flat-footed determined the standing reach height. The same reach height was used for both trials. The vertical column was then raised to accommodate the jumping ability of the subject. The subject was allowed a countermovement with no approach step and then jumped to the highest vane possible. Subjects were allowed to jump until they were unable to touch a higher vane on 2 consecutive trials. Jump height was determined by subtracting the distance between the highest vane touched and the standing reach. Jump height and body mass were used to calculate absolute VJpower using the Harman equation (19). This test was chosen because it is a commonly used field test to measure power. The test has been found to have high reliability (r = 0.94) (38).
Statistical comparisons were made for the 2 conditions of the independent variable of the study (stretch and nonstretch). Differences of the mean values of the dependent variables for the 2 conditions were evaluated using paired t-tests. The dependent variables were SlowKE, FastKE, VJheight, VJpower, and normalized EMG activity for the vastus lateralis and the long head of the biceps femoris during knee extension tests. Significant differences were assumed with p ≤ 0.05. Effect sizes were also calculated using Cohen's d for each dependent variable. All analyses were executed using Statistical Package for Social Sciences (SPSS 18.0; IBM, Somers, NY, USA).
The results for knee extension torque are summarized in Table 1. Stretching the antagonist elicited significantly greater torque production for the FastKE but not the SlowKE. According to Cohen (8), the effect sizes for both trials were moderate.
Vertical Jump Height and Power
Both VJheight and VJpower were significantly higher after the stretching protocol (Table 1). The effect sizes were small (8) for both the variables.
Paired samples t-test indicated no significant (p > 0.05) differences between the trials for EMG, represented as a percentage of the MVC. The results for all conditions of the vastus lateralis are summarized in Figure 5. The results for all conditions of the biceps femoris are summarized in Figure 6.
There was considerable variability among subjects as several had dramatic improvements, particularly in SlowKE and FastKE, following the stretching treatment while others had minimal or even negative responses. The percent change from nonstretching to stretching trials for the 4 dependent variables for each subject is illustrated in Figure 7.
Despite a large amount of research on the effects of statically stretching the agonist musculature before rapid force production, this is the first published study to our knowledge that has examined the effects of statically stretching the antagonist musculature before a power movement. We hypothesized that a static stretch of the antagonists could inhibit this muscle group, allowing for greater expression of strength and power production of the agonists. Indeed, the primary finding of this study was that there were statistically significant increases in isokinetic knee extension torque at 300°.s−1 as well as VJheight and VJpower following static stretches of the antagonist musculature. However, these results are tempered by the fact that the magnitudes of the increases were small (8). Furthermore, there was no significant difference in EMG between stretching and nonstretching trials.
The increase in isokinetic knee extension torque occurred only at 300°.s−1 and not 60°.s−1, suggesting that the effects of antagonist stretching on torque production could be velocity specific. Nelson et al. (24) found an opposite and negative effect for agonist stretching on peak knee extension torque. These researchers reported that knee extension torque was reduced at the slowest velocities after static stretching of the quadriceps. In the current study, antagonist stretching exhibited the opposite effect vs. the agonist stretching in the study by Nelson et al. (24); in the current study, torque increased 9.3% at the fastest velocity after antagonist stretching. However, the majority of other studies that have examined torque at different speeds after agonist stretching have found no velocity-specific effect (10–12,23,32,39).
Decrements in torque production of 2–3% after static stretching of the agonists have been reported in several studies (10–12,23), whereas others have reported slightly larger drops in isokinetic peak torque of 5–12% (24,31,32,39). However, as Sobolewski et al. (32) observed, the magnitudes of these decrements were often within the standard error of measurement of isokinetic testing, and the effect sizes were small. Similarly, we observed an 8% increase for SlowKE and a 9% increase for FastKE after stretching of the antagonist, but the magnitudes of these increases exhibited small effect sizes (8). Thus, although there is a trend (and often a statistically significant difference) for static stretching to impact torque output, whether these differences between stretching and nonstretching conditions have practical meaning is not clear. Nevertheless, the magnitude of torque increase observed in the present study after static stretching of the antagonist was remarkably similar to the magnitude of decrease in torque output that has been reported after static stretching of the agonist.
Similar to the change in isokinetic peak torque after static stretching of the antagonist, there was a statistically significant increase in VJheight and VJpower after static stretching of the hip flexors and dorsiflexors, the antagonists of the hip extensors and plantar flexors, but the effect sizes associated with these changes were small. The magnitude of change in VJheight observed in the present study is similar to, but in the opposite direction of, that reported by other researchers who have examined the effects of agonist stretching on VJheight; we observed an increase in VJheight after static stretching of the antagonists, whereas other researchers observed a decrease in VJheight after static stretching of the agonists.
Church et al. (7) reported a decrease of 1.5 cm in VJheight following proprioceptive neuromuscular facilitation (PNF) stretching of the hamstrings and quadriceps. This is similar to the increase of 1.2 cm (2%) in VJheight seen in the current study. Robbins and Scheuermann (27) investigated the effects of 3 different volumes of static agonist stretching on vertical jump height. The greatest stretching volume of 6 sets of 15 seconds decreased VJheight by an average of 1.9 cm. Cornwell et al. (9) also found a significant decrease in VJ height after acute static stretching of the plantar flexors.
Two hypotheses have been proposed for agonist stretch–induced force deficits (9,16,20,33). One proposed mechanism for decreased strength and power after agonist preactivity static stretching involved mechanical adaptations, namely, reduction in stiffness and increase in length between resting sarcomeres that alters the length-tension relationship of the muscle. The second proposed mechanism involved neural factors such as decreased recruitment or reflex sensitivity or both.
It was hypothesized in the current study that stretching the antagonist musculature would result in an increased performance by increasing the neural drive to the agonist, decreasing neural drive to the antagonist, reducing stiffness of the antagonist and braking forces to the agonist, or a combination of these factors. Despite an increase in isokinetic peak torque for FastKE after hamstring stretching and a 9.7% increase in EMG activity of the vastus lateralis following the stretching trial compared with the nonstretching trial, the differences of EMG activity between test conditions were not statistically significant. Similarly, the EMG activity of the antagonist biceps femoris during FastKE was 16% lower after the stretching trial compared with the nonstretching trial, but again this difference was not statistically significant. The lack of a statistically significant finding for EMG activity suggests that the difference in torque observed was not related to increased activation of the prime movers.
Herda et al. (20) found no change in EMG after agonist static stretching. They hypothesized that decrements in force following stretching were related to mechanical factors. Cornwell et al. (9) found a decrease in EMG activity and stiffness after agonist static stretching. However, these authors hypothesized that reductions in stiffness were insufficient to cause a decrease in force production. Fowles et al. (16) found that EMG activity was significantly decreased for the first 15 minutes following static stretching, and force decrements were greatest during this time. However, electrical activity did return to normal after 15 minutes while force decrements remained for 60 minutes. These authors theorized that neural factors played a bigger role in strength decreases early, but as time passed, the reduction in maximum voluntary contractions originated peripherally in the muscle. It is possible that the improvement in knee extension torque after antagonist stretching at 300°.s−1 in a mechanically mediated response. If the length-tension relationship of the hamstrings was disrupted, this could lead to a reduction in braking forces, which would allow the quadriceps to produce more torque. This is speculation because no measure of mechanical adaptation was taken.
With each measured variable, there was a large amount of subject-to-subject variability, as shown by the relatively large SDs (Table 1). The differences of interindividual strength and power responses to antagonist stretching may have been due to initial levels of flexibility. One of the limitations of this study is that there was no initial flexibility assessment taken. There is evidence that tight or short antagonist musculature may result in decreased function of the agonist musculature (29,34,35). It is possible that individuals with lower initial levels of flexibility in antagonist musculature experienced a greater training effect with stretching than those with higher initial levels. The researchers are not aware of any studies that have investigated whether initial levels of flexibility affect the magnitude of treatment effect from stretching, and this is a potential avenue of investigation.
Antagonist stretching of the hamstrings resulted in significantly greater torque during a high-velocity (300°.s−1) isokinetic knee extension. The take home message for the practitioner is that stretching the antagonist to the hip extensors (hip flexors) and plantarflexors (dorsiflexors) before jumping resulted in significantly greater VJheight and VJpower. These findings suggest that stretching the antagonist musculature immediately before a high-velocity activity may enhance the performance of that activity. Practitioners could apply the results of the current study in designing the warm-up procedures before plyometric training sessions or other training sessions that require high-velocity movements.
However, the findings of the current study are mitigated by the fact that the magnitude of improvement was small, and there was substantial interindividual variability in the response to antagonistic stretching. Furthermore, the difference in EMG activity after antagonistic stretching compared with a nonstretching trial was not significant, leaving the mechanism of improvement in doubt. Nevertheless, there is justification for practitioners to experiment with stretching the antagonist musculature to improve performance in high-velocity activity.
For the researcher, there is ample opportunity for further investigation on this topic. Future research should investigate other muscle groups and movement patterns. It should also be determined if initial levels of flexibility affect responses to antagonist stretching before strength and power–related performance. The effects of antagonist stretching using other stretching techniques (e.g., PNF, dynamic) should be investigated, as well as gender effects to antagonist stretching. Future research should also attempt to determine possible mechanisms, whether mechanical, neural, or a combination of both.
The authors thank the subjects for their effort and Dr. Dennis Dolny for his expertise and guidance with the Biopac EMG system. This study was conducted without external funding but was supported by the Human Movement Science Program at Utah State University.
1. Bandy WD, Irion JM, Briggler M. The effect of time and frequency of static stretching on flexibility of the hamstrings muscles. Phys Ther 77: 1090–1096, 1997.
2. Baratta R, Solomomow M, Zhou B, Letson D, Chuinard R, D'Abrosia R. Muscular co-contraction. The role of antagonist musculature in maintaining knee stability. Am J Sports Med 16: 113–122, 1988.
3. Basmajian JV, DeLuca CJ. Muscles Alive: Their Function Revealed by Electromyography. Baltimore, MD: Willams & Wilkins, 1985.
4. Burden A. How should we normalize electromyograms obtained from healthy participants? What we have learned from over 25 years of research. J Electromyogr Kinesiol 20: 1023–1035, 2010.
5. Carolan B, Cafarelli E. Adaptations in coactivation after isometric training. J Appl Physiol 73: 911–917, 1992.
6. Chan SP, Hong Y, Robinson PD. Flexibility and passive resistance of the hamstrings of young adults using two different static stretching protocols. Scand J Med Sci Sports 11: 81–86, 2001.
7. Church BJ, Wiggins MS, Moode M, Crist R. Effect of warm-up and flexibility treatments on vertical jump performance. J Strength Cond Res 15: 332–336, 2001.
8. Cohen J. Statistical Power Analysis for Behavioral Sciences. Hillsdale, NJ: Erlbaum Associates, 1988.
9. Cornwell A, Nelson G, Sidaway B. Acute effects of stretching on the neuromechanical properties of the triceps surae muscle complex. Eur J Appl Physiol 86: 428–434, 2002.
10. Cramer JT, Beck TW, Housh TJ, Massey LL, Marek SM, Danglemeier S, Purkayastha S, Culbertson JY, Fitz KA, Egan AD. Acute effects of static stretching on characteristics of the isokinetic
angle-torque relationship, surface electromyography, and mechanomyography. J Sports Sci 25: 687–698, 2007.
11. Cramer JT, Housh TJ, Johnson GO, Miller JM, Coburn JW, Beck TW. Acute effects of static stretching on peak torque in women. J Strength Cond Res 18: 236–241, 2004.
12. Cramer JT, Housh TJ, Weir JP, Johnson GO, Coburn JW, Beck TW. The acute effects of static stretching on peak torque, mean power output, electromyography, and mechanomygraphy. Eur J Appl Physiol 93: 530–539, 2005.
13. DeLuca C. The use of surface electromyography in biomechanics. J Appl Biomech 13: 153–163, 1997.
14. Draganich LF, Jaeger RJ, Kralj AR. Coactivation of the hamstrings and the quadriceps during extension of the knee. J Bone Joint Surg 71: 1078–1081, 1989.
15. Feiring DC, Ellenbecker TS, Derscheid GL. Test-retest reliability of the biodex isokinetic
dynamometer. J Orthop Sports Phys Ther 11: 298–300, 1990.
16. Fowles JR, Sale DG, MacDougall JD. Reduced strength after passive stretch of the human plantarflexors. J Appl Physiol 89: 1179–1188, 2000.
17. Häkkinen K, Kallinen M, Izquierdo M, Jokelainen K, Lassila H, Malkia E, Alen M. Change in agonist-antagonist EMG, muscle CSA, and force during strength training in middle aged and older people. J Appl Physiol 84: 1341–1349, 1998.
18. Harman E, Garhammer J, Pandorf C. Administration, scoring, and interpretation of selected tests. In: Essentials of Strength Training and Conditioning (2nd ed.). Baechle TR, Earle RW, eds. Champaign IL: Human Kinetics, 2000. pp. 293.
19. Harman EA, Rosenstein MT, Frykman PN, Rosenstein RM, Kraemer WJ. Estimation of human power output from vertical jump. J Appl Sport Sci 5: 116–120, 1991.
20. Herda TJ, Cramer JT, Ryan ED, McHugh MP, Stout JR. Acute effects of static versus dynamic stretching on isometric peak torque, electromyography, and mechanomyography of the biceps femoris muscle. J Strength Cond Res 22: 809–817, 2008.
21. Johagen SG, Nemeth G, Eriksson E. Hamstring injuries in sprinters: The role of concentric and eccentric strength and flexibility. Am J Sports Med 22: 262–266, 1994.
22. Kokkonen J, Nelson AG, Cornwell A. Acute muscle stretching inhibits maximal strength performance. Res Q Exerc Sport 69: 411–415, 1998.
23. Marek SM, Cramer JT, Fincher AL, Massey LL, Dangelmaier SM, Purkayastha S, Fitz KA, Culbertson JY. Acute effects of static and proprioceptive neuromuscular facilitation stretching on muscle strength and power output. J Athletic Train 40: 94–103, 2005.
24. Nelson AG, Guillory IK, Cornwell A, Kokkonen J. Inhibition of maximal voluntary isokinetic
torque production following stretching is velocity-specific. J Strength Cond Res 15: 241–246, 2001.
25. Nelson AG, Kokkonen J. Acute ballistic muscle stretching inhibits maximal strength performance. Res Q Exerc Sport 72: 415–419, 2001.
26. Power K, Behm D, Cahill F, Carroll M, Young W. An acute bout of static stretching: effects on force and jumping performance. Med Sci Sports Exerc 36: 1389–1396, 2004.
27. Robbins JW, Scheuermann BR. Varying amounts of acute static stretching and its effect on vertical jump performance. J Strength Cond Res 22: 781–786, 2008.
28. Safran MR, Seaber AV, Garrett WE. Warm-up and muscular injury prevention: An update. Sports Med 8: 239–249, 1989.
29. Sahrmann SA. Diagnosis and Treatment of Movement Impairment Syndromes. St Louis, MI: Mosby, 2002.
30. Shrier I. Stretching before exercise does not reduce the risk of injury: A critical review of the clinical and basic scientific literature. Clin J Sports Med 9: 221–227, 1999.
31. Siatras TA, Mittas VP, Mameletzi DN, Eystroatios AV. The duration of the inhibitory effects with static stretching on quadriceps peak torque production. J Strength Cond Res 22: 40–46, 2008.
32. Sobolewski EJ, Wagner DR, Bressel E. Effect of static stretching and jogging on knee extension isokinetic
peak torque. Isokinet Exerc Sci 19: 157–162, 2011.
33. Vujnovich AL, Dawson NJ. The effect of therapeutic muscle stretch on neural processing. J Orthop Sports Phys Ther 20: 145–153, 1994.
34. White SG, Sahrmann SA. A movement system balance approach to management of musculoskeletal pain. In: Physical Therapy of the Cervical and Thoracic Spine. Grant R, ed. New York, NY: Churchill Livingstone, 1994. pp. 339–357.
35. Winters MV, Blake CG, Trost JS, Marcello-Brinker TB, Lowe L, Garber MB, Wainner RS. Passive versus active stretching of the hip flexor muscles in subjects with limited hip extension: A randomized clinical trial. Phys Ther 84: 800–807, 2004.
36. Wordell TW, Smith TL, Winegardner J. Effect of hamstring stretching on hamstring performance. J Sport Phys Ther 20: 154–159, 1994.
37. Young W, Elliot S. Acute effects of static stretching, proprioceptive neuromuscular facilitation stretching, and maximum voluntary contractions on explosive force production and jumping performance. Res Q Exerc Sport 72: 273–279, 2001.
38. Young W, MacDonald CH, Heggen T, Fitzpatrick J. An evaluation of the specificity, validity and reliability of jumping tests. J Sports Med Phys Fitness 37: 240–245, 1997.
39. Zakas A, Doganis G, Papakonstandinou V, Sentelidis T, Vamvakoudis E. Acute effects of static stretching duration on isokinetic
peak torque production of soccer players. J Bodywork Mov Ther 10: 89–95, 2006.