Secondary Logo

Share this article on:

Assessing Lower-Body Peak Power in Elite Rugby-Union Players

Argus, Christos K; Gill, Nicholas D; Keogh, Justin WL; Hopkins, Will G

Journal of Strength and Conditioning Research: June 2011 - Volume 25 - Issue 6 - p 1616-1621
doi: 10.1519/JSC.0b013e3181ddfabc
Original Research

Argus, CK, Gill, ND, Keogh, JWL, and Hopkins, WG. Assessing lower-body peak power in elite rugby-union players. J Strength Cond Res 25(6): 1616-1621, 2011—Resistance training at the load that maximizes peak power (P max) may produce greater increases in peak power than other loads. P max for lower-body lifts can occur with no loading but whether P max can be increased further with negative loading is unclear. The purpose of this investigation was therefore to determine lower-body P max (jump squat) using a spectrum of loads. Box squat 1 repetition maximum (1RM) was measured in 18 elite rugby-union players. P max was then determined using loads of −28 to 60%1RM. Elastic bands were used to unload body weight for negative loads. Jump squat P max occurred with no loading (body weight: 8,880 ± 2,186 W) in all but 2 subjects. There was a discontinuity in the power-load relationship for the jump squat, possibly because of the increased countermovement in the body weight jump. The self-selected depth (dip) before the propulsive phase of the jump was greater by 24 ± 11 to 40 ± 16% (moderate to large effect size) than all positive loads. These findings highlight methodological issues that need to be taken into consideration when comparing power outputs of loaded and unloaded jumps.

Division of Sport and Recreation, Institute of Sport and Recreation Research New Zealand, AUT University, Auckland, New Zealand

Address correspondence to Christos K. Argus,

Back to Top | Article Outline


Rugby union is a competitive sport that requires high levels of muscular power. As such, training methods that enhance muscular power are of extreme importance for the physical preparation in these athletes. The load that maximizes peak power (P max) has been discussed for >20 years and has been suggested to enhance power and performance in explosive exercise (3,18,24). It has been proposed that training at the load that maximizes power may provide favorable neural and muscular adaptations (18,24,28).

To accurately determine the effects of training at P max, P max must be firstly identified. However, large variations in the load that produces P max have been reported (2,3,7,9,14,15,18). Traditionally, findings suggest that P max is typically expressed at loads ranging from 30 to 70% of maximum strength (3,15). More recently, some studies have reported that P max occurs at loads <30% of maximal strength (7,9). The large between-study variation in P max appears to be because of differences in exercise performed, methods used to assess power, and participants recruited (11). As such, using a P max load from the literature in the overall power training program of your athletes may not match their P max load, thereby producing a suboptimal loading and response.

To accurately quantify P max, power outputs across multiple loads need to be investigated. Recently, researchers have reported that vertical jump P max occurred when using body weight only (7,9). However, power was not assessed at loads less than body weight. As such, whether P max can be increased further with negative loading is unclear. A novel approach to assess body weight at negative loads is with the use of elastic bands that may be attached in a manner, which provides upward tension, thereby reducing the effective body weight of the subject.

Another methodological issue is that many investigations have assessed P max using single efforts (repetitions) at each load (7,8,10,14,15), whereas power training typically consists of performing sets of 3-5 consecutive repetitions (4). Furthermore, recent literature has revealed that power is not maximized until the second or third repetition of a set (5). If the overall aim is to train at P max (using consecutive repetitions); then P max should be assessed in the same manner. To date, only Baker and colleagues (2,3) have assessed P max in a training environment performing multiple consecutive repetitions and reported that P max occurred between 40 and 70% of maximum strength. Finally, the experience level (or training history) of subjects assessed may produce variation in the findings. Baker (2) reported that stronger athletes may produce P max at lower intensities than weaker athletes. Therefore, to make accurate comparison between investigations, subjects need to be of similar strength levels.

Elite rugby-union athletes typically have high levels of strength and regularly perform resistance training with multiple sets and consecutive repetitions. If methodology issues have an effect on the load that maximizes peak power, then specific population assessment needs to occur to accurately identify P max. Determining P max in this population will provide athletes with specific training intensities that allow maximal peak power to be achieved during training, which in turn may lead to enhanced performance gains (18,24,28).

Therefore, the purpose of this investigation was to determine lower-body P max in elite rugby-union players. Point of difference from previous investigations included the assessment of P max load at negative through to positive loads.

Back to Top | Article Outline


Experimental Approach to the Problem

To more accurately quantify P max in terms of how it is commonly applied to training programs, elite rugby-union subjects were assessed for lower-body maximal strength and power (via a spectrum of loads including negative loading) across 4 separate sessions, with each session separated by 24 hours (Table 1). Multiple repetitions were performed in each set (1 set of 4 repetitions at each load) to be more representative of a typical training session. Peak power was selected as the dependent measure because it has been reported to have the greatest association with athletic performance (13). Power was assessed using the jump squat exercise because of its common usage in power training programs and research studies and its ability to represent lower-body power (1,3).

Table 1

Table 1

Back to Top | Article Outline


Eighteen elite rugby-union players from a Super 14 professional rugby team during the preseason phase of their campaign volunteered to take part in this study (mean ± SD; age, 23.8 ± 2.2 years; height, 185.8 ± 6 cm; mass, 103.8 ± 10.6 kg). Each subject had undergone at least 2 years of intensive and regular resistance training exercise and must have been competing in a prior national or international rugby competition to be included in this study. Subjects were informed of the experimental risks and signed an informed consent document before the investigation. The investigation was approved by an Institutional Review Board for use of Human subjects (Auckland University of Technology Ethics Committee). Four subjects were unable to attend session 2 because of unforeseen circumstances.

Back to Top | Article Outline



Maximal strength was assessed using the box squat exercise using methods previously outlined (1). Briefly, after 3 submaximal sets of box squat, each athlete then performed 1 set to failure of 1-4 repetitions. Participants used a self-selected foot position and were required to lower themselves to a sitting position briefly on the box and then return to a standing position. The box height was adjusted for each athlete to allow the top of the thighs to be parallel to the floor while in the seated position. The box squat was performed using free weights. Three minutes rest was allowed between each set. Each set to failure was used to predict the athletes' 1 repetition maximum (1RM).

The following equation was used to predict box squat 1RM (20). This equation is a valid measure of 1RM strength because it has been shown to have a correlation between actual and predicted 1RM of r = 0.969 (21):

Back to Top | Article Outline

Jump Squat

Lower-body power was assessed using a jump squat exercise performed in a Smith machine. Subjects warmed up with 2 sets of 4 repetitions lowering the bar to a 90° knee angle using a load of 50% of their 1RM box squat. Subjects then completed 1 set of 4 repetitions of jump squats at −28% (±5%), −15% (±3%), 0% (body weight), 20, 30, 40, 50, or 60% 1RM box squat. Subjects used a self-selected foot position and lowered the bar to a self-selected depth during these performance tests. Subjects were then required to jump as explosively as possible trying to jump as high as they could (1). Three minutes rest was allowed between each set. The body weight jump was assessed using a broomstick, which was placed behind the neck and on the top of the shoulders. The −28% (±5%) and −15% (±3%) jump squats were an assisted jump, performed in a squat cage wearing a climber's harness with an elastic band (Iron Woody LLC, Olney, MT, USA) attached to either side of the harness (at the hip level), with the other end attached above the participant to the top of the squat cage. Two thicknesses of elastic bands were used. The elastic bands provided vertical tension, which reduced the body weight of each participant when the participant was in a standing position with hip and knee fully extended. The reduction in weight was assessed by having subjects stand on scales with and without the attachment of the elastic bands.

The power and displacement produced during each repetition was quantified with a Gymaware™ optical encoder (50-Hz sample period with no data smoothing or filtering; Kinetic Performance Technology, Canberra, Australia) using the methods described elsewhere (12). Quantification of the power produced included body weight and bar mass (system mass) in the calculation (13).

Back to Top | Article Outline

Statistical Analyses

To estimate the load that maximized mechanical power output, a quadratic was fitted to each participant's power output (in watts) and load (% of 1RM). However, in all but 2 subjects, power at body weight was clearly above any quadratic curve fitted to the points (Figure 1). Additionally, for the 4 subjects that did not complete the assisted jumps, the quadratic curves all had positive curvature where theory predicts negative curvature. Therefore, for all subjects we used the value observed at body weight for P max. Findings were discussed as means and SDs.

Figure 1

Figure 1

In addition to fitting a quadratic, standardized differences of the mean were used to assess magnitudes of effects between each individual load assessed by dividing the differences by the appropriate between-athlete SD. Standardized changes of <0.20, <0.60, <1.2, <2.0, and >2.0 were interpreted as trivial, small, moderate, large, and very large effects, respectively (6,17). Lastly, displacement data were log transformed to reduce nonuniformity of error, and the differences were derived by back transformation as percent changes (16). To make inferences about the true (large-sample) value of an effect, the uncertainty in the effects was expressed as 90% confidence limits.

The interclass correlation (ICC) and coefficient of variation (CV) for box squat was r = 0.915 and 4.6%, respectively. The ICCs and CV% for jump squat at 0 and 50% of 1RM box squat were 0.834 and 4.2%, and 0.904 and 4.8%, respectively. All test-retest reliabilities were assessed 7 days apart. Validity of the Gymaware™ optical encoder has been previously reported elsewhere (12). The sample size for this investigation was limited to the number of athletes in the squad. All athletes in the squad who were injury free were included and therefore no more athletes could be obtained.

Back to Top | Article Outline


The mean predicted 1RM box squat was 147.9 kg (±26.8 kg). The greatest lower-body peak power was 8,880 W (±2,186 W) and occurred at body weight (Figure 2). The peak power produced during the body weight jump was greater (moderate to large effect size) than that of all other intensities assessed. Sixteen of the 18 subjects produced peak lower-body power at body weight. Because of the irregularity in the lower-body power results, whereby a quadratic could not be fitted to the points (see Statistical Analyses; Figure 1); we reexamined the GymAware™ data to gain some insight into the potential reasons underlying this result. As the GymAware™ system is a linear position transducer, we started by examining the displacement data to ascertain whether differences in technique between the different jump intensities may have contributed to this finding.

Figure 2

Figure 2

Analysis of the displacement data revealed that during the body weight jump, the self-selected depth (dip) before the propulsive phase of the jump was greater by 24 ± 11 to 40 ± 16% (moderate to large effect size) than all positive loads. As the loads increased, the subjects continued to reduce the depth of their countermovement. Small differences in the countermovement depth ranging from 11% (±11%) to 17% (±14%) were observed between 20 and 40%, 20 and 50%, and 20 and 60% 1RM box squat load. Additionally, small differences ranging from 7% (±9%) to 14% (±9%) were also observed between 30 and 50%, 30 and 60%, and 40 and 60% 1RM box squat load.

Back to Top | Article Outline


It has been hypothesized that training at P max is beneficial for increasing muscular power (3,18,24). Therefore, the purpose of this investigation was to determine lower-body P max in elite rugby-union players. To the author's knowledge, this was the first investigation that assessed P max using negative loads. By assessing power at negative loading, we were able to identify a decline in power either side of the maximum power output which previous authors have not considered (7,9). Peak lower-body power occurred with no loading (body weight) in 16 of the 18 subjects. However, discontinuity in the power outputs of the lower body was observed between body weight and all loaded jumps.

An interesting phenomenon occurred when assessing lower-body power across this spectrum of negative and positive loads. In all but 2 of the subjects assessed, power with no loading was substantially higher than all other loads assessed and was clearly above any quadratic curve that was fitted to all the points (Figure 1). On closer observation, there appeared to be discontinuity of the power outputs between body weight and all positive loads. Indeed negative and body weight loads appeared to have a different power-load relationship than the positive loads. As such, it may be that a separate quadratic needs to be fitted to each power-load relationship when loaded and unloaded intensities are assessed. However, this would result in 2 P max intensities, one for training with unloaded jumps and the other for loaded jumps.

The separate power-load relationships may suggest that something substantially affects power output when subjects jump with an additional load. We reanalyzed the position data produced by Gymaware™ and found that during the body weight jump, the self-selected depth (dip) before the propulsive phase of the jump was greater than all loaded jumps (24-40%). Furthermore, as the loads increased to a greater percent of 1RM, there was a further reduction in the depth of the countermovement. As such, the disproportionally higher power output at body weight may be because of the larger dip used in this jump. The use of a greater dip with the body weight load may have afforded this jump some biomechanical advantages that contributed to the greater power outputs. The deeper countermovement would have increased the time to produce force. According to the impulse-momentum relationship, greater time to produce force would increase the amount of impulse (force multiplied by time) generated, which in turn would result in a greater change in the momentum (velocity) of the system (19). Additionally, the greater dip would have increased the amount of stretch placed on the agonist musculature and via the force-length relationship allow greater forces to be generated (25).

The methodological concerns observed could be controlled by keeping the depth consistent for all jumps. However, what should the constant depth be? If it is too low, velocity of the movement may be compromised and there is chance of increasing the likelihood of injury when jumping with heavy loads. If it is not low enough, it may prevent an optimal combination of force and velocity reducing power output and defeating the purpose of assessing P max. Additionally, how should depth be controlled? Cormie et al. (9) attempted to control depth by visually monitoring knee angle to a depth of 90°. However, Cormie et al. (9) still reported significant differences in depth between the different loading intensities. Harris et al. (14) controlled depth by performing a concentric only jump squat starting at a fixed knee angle of 110°. However, what if the purpose of your training was to improve stretch shortening cycle and countermovement peak power? Young et al. (29) suggested that executing a countermovement at a self-selected depth encouraged subjects to find their own optimum jumping conditions. Furthermore, as previously alluded to in the Introduction, if the goal is to train at the P max load, P max needs to be assessed in the manner it is trained. For most athletes, they will train using a self-selected depth.

The discontinuity in jump technique (amount of dip) between each load makes determining P max for the lower-body problematic. If lower-body P max can not be accurately determined; then the contention that training at the load that maximizes power may provide favorable neural and muscular adaptations (18,24,28) would appear somewhat problematic, at least for the lower body.

Lower-body peak power occurred at body weight, a finding similar to Cormie et al. who reported that lower-body peak power occurred at body weight in well trained (football players, long jumpers, and sprinters) (7) and untrained men (9). In contrast, Siegel et al. (26) reported that peak power occurred between 50 and 70% 1RM squat in untrained subjects, whereas Sleivert and Taingahue (27) reported that peak power occurred at 60% of 1RM squat in trained athletes. The difference in findings is likely because of the inclusion or exclusion of system mass (i.e., bar mass plus body weight) in the calculation of power. In this investigation, and investigations by Cormie et al. (7,9), all of which found peak power to occur at body weight, system mass was included in the calculation of power, whereas the investigation by Sleivert and Taingahue (27) used bar mass only. Additionally, Siegel et al. (26) did not state that system mass was included in their calculations. This becomes extremely important when comparing findings as the inclusion or exclusion of body weight can cause a shift in peak power from 20% (system mass included) to 70% of 1RM (system mass excluded) (13). Therefore, the higher P max observed in the 2 investigations may be artificially high because of the exclusion of system mass from the calculation.

Heavy strength training and high velocity training have been shown to be effective in improving explosive performance in some studies (18,22,23). However, it has been suggested that training at P max may enhance power and performance in explosive exercise more so than heavy strength P max high velocity training (3,18,24). It should be noted that there is only a limited and equivocal literature involving the comparisons of training at P max vs. heavier or lighter loads, and as such the load that maximizes performance adaptation is still somewhat unknown.

The load that maximizes peak power may be influenced by several factors including the spectrum of loads assessed and whether comparisons are made between loaded and unloaded conditions. Additionally, data calculation and reporting methods (i.e., inclusion or exclusion of body weight) may influence P max.

Back to Top | Article Outline

Practical Applications

Lower-body P max occurred at body weight in 16 of the 18 subjects. However, results indicated there was a discontinuity between loaded and unloaded jumps. As such lower-body current P max assessment procedures may be flawed because of the inability to accurately determine the load that maximizes peak power. Methods that can assess and improve lower-body power in a training environment need to be developed. We suggest assessment using a range of heavy and lighter intensities for each individual in each exercise, in a manner similar to how he or she trains. This will increase external validity and possibly result in a greater likelihood of enhanced training adaptations.

Back to Top | Article Outline


The Waikato Rugby Union and the Tertiary Education Commission provided financial support by way of scholarship for the primary author. The results of the present study do not constitute endorsement by the NSCA.

Back to Top | Article Outline


1. Argus, CK, Gill, ND, Keogh, JWL, Hopkins, WG, and Beaven, CM. Changes in strength, power and steroid hormones during a professional rugby union competition. J Strength Cond Res 23: 1583-1592, 2009.
2. Baker, D. A series of studies on the training of high-intensity muscle power in rugby league football players. J Strength Cond Res 15: 198-209, 2001.
3. Baker, D, Nance, S, and Moore, M. The load that maximizes the average mechanical power output during jump squats in power-trained athletes. J Strength Cond Res 15: 92-97, 2001.
4. Baker, D and Newton, RU. Methods to increase the effectiveness of maximal power training for the upper body. Strength Cond J 27: 24, 2005.
5. Baker, DG and Newton, RU. Change in power output across a high-repetition set of bench throws and jump squats in highly trained athletes. J Strength Cond Res 21: 1007-1011, 2007.
6. Batterham, AM and Hopkins, WG. Making meaningful inferences about magnitudes. Int J Sports Physiol Perform 1: 50-57, 2006.
7. Cormie, P, Grant, OM, Triplett, NT, and McBride, JM. Optimal loading for maximal power output during lower-body resistance exercises. Med Sci Sport Exerc 39: 340-349, 2007.
8. Cormie, P, McBride, JM, and McCaulley, GO. The influence of body mass on calculation of power during lower-body resistance exercises. J Strength Cond Res 21: 1042-1049, 2007.
9. Cormie, P, McBride, JM, and McCaulley, GO. Power-time, force-time, and velocity-time curve analysis during the jump squat: Impact of load. J Appl Biomech 24: 112-120, 2008.
10. Cronin, J, McNair, PJ, and Marshall, RN. Developing explosive power: A comparison of technique and training. J Sci Med Sport 4: 59-70, 2001.
11. Cronin, J and Sleivert, G. Challenges in understanding the influence of maximal power training on improving athletic performance. Sport Med 35: 213-234, 2005.
12. Drinkwater, EJ, Galna, B, McKenna, MJ, Hunt, PH, and Pyne, DB. Validation of an optical encoder during free weight resistance movements and analysis of bench press sticking point power during fatigue. J Strength Cond Res 21: 510-517, 2007.
13. Dugan, EL, Doyle, TLA, Humphries, B, Hasson, CJ, and Newton, RU. Determining the optimal load for jump squats: A review of methods and calculations. J Strength Cond Res 18: 668-674, 2004.
14. Harris, NK, Cronin, JB, and Hopkins, WG. Power outputs of a machine squat-jump across a spectrum of loads. J Strength Cond Res 21: 1260-1264, 2007.
15. Harris, NK, Cronin, JB, Hopkins, WG, and Hansen, KT. Squat jump training at maximal power loads vs. heavy loads: Effect on sprint ability. J Strength Cond Res 22: 1742-1749, 2008.
16. Hopkins, WG. Spreadsheets for analysis of controlled trials with adjustment for a predictor. Sportscience 10. Available at: 2006.
17. Hopkins, WG, Marshall, SW, Batterham, AM, and Hanin, J. Progressive statistics for studies in sports medicine and exercise science. Med Sci Sport Exerc 41: 3-12, 2009.
18. Kaneko, M, Fuchimoto, T, Toji, H, and Suei, K. Training effect of different loads on the force-velocity relationship and mechanical power output in human muscle. Scand J Sports Sci 5: 50-55, 1983.
19. Kreighbaum, E and Barthels, K, eds. Linear Momentum and Kinetic Energy. Biomechanics. A Qualitative Approach for Studying Human Movement. Needhham Heights, MA: Allyn and Bacon, 1996. pp. 290-299.
20. Lander, J. Maximums based on reps. Natl Strength Cond Assoc 6: 60-61, 1985.
21. Lesuer, DA, McCormick, JH, Mayhew, JL, Wasserstein, RL, and Arnold, MD. The accuracy of prediction equations for estimating 1-RM performance in the bench press, squat, and deadlift. J Strength Cond Res 11: 211-213, 1997.
22. Mangine, GT, Ratamess, NA, Hoffman, JR, Faigenbaum, AD, Kang, J, and Chilakos, A. The effects of combined ballistic and heavy resistance training on maximal lower- and upper-body strength in recreationally trained men. J Strength Cond Res 22: 132-139, 2008.
23. McBride, JM, Triplett-McBride, T, Davie, A, and Newton, RU. The effect of heavy- vs. light-load jump squats on the development of strength, power, and speed. J Strength Cond Res 16: 75-82, 2002.
24. Newton, RU and Kraemer, WJ. Developing explosive muscular power: Implications for a mixed methods training strategy. Strength Cond J 16: 20-31, 1994.
25. Schachar, R, Herzog, W, and Leonard, TR. Force enhancement above the initial isometric force on the descending limb of the force-length relationship. J Biomech 35: 1299, 2002.
26. Siegel, JA, Gilders, RM, Staron, RS, and Hagerman, FC. Human muscle power output during upper- and lower-body exercises. J Strength Cond Res 16: 173-178, 2002.
27. Sleivert, G and Taingahue, M. The relationship between maximal jump-squat power and sprint acceleration in athletes. Eur J Appl Physiol 91: 46-52, 2004.
28. Wilson, GJ, Newton, RU, Murphy, AJ, and Humphries, BJ. The optimal training load for the development of dynamic athletic performance. Med Sci Sport Exerc 25: 1279-1286, 1993.
29. Young, WB, Jenner, A, and Griffiths, K. Acute enhancement of power performance from heavy load squats. J Strength Cond Res 12: 82-84, 1998.

elite athletes; jump squat; power profile

© 2011 National Strength and Conditioning Association