In most team sports, players are required to repeatedly perform short, explosive efforts, such as sprints, accelerations, decelerations, and jumps (4,5,13,14). As such, the ability to repeat high-intensity efforts has been suggested to be an important fitness determinant in such sports (29). In addition to lower-body explosive actions (e.g., sprints and jumps), in most team sports, such as rugby, handball, basketball, and football, players have to perform high-intensity upper-body actions. For example, the upper-body movements associated with match play in basketball players have been previously quantified, and it has been shown that elite players perform more upper-body actions than their subelite counterparts (28). Furthermore, upper-body maximal dynamic strength (assessed via a bench press exercise) has been reported to be significantly greater in a senior basketball team than in the U-20 and U-18 squads (4) and that elite basketball players achieved significantly greater maximal performance than average-level players in the bench press exercise (9). It has also been reported that bench press maximal strength was able to differentiate between playing positions in basketball players (4). Therefore, upper-body strength seems to be an important physical fitness attribute for basketball players.
Several studies have assessed the determinants and the reliability of different lower-body (i.e., sprints and jumps) repeated effort tests in team sport players (6,8,22,30). For example, Mendez-Villanueva et al. (26) reported high correlations between maximal sprinting speed (attained during a single, 40-m sprint test) and the mean sprint time during a repeated sprint ability (RSA) test (10 × 30 m sprints) in highly trained, young soccer players. Similarly, very large correlations have also been found between maximal sprinting speed and RSA in team sports athletes (7). Collectively, these results suggest that neuromuscular factors are likely to be the main determinants of the ability to repeat lower-body (i.e., sprints), high-intensity actions in trained team sport athletes. To the best of our knowledge, no data are currently available on such associations in upper-body exercise modes.
The reliability of several tests that evaluate the ability to repeat high-intensity efforts such as the RSA (6,22), the repeated jump and sprint ability (RJSA) (8), and the repeated change of direction ability (RCODA) (32) has been previously reported. In addition to the reliability of a test, the concept of “usefulness” is required to increase the relevance of a field test because it extends its practicality to confidently monitor the progression of a player (8). To date, the reliability and usefulness of a repeated effort, upper-body test has never been evaluated. Therefore, the aims of this study were (a) to investigate the main determinants of an upper-body (bench press) repeated power ability (RPA) test, (b) to examine the reliability of this RPA test, and (c) to determine its potential usefulness and practicality in the field to monitor meaningful changes in player's performance.
Experimental Approach to the Problem
Highly trained, young basketball players were assessed using the bench press exercise to determine their ability to repeat maximum power output with the upper body. The bench press exercise was chosen because it is one of the most common exercises used to develop upper-body strength and power in team sports (1). Furthermore, it is the most used exercise to assess and monitor upper-body strength and power in contact and collision sports (20). The experimental schedule was composed by 4 consecutive testing sessions: first, players did a 1-repetition maximum (1RM) bench press test. Second, an incremental load test to determine the load that maximized power output was carried out. Finally, the RPA tests were performed twice within 5–7 days. Average power (AP) output was used to quantify performance in the RPA test. Relative (intraclass correlation coefficient [ICC]) and absolute (coefficient of variation [CV]) reliability were used to assess the reproducibility of the RPA test. The usefulness or sensitivity of the test was analyzed to understand the real changes in test performance.
Forty-five young (U-16, U-17, and U-18), highly trained, male basketball players (age, 16.6 ± 1.0 years; height, 189.5 ± 9.5 cm; body mass, 81.1 ± 9.7 kg) volunteered to participate in this study. Data collection took place in the second month of the competitive season after a 2-month preseason period. All players were training in a basketball club for at least 7 years and participated on average in approximately 10 hours of combined basketball (5–6 sessions), strength/power (2 sessions), and 2 competitive matches per week. At the time of the study, all players were competing at national level (i.e., Spanish Basketball National League). Furthermore, some players (n = 15) were also competing at international level (i.e., European and World Basketball Championships). Every subject had a minimum of 1-year experience training with the bench press exercise. Written informed consent was obtained from both the players and their parents before the investigation. The present study was approved by the institutional research ethics committee and conformed to the recommendations of the Declaration of Helsinki.
Before commencement of the RPA tests, players were familiarized with the exercise procedure and chose the grip width that was the most comfortable, which was measured and kept constant throughout all testing sessions. The test was completed twice in the same week with 5–7 days difference between each trial. Every trial was carried out at the same time of the day (18:00–20:00). Each testing session was preceded by a 15-minute standardized warm-up, which included jogging, arm and shoulder mobilization, 1 set of 8 repetitions at 40% of 1RM, 1 set of 4 repetitions at 60% of 1RM, and 1 set of 3 repetitions at 45% of 1RM, all performed in an explosive manner. After warming up, subjects rested 5 minutes before starting the RPA test.
Subjects performed their bench presses in a free weight setting on a bench using a standard 20-kg Olympic barbell (York International Olympic Training Bar, Cranston, RI, USA). The tether of a linear encoder (MuscleLab Ergotest Technology, Langesund, Norway) was attached to the bar and used to measure its vertical displacement. Data were sampled at a frequency of 100 Hz (time resolution of 10 milliseconds) with an accuracy of 0.075 mm, recorded by an acquisition unit, and stored on a portable computer equipped with the data acquisition and analysis program. The software (version 8; MuscleLab) displays the time course of displacement and velocity. This device has been widely used to evaluate dynamic muscle work, and good reliability scores have been reported (10), and in pilot studies, the maximum power output load calculated by the software has shown a good reliability (ICC = 0.96 and CV = 3.70%).
One Repetition Maximum Assessment
At least 72 hours before beginning the incremental load test, the 1RM in the bench press exercise was determined. The 1RM test was conducted as previously described (15). The warm-up consisted in 1 set of 10 repetitions with 50% of estimated 1RM and 1 set of 4 repetitions with 60% of estimated 1RM. Then, players performed 4 subsequent attempts to determine 1RM. Previously documented training experience was used as guidance for selecting the initial test load (i.e., 4 subsequent attempts). The subjects lowered the bar to the chest but were not allowed to bounce the bar on the chest. They were required to press the bar to full elbow extension to compute a valid lift. Five minutes of passive recovery were given between each attempt.
Incremental Load Test
An incremental load test was used to determine the load that maximized power output (23). Four loads (20, 40, 60, and 80% of 1RM) were used to determine the force-velocity relationship (31). Subjects started with the bar at arm's length and lowered the bar until the chest was touched lightly without bounding the bar. Thereafter, without pause between eccentric and concentric phases, they moved the bar up at maximum velocity. The test began with 20% of 1RM with subsequent increments up to 80% of 1RM. At least 5-minute recovery was provided between loads. The maximum power output load was calculated by the software (version 8; Musclelab).
Repeated Power Ability Test
To test the RPA, a test that consisted of 5 sets of 5 repetitions, with the load that maximized power output, with 30 seconds of passive recovery between sets was employed. Players started with the barbell at arm's length and performed 5 repetitions without pause between the eccentric and concentric phases. The concentric phase was performed as fast as possible, and the eccentric phase was executed at slower velocity (i.e., self-selected and never exceeding the 3 seconds) than the concentric phase. Each set was only accepted if during the 5 repetitions, the bar was lowered until the chest was touched lightly during the eccentric phase, and the concentric phase terminated with a full elbow extension. In addition, bouncing the bar on the chest was not allowed. During the 30-second recovery between sets, subjects were required to stand passively. Five seconds before starting the set, the subjects were asked to assume the starting position, maintaining the bar with a full elbow extension and waiting for the start signal (5-second countdown). Strong verbal encouragement was provided to each subject during each repetition. The variables used for the analysis were AP of each set, AP over 5 sets (APmean), and percentage of AP decrement (%Dec). Percentage of AP decrement was calculated using the following formula: %Dec = 100 − (APmean/AP in set 1 [APbest] × 100), which was proposed by Fitzsimons et al. (12) and modified by Edge et al. (11).
Data are presented as mean ± SD. The distribution of each variable was examined with the Shapiro-Wilk normality test. Relative reliability analysis was examined by the ICC. An ICC ≥ 0.70 was considered acceptable (3,19). To examine absolute reliability, pairwise comparisons were first applied with paired t-test to assess any significant differences between the sessions. The significance level was set at p ≤ 0.05. The magnitude of between-session differences was also expressed as standardized mean difference (Cohen effect sizes, ES). The criteria to interpret the magnitude of the ES were as follows: <0.2 = trivial, 0.2–0.6 = small, 0.6–1.2 = moderate, and >1.2 = large (17). The spreadsheet of Hopkins (16) was also used to determine the change in the mean between trials and the typical error of measurement (TEM), expressed as a CV (%). A CV of <5% was set as the criterion for reliability. The usefulness of the test was assessed while comparing the smallest worthwhile change (SWC) with the TEM (18). The SWC was determined by multiplying the between-subject SD by 0.2 (SWC0.2) (18), which corresponds to a small effect, 0.6 (SWC0.6), which correspond to a moderate effect, and 1.2 (SWC1.2), which corresponds to a large effect. If the TEM was below the SWC, the test was rated as “good”; if the TEM was similar to the SWC, the rating was “OK”; and if the TEM was higher than the SWC, a rating of “marginal” was given (21). The Pearson product moment correlation coefficient was used to determine the relationship between different variables. The following criteria were adopted for interpreting the magnitude of correlation (r) between test measures: ≤0.1 = trivial, >0.1–0.3 = small, >0.3–0.5 = moderate, >0.5–0.7 = large, >0.7–0.9 = very large, and >0.9–1.0 = almost perfect (17). If the 90% confidence limits overlapped small positive and negative values, the magnitude of the correlation was deemed unclear; otherwise, the magnitude was deemed to be the observed magnitude (17).
1RM Test and Maximum Power Output Load
One repetition maximum was 64.7 ± 10.3 kg in the bench press exercise. The estimated load that maximized power output was 29.4 ± 5.0 kg, and it was established at 45.4 ± 1.8% of 1RM.
Test Reliability and Usefulness
Results of RPA test are presented in Table 1. There were no significant differences (p ≥ 0.05) between trial 1 and trial 2 in any variable. Between-trial ESs were all trivial (ES < 0.2) (Table 2). All the other measures of reliability of the RPA test are presented in Table 2. The changes in performance to be considered small, moderate, and large are displayed in Table 2.
Relationship Between Maximal Power and Repeated Power Ability
Correlation coefficient between each variable is presented in Table 3; APbest (i.e., maximal power output) and APmean correlated almost perfectly (r = 0.99; 90% confidence interval, 0.98–0.99) (Figure 1). However, APbest and %Dec showed a trivial correlation (r = 0.06) (Figure 2).
The aim of the present study was, for the first time, to analyze the main determinants of an upper-body (i.e., bench press) RPA test and to examine its reliability and usefulness. The main findings of our study were the almost perfect correlations between APmean and APbest, whereas %Dec showed trivial or small correlations with APmean or any other variable. With the exception of the %Dec, the other variables examined showed very high reliability scores with an apparent absence of a learning effect (i.e., all between trial standardized differences were rated as trivial). Based on the usefulness analysis, the bench press RPA test can detect small performance changes (SWC calculated as 0.2 × SD) in maximum (APbest) and mean (APmean) power output in young basketball players.
The RPA test provided very high reliability scores with APmean as the most reliable variable. Previous researchers have found good reliability for measures of maximum power output at different 1RM loads in a bench press exercise (2,23). The CV attained for the APbest in the present study (3.6%) was similar to the values obtained for the maximum power output with the load that maximizes power output (50% 1RM) in professional soccer players (4.1%) (23) and higher (2.1%) than what has been reported in a mixed group of athletes (i.e., basketball, handball, volleyball, sprinters) (2). The high ICC value for APbest (0.97) and APmean (0.97) also indicate a high level of relative reliability. These ICC values are similar to those reported for maximum power output in male physical education students (0.97) (24). Nevertheless, lower ICC scores have been previously reported in highly trained athletes (0.87) (2) and soccer players (0.86) (23). Differences in subjects' training background, sporting activity, or measurement device might explain these between studies differences in both absolute and relative reliability.
The %Dec showed the worst reliability scores among all the other RPA parameters analyzed. Surprisingly, to the best of our knowledge, no study has previously analyzed this parameter in an upper-body action (i.e., bench press). Therefore, it is not possible to compare our data with previous studies. Nevertheless, fatigue measurements (i.e., %Dec) in different repeated explosive lower-body test, such as RSA, RCODA, or RJSA, have been typically reported to be the less reliable variable (7,8,32).
Because the TEM values were lower than those calculated for the SWC0.2, the usefulness of the RPA test was rated as “good” for APbest, AP in set 3 (AP3), AP in last set (APlast), and APmean. Average power in set 2 (AP2) and AP in set 4 (AP4) were rated as “OK” because of their similar SWC0.2 and TEM, whereas %Dec showed higher TEM than that reported to SWC0.2 rating as “marginal.” From a practical viewpoint, this means that even small changes (0.2 times the APbest and APmean between-subject SD) would be detected by this test. In addition, the calculated SWC0.2 values for APbest (4.3%) in the current study are well below the range of reported changes in maximal power output performance after different short strength and power training periods (10,25). For example, a group of male subjects with similar 1RM values (59.3–61.3 kg) than our players on the bench press improved their maximum power output by an average of 7.6% after a conventional maximal strength training (i.e., 80–95% of 1RM) and by 16.7% after a maximal power training (i.e., 20–50% of 1RM) (25). Similarly, a 10.6% and a 6.5% improvement in maximal power output in bench press exercise was reported in highly trained young basketball and soccer players after a repetition failure training and nonrepetition failure training, respectively (9). All together, the ability of the present RPA test to detect small changes in power output and the typically greater magnitude of the changes in responses to different training interventions previously reported using similar populations seem to confirm its usefulness and sensitivity in real-world scenarios.
Pearson correlation coefficients showed almost perfect correlations between each set and APmean. An almost perfect relationship (Figure 1) was established between APbest and APmean, which is in agreement with other RSA (7,26,27) or RCODA (32) studies. On the contrary, %Dec showed only small and trivial correlations with all other power output variables (Table 3). Although correlations do not imply causality, the strong correlation between APbest and APmean suggests that the main RPA determinant in highly trained young basketball players is APbest. These associations might be taken into account when determining which factors should be trained to improve RPA in young basketball players. However, it should be acknowledged that the RPA protocol (e.g., number of repetitions and recovery duration) is likely to influence APmean and therefore should be taken into account when testing and training to improve RPA (26).
In summary, because of the high reliability scores found in this test, this protocol may be used to assess upper-body RPA in relatively experienced subjects. Further studies are needed to establish the sensitiveness of this test to different training programs and the trainability of the upper-body RPA.
The upper-body RPA test used here has been shown to be reliable and useful to assess power capabilities in highly trained young basketball players. The current RPA test seems to be appropriate to monitor short- and long-term changes in power output during a bench press exercise because of the low TEMs for maximal and AP output. Based on the present results and previously published training data, we suggest that in young, male basketball players, coaches might interpret a performance improvement of approximately 4% in either APbest or APmean as the smallest real increase in bench press performance. In this regard, preliminary data (Tous-Fajardo and Moras, unpublished observations, 2009) have shown improvements of approximately 20% in APbest after a 7-week RPA training intervention. These authors propose 2 training strategies: maximal power endurance (APbest) and RPA (APmean) that seem to be closely associated in lights of the findings of the present study. Finally, the RPA main determinant was the AP in the first/best set (APbest). Thus, it would be expected that to enhance the ability to repeat power during upper-body actions in highly trained basketball players, training efforts should be directed to improve the maximum power output over one set. Additional research with players of different sport, age, gender, and playing standards should be conducted before the applicability of the current results can be generalized.
1. Argus CK, Gill ND, Keogh JW, Hopkins WG, Beaven CM. Changes in strength
, power, and steroid hormones during a professional rugby union competition. J Strength
Cond Res 23: 1583–1592, 2009.
2. Asci A, Acikada C. Power production among different sports with similar maximum strength
. J Strength
Cond Res 21: 10–16, 2007.
3. Baumgartner TA, Chung H. Confidence limits for intraclass reliability coefficients. Meas Phys Educ Exerc Sci 5: 179–188, 2001.
4. Ben Abdelkrim N, Chaouachi A, Chamari K, Chtara M, Castagna C. Positional role and competitive-level differences in elite-level men's basketball players. J Strength
Cond Res 24: 1346–1355, 2010.
5. Ben Abdelkrim N, El Fazaa S, El Ati J. Time-motion analysis and physiological data of elite under-19-year-old basketball players during competition. Br J Sports Med 41: 69–75, 2007; discussion 75.
6. Bishop D, Spencer M, Duffield R, Lawrence S. The validity of a repeated sprint ability test. J Sci Med Sport 4: 19–29, 2001.
7. Buchheit M. Repeated-sprint performance
in team sport players: Associations with measures of aerobic fitness, metabolic control and locomotor function. Int J Sports Med 33: 230–239, 2012.
8. Buchheit M, Spencer M, Ahmaidi S. Reliability, usefulness, and validity of a repeated sprint and jump ability test. Int J Sports Physiol Perform 5: 3–17, 2010.
9. Delextrat A, Cohen D. Physiological testing of basketball players: Toward a standard evaluation of anaerobic fitness. J Strength
Cond Res 22: 1066–1072, 2008.
10. Drinkwater EJ, Lawton TW, Lindsell RP, Pyne DB, Hunt PH, McKenna MJ. Training leading to repetition failure enhances bench press strength
gains in elite junior athletes. J Strength
Cond Res 19: 382–388, 2005.
11. Edge EJ, Bishop D, Hill-Haas S, Dawson B, Goodman C. Comparison of muscle buffer capacity and repeated-sprint ability of untrained, endurance-trained and team-sport athletes. Eur J Appl Physiol 96: 225–234, 2006.
12. Fitzsimons M, Dawson B, Ward D, Wilkinson A. Cycling and running test of repeated sprints ability. Aust J Sci Med Sport 25: 82–87, 1993.
13. Gabbett T, King T, Jenkins D. Applied physiology of rugby league. Sports Med 38: 119–138, 2008.
14. Gorostiaga EM, Granados C, Ibanez J, Gonzalez-Badillo JJ, Izquierdo M. Effects of an entire season on physical fitness changes in elite male handball players. Med Sci Sports Exerc 38: 357–366, 2006.
15. Hoffman JR. Norms for Fitness, Performance
and Health. Champaign, IL: Human Kinetics, 2006.
16. Hopkins WG. Reliability from consecutive pairs of trials (Excel spreadsheet). Internet Society for Sport Science. Available at: sportsci.org/resource/stats/xrely.xls
. Accessed March 13, 2012.
17. Hopkins WG, Marshall SW, Batterham AM, Hanin J. Progressive statistics for studies in sports medicine and exercise science. Med Sci Sports Exerc 41: 3–13, 2009.
18. Hopkins WG, Schabort EJ, Hawley JA. Reliability of power in physical performance
tests. Sports Med 31: 211–234, 2001.
19. Hori N, Newton RU, Kawamori N, McGuigan MR, Kraemer WJ, Nosaka K. Reliability of performance
measurements derived from ground reaction force data during countermovement jump and the influence of sampling frequency. J Strength
Cond Res 23: 874–882, 2009.
20. Hrysomallis C. Upper-body strength
and power changes during a football season. J Strength
Cond Res 24: 557–559, 2010.
21. Impellizzeri FM, Marcora SM. Test validation in sport physiology: Lessons learned from clinimetrics. Int J Sports Physiol Perform 4: 269–277, 2009.
22. Impellizzeri FM, Rampinini E, Castagna C, Bishop D, Ferrari Bravo D, Tibaudi A, Wisloff U. Validity of a repeated-sprint test for football. Int J Sports Med 29: 899–905, 2008.
23. Jandacka D, Uchytil J. Optimal load maximizes the mean mechanical power output during upper extremity exercise in highly trained soccer players. J Strength
Cond Res 25: 2764–2772, 2011.
24. Jandacka D, Vaverka F. A regression model to determine load for maximum power output. Sports Biomech 7: 361–371, 2008.
25. Jidovtseff B, Croisier J, Scimar N, Demoulin C, Maquet D, Crielaard J. The ability of isoinertial assessment to monitor specific training effects. J Sports Med Phys Fitness 48: 55–64, 2008.
26. Mendez-Villanueva A, Buchheit M, Kuitunen S, Douglas A, Peltola E, Bourdon P. Age-related differences in acceleration, maximum running speed, and repeated-sprint performance
in young soccer players. J Sports Sci 29: 477–484, 2011.
27. Pyne DB, Saunders PU, Montgomery PG, Hewitt AJ, Sheehan K. Relationships between repeated sprint testing, speed, and endurance. J Strength
Cond Res 22: 1633–1637, 2008.
28. Scanlan A, Dascombe B, Reaburn P. A comparison of the activity demands of elite and sub-elite Australian men's basketball competition. J Sports Sci 29: 1153–1160, 2011.
29. Spencer M, Bishop D, Dawson B, Goodman C. Physiological and metabolic responses of repeated-sprint activities specific to field-based team sports. Sports Med 35: 1025–1044, 2005.
30. Spencer M, Pyne D, Santisteban J, Mujika I. Fitness determinants of repeated-sprint ability in highly trained youth football players. Int J Sports Physiol Perform 6: 497–508, 2011.
31. Tihanyi J, Apor P, Petrekanits M. Force-velocity-power characteristics for extensors of lower extremities. In: Biomechanics X-B. Jonsson, B, ed. Champaign, IL: Human Kinetics, 1987. pp. 707–712.
32. Wong DP, Chan GS, Smith AW. Repeated sprint and change-of-direction abilities in physically active individuals and soccer players: Training and testing implications. J Strength
Cond Res 26: 2324–2330, 2012.
Keywords:Copyright © 2014 by the National Strength & Conditioning Association.
high-intensity efforts; strength; performance; upper body