A common element of success in many sports is the ability to sprint at a high velocity. For athletes to continuously increase their sprinting speed, it becomes necessary to begin strength training (26). A key factor in implementing strength training with the aim of improving sprinting speed is selecting exercises that transfer to sprinting (6). Identifying key exercises and tracking improvement in them is a key element of sprint-specific strength training.
A common form of strength training that is used to increase sprinting speed is stretch shortening cycle training. Stretch shortening cycle training or “plyometrics” are exercises characterized by the use of jumping exercises that involve a prestretch of muscles, a short transition period from eccentric to concentric, and a powerful contraction of the joints involved (12). Drop jumping is a popular method for training the stretch shortening cycle (2). Drop jumps (DJs) originated in the former Soviet Union from the research of Yuri Verkhoshansky who developed this type of training and referred to it as the “shock method” (20).
Drop jumps are also used as an assessment tool to analyze the effectiveness of an athlete's stretch shortening cycle. There are 3 important parameters of DJs that must be taken into consideration when using them during testing and training (4,5,21). They are as follows: (a) Drop Height—box height that the athlete is dropping from; (b) Ground Contact Time—the time the athlete spends on the ground after landing and before jumping; and (c) Jump Height—the height the athletes raise their center of gravity.
A higher dropping height will result in a lower jump height in untrained athletes, whereas skilled jumping athletes will actually jump higher (12). A higher dropping height will lead to a longer ground contact time (21) and greater forces acting on the knee and ankle joints (4). When the athlete intentionally attempts to shorten ground contact time, it leads to an increase in peak force (3) but a decrease in jump height (21).
Several methods have been devised to assess DJ performance. A simple method is measuring jump height (DJ H) from several different standard drop heights and not considering ground contact time (10). Another method includes measuring jump height from several standard drop heights and excluding any jumps that have a ground contact time >0.25 seconds (18). Another popular method has been to standardize the drop height but take the jump height and divide it by the ground contact time. This gives a measure called the reactive strength index (RSI) (23). This method has become a very popular assessment tool among practitioners when measuring DJs.
With all these different methods of assessing DJs, it becomes necessary to ask which method is most relevant. If a DJ measure has a high correlation with sprinting speed, it would be assumed that it is the most relevant measure of DJ performance. Although correlation does not mean cause and effect, it does suggest that a relationship may be important. The aim of this study was thus to examine (a) which DJ measure most highly correlates with sprinting speed and (b) compare the correlations with other common tests such as squatting strength, countermovement jump (CMJ), and squat jump (SJ). It was hypothesized that RSI would have the highest correlation of all the different tests with sprinting speed.
Experimental Approach to the Problem
This study was designed to identify which measure of a drop jump, DJ H or RSI, has the highest relationship with sprinting speed over 60 m. These results were to be compared with maximal squat strength relative to body weight, CMJ, and SJ. The rationale of the experiment was to help identify key tests for a testing battery in female rugby players. Specifically, the study wanted to confirm the importance of using DJs as a testing measure and determine what testing result was most important for female rugby players. Sprint times, DJ measures, and maximal front squat strength were assessed, and Pearson product correlation coefficients were calculated between selected variables.
The subjects involved in the study were 15 female university rugby players (height = 1.71 ± 0.5 m, body mass = 71.65 ± 9.99 kg, strength training background = 2.67 ± 1.11 years) who would be considered high-level athletes in their sport with 14 of the 15 subjects having been Canadian national team members at the senior or junior level. All of the players had a minimum of 3 months of resistance training background. The subjects regularly undertook sprint and strength training as part of their regular team training under the supervision of a strength and conditioning coach and would be considered competent at sprinting and basic strength training exercises. Regular strength sessions involved exercises such as power cleans, full squat cleans, backs squats, front squats, bench press, military press, power jerk, and various other exercises. Before any testing, all of the subjects signed consent forms approved by the University of Western Ontario Research Ethics Board in compliance with human testing.
The subjects were tested for maximal front squat strength, a 60-m sprint with 0- to 10-m, 10- to 30-m, and 30- to 60-m splits, CMJs, SJs, and DJs from 0.12, 0.24, 0.36, 0.48, 0.60, 0.72, and 0.84 m. One week before the testing, the subjects underwent a familiarization session for the DJs. The athletes were all familiarized to the other tests through their regular team testing. The testing took place over 2 days with the sprint testing followed by the maximal front squat test happening on the first day. Two to three days after the sprint and strength testing, the subjects were tested for all of the jumping tests. In addition to the physical testing, the subjects heights were also measured using a measuring tape, and their mass was taken using a Kistler force plate (model 9287BA, Winterthur, Switzerland). Height and mass were measured on the day of the jump testing. All of the tests took place in the late afternoon between 15:00 and 17:00 hours during regular team training times. Hydration status was not measured; however, hydration protocols were previously established with the team, and athletes were reminded to follow them during the testing sessions.
The subjects were at the end of a 3-week transition period between the competitive season and their off season and were not currently undertaking any heavy physical training. The absence of any heavy physical training meant that the subjects showed up completely fatigue-free to the testing sessions, but detraining would be minimal. The testing sessions were also part of the subjects' regular team testing, so it could be assumed that the subjects were highly motivated for the testing.
The subjects underwent a 15-minute dynamic warm-up involving dynamic stretching, skipping, and bounding movements. This type of warm-up has been shown to optimize performance (22). In addition to the dynamic warm-up, the subjects also did several sprints ranging from 10 to 60 m before the sprint tests.
The sprint testing involved using an infrared timing system (Equine Electronic, Peculiar, MO, USA) sampling at 1,000 Hz with gates set at 0, 10, 30, and 60 m. Splitting the 60-m sprint into 3 sections allowed for different components the 60-m sprint to be analyzed (7). The splits represented initial acceleration, acceleration to maximal speed, and maintenance of maximal speed. The gates consist of 2 infrared timing devices sitting on 1.03-m tripods facing each other perpendicular to the plane of motion. When an object breaks the beams between the 2 devices, a signal is sent to a handheld receiver that is attached to a stopwatch (Robic, Oxford, CT, USA). The subjects were instructed to begin with their front foot beside a marker that was placed 0.5 m in front of the first gate. Each subject was given 3 trials with a rest time of 5 minutes between each trial. The trial that had the fastest 60-m time was kept for analysis.
The strength testing involved the subjects doing a maximal number of front squats with a mass that allowed them to do no more than 5 repetitions. The Epley prediction equation (1 repetition maximum [1RM] = ((0.033*# of Repetitions) × Mass Lifted) + Mass Lifted), which has been validated in women before (9), was used to predict the 1RM of each subject.
Subjects were encouraged to choose a weight that they thought they could lift twice. The amount of repetitions that could be done was capped at 5 so that maximal set was done with >85% of their 1RM and to reduce some of the error in the prediction equation. The predicted 1RM squatting score was kept as an absolute value and as a value expressed relative to body weight for analysis (1RM/BW). Every subject could comfortably hold the bar in the “clean catch” position and was required to squat down to a thighs below parallel position where there hamstrings would touch their calf muscles. All of the participants regularly did front and back squats in their training and were competent with front squats. The front squat was chosen over the back squat because it was an easier method of dropping a weight during a failed maximal attempt.
Jump Testing Data Collection
All DJs, CMJs, and SJs (10) were performed and collected on a Kistler force plate (model 9287BA). The subjects were all familiar with this method from their regular team fitness testing. Before the jump testing, subjects underwent a 15-minute dynamic warm-up involving dynamic stretching, skipping, and bounding movements. The subjects were given 3 trials for both SJ and CMJ with 30 seconds between each jump.
Two Kistler forces plates (models 9287BA and 9287B) were used to collect the DJ data using the method suggested by Baca (1). Wooden boxes at the designated heights were placed on 1 of the force plates (Kistler 9287BA). Subjects would begin the DJ by stepping off the wooden box without lifting their center of gravity and landing on the other force plate (Kistler 9287B). The subjects were encouraged to “jump as high as they can while trying to minimize ground contact time.” Each force plate was sampling at 1,000 Hz.
The subjects were given 3 trials at each dropping height. The subjects were given 45 seconds in between each DJ at each height, which has been shown to be adequate rest between DJs (14). The subjects were given 3 minutes in between dropping heights. The tests began at the lowest (0.12 m) dropping height and progressively grew higher until they reached the highest (0.84 m) dropping height.
Jump Testing Data Analysis
Force data were amplified, converted from analog to digital data and transferred to Bioware (Winterthur, Switzerland) software on the attached computer. The data were then exported into Microsoft Excel (Seattle, WA, USA) for further analysis. Using a forward dynamics approach, the force data were integrated to calculate velocity and displacement. The suggestions of Street et al. (19) were used, which have been shown to allow for the most valid calculation of jump height.
The highest displacement of the center of gravity obtained during CMJs and SJs was kept as the jump height (CMJ H and SJ H). For the DJs, the highest point the subject jumped after landing from the box was kept as the jump height (DJ H) (see Figure 1 for sample displacement curve). The RSI (21) was taken as the maximal height the subject reached during the DJ divided by the ground contact time.
The jumping trial with the highest result for the examined variable (jump height and RSI) was kept for analysis. The Pearson product-moment correlation was calculated to determine the relationship between each of the performance. The level of significance was set at p ≤ 0.05. All statistical analyses were conducted using Statistical Practice for Social Sciences Version 15 (Chicago, IL, USA). The reliability, as estimated with intraclass correlations coefficients, of the measurements are presented in Table 1 and were all high (0.80-0.97).
Mean performance scores are given in Table 2. Statistically significant relationships were found between 0- and 10-m split times and DJ heights from all dropping heights, CMJ JH and SJ JH (Table 3). Statistically significantrelationships were found between 10- and 30-m split times and DJ JH from all dropping heights, RSI from 0.72- and 0.84-m dropping heights, CMJ height, and SJ height (Table 3). Statistically significant relationships were found between 30- and 60-m split times and DJ heights from all dropping heights, RSI from 0.72- and 0.84-m dropping heights, CMJ H and SJ H (Table 3). The CMJ H, SJ H, and 0.84-m DJ H all had stronger relationships with each sprinting phase than the 1RM/BM did. The DJ H had a stronger relationship with the sprinting times than the RSI at all of the dropping heights. Both of these measures had their strongest correlations with sprinting times from the 0.84-m dropping height (Figures 2-7).
This investigation found that DJ H had a significant correlation with sprinting ability in female rugby athletes from every dropping height. A similar relationship was previously found by Mero (16) with track and field sprinters (r = 0.72). This study did not find that RSI had as strong a relationship, as previously hypothesized, with sprinting speed as DJ H. The RSI had moderate correlations only with the 10- to 30-m split and only at 0.72- and 0.84-m dropping heights. Hennessy and Kilty (11) found a relationship between sprinting speed and RSI (r = 0.75). That particular study did not, however, examine the relationship between DJ H and sprinting speed. Young et al. (25) found that RSI was unrelated to early acceleration or maximal velocity in a group of sprinters. Cronin and Hansen also did not find a relationship between RSI and sprinting speed (8). It is of interest to note that the higher dropping heights (0.72 and 0.84 m) had the highest correlation with sprinting speed. This may suggest that the ability to store elastic energy from high dropping heights loads the lower body in a fashion similar to the large eccentric forces during high-velocity sprinting.
It is interesting to note that this study found that DJ H had a higher correlation with sprinting speed than RSI. It is worthwhile to note the influence that ground contact time has on RSI, and this may explain why RSI had lower correlations than DJ H did. In the Young et al. (24) study, the authors examined the effects of changing the instruction for the execution of the DJs. There are some interesting comparisons with this study during DJs where the subject was told to maximize jump height from a 0.45-m dropping height. The closest dropping height in this study was 0.48 m, so it is used as a base of comparison. The average maximal jump height in the male subjects was slightly higher than the results of this study (0.40 vs. 0.38 m), but the RSI was lower than that in this study (101 vs. 127 cm·s−1) because of a higher ground contact time (0.41 vs. 0.30 seconds). They saw changes in jump height and ground contact time when the subjects were given instructions to maximize RSI. Average jump height dipped below what was recorded in this study (0.32 vs. 0.37 m), but RSI was higher (188 vs. 127 cm·s−1) via a decrease in ground contact time (0.18 vs. 0.30 seconds). In this study, the subjects were given the instructions to jump as high as they could but to do so in the minimal amount of time. They were not given feedback after every trial to manipulate their jumping parameters to maximize RSI.
It is the contention of Yuri Verkhoshansky (personal communication, January 2008) that the ground contact time is an important parameter of drop jumping, but it is secondary to jump height. Verkhoshansky bases this opinion on the extensive research he did with national team track athletes in the former Soviet Union. Although it is intellectually appealing to use RSI as a way of monitoring stretch shortening cycle function, it has to be questioned for a couple of reasons. Drop jump height has been shown to have a strong relationship with sprinting speed, whereas RSI has been equivocal. The RSI is also heavily dependent on ground contact time, which can distort the results. A better approach while monitoring athletes over a training period may be their ability to increase DJ H and then secondarily their ability to concurrently decrease ground contact time. For example, athletes could do a DJ where they had touched a preset target with their hand or land on another box and then measure the ground contact time with a switch mat or force plate.
The other area that was being examined was how DJs compared against other commons tests as a predictor of sprinting speed. The 1RM/BM showed a significant relationship with the 10- to 30-m split but not the 0- to 10-m and 30- to 60-m split. Strong relationships between 1RM/BM and sprinting speed have been previously identified (2,15). The results of this study support those studies. The CMJ H and SJ H both showed strong relationships with sprinting speed, which has also been previously reported (8,13,16). The reason why the DJ H had a stronger correlation than 1RM/BM, CMJ H, and SJ H with sprinting speed may be because of neuromuscular specificity. It is possible that DJs more closely simulate the joint angles, contraction velocities, and the specific stretch shortening cycle function of sprinting.
The results of this study would suggest that there may be a relationship between drop jumping and sprinting speed. It would appear that the height achieved in a DJ may be a more beneficial measure than the RSI. It would be recommended that coaches and practitioners use 2 different tests while assessing stretch shortening cycle function rather than RSI. The maximal height during a DJ can first be assessed to monitor progress. If a coach also desires to specifically monitor ground contact time, it may be useful to set the jump height for the athlete by having the athlete jump onto a box or touch a target overhead set at a standard height and measure the ground contact time with a switch mat or force plate. The athlete's ability to shorten the ground contact time while still jumping to the target can be monitored to see if progress is being made in training. Although correlation does not mean cause and effect, the results could suggest that including DJs in a training program may help improve sprinting speed.
1. Baca, A. A comparison of methods for analyzing drop jump performance. Med Sci Sports Exerc
31: 437-442, 1999.
2. Baker, D and Nance, S. The relation between running speed and measures of strength and power in professional rugby league players. J Strength Cond Res
13: 230-235, 1999.
3. Bobbert, MF. Drop jumping as a training method for jumping ability. Sports Med
9: 7-22, 1990.
4. Bobbert, MF, Huijing, PA, and van Ingen Schenau, GJ. Drop jumping. I. The influence of jumping technique on the biomechanics of jumping. Med Sci Sports Exerc
9: 332-338, 1987.
5. Bobbert, MF, Huijing, PA, and van Ingen Schenau, GJ. Drop jumping. II. The influence of dropping height on the biomechanics of drop jumping. Med Sci Sports Exerc
6. Bondarchuck, A. Transfer of Training in Sports
. Muskegon, Michigan: Ultimate Athlete Concepts, 2007.
7. Brown, T, Vescovi, J, and Van Heest, JL. Assessment of linear sprinting performance: A theoretical paradigm. J Sport Sci Med
3: 203-210, 2004.
8. Cronin, J and Hansen, K. Strength and power predictors of sports speed. J Strength Cond Res
19: 349-357, 2005.
9. Cummings, B and Finn, J. Estimation of 1 repetition maximum bench press for women. J Strength Cond Res
12: 262-265, 1998.
10. Häkkinen, K and Komi, P. Alterations of mechanical characteristics of human skeletal muscle during strength training. Eur J Appl Phys Occup Phys
50: 167-172, 1983.
11. Hennessy, L and Kilty, J. Relationship of the stretch-shortening cycle to sprint performance in trained female athletes. J Strength Cond Res
15: 326-331, 2001.
12. Komi, P. Stretch-shortening cycle. In: Strength and Power in Sport
. P. Komi, ed. Oxford, United Kingdom: Blackwell, 2003. pp. 184-202.
13. Komi, P and Bosco, C. Utilization of stored elastic energy in leg extensor muscles by men and women. Med Sci Sports Exerc
10: 261-265, 1978.
14. Laffaye, G, Bardy, B, and Taiar, R. Upper-limb motion and drop jump: The effect of expertise. J Sports Med Phys Fitness
46: 238-247, 2005.
15. McBride, J, Blow, D, Kirby, T, Haines, T, Dayne, A, and Triplett, NT. Relationships between maximal squat strength and five, ten, and forty yard sprints. J Strength Cond Res
23: 1633-1636, 2009.
16. Mero, A. Relationship between the muscle fibre characteristics, sprinting and jumping of sprinters. Scand J Sports Sci
3: 16-22, 1985.
17. Read, M and Cisar, C. The influence of varied rest interval lengths on depth jump
performance. J Strength Cond Res
15: 279-283, 2001.
18. Schmidtbleicher, D. Training for power events. In: Strength and Power in Sport
. Komi, PV, ed. Boston, MA: Blackwell Scientific, 1992. pp. 381-395.
19. Street, GM, Mcmillan, S, Board, W, Rasmussen, M, and Heneghan, JM. Sources of error in determining countermovement jump height with the impulse method. J Appl Biomech
17: 43-54, 2001.
20. Verkhoshansky, Y. Special Strength Training: A Practical Manual for Coaches
. Muskegon, Michigan: Ultimate Athlete Concepts, 2006.
21. Walsh, M, Arampatzis, A, Schade, F, and Bruggeman, G.The effect of drop jump starting height and contact time on power, work performed, and moment of force. J Strength Cond Res 18:561-566, 2004.
22. Winchester, J, Nelson, AG, Landin, D, Young, MA, and Schexnayder, IC. Static stretching impairs sprint performance in collegiate track and field athletes. J Strength Cond Res
22: 13-19, 2008.
23. Young, W. Laboratory strength assessment of athletes. New Stud Athl
10: 89-96, 1995.
24. Young, W, Pryor, JF, and Wilson, G. Effect of instructions on characteristics of countermovement and drop jump performance. J Strength Cond Res
9: 232-236, 1995.
25. Young, W, Pryor, JF, and Wilson, G. Relationship between strength qualities and sprinting performance. J Sports Med Phys Fitness
35: 13-19, 1995.
26. Zatsiorsky, V and Kraemer, W. Science and Practice of Strength Training
. Champaign, IL: Human Kinetics, 2006.