Resisted sprint (RS) training is commonly used as a means to improve speed and acceleration in running. This method of training may involve the athlete sprinting with an added load using a weighted sledge, a weighted vest, or a speed parachute, or performing uphill or sand dune training. The underlying assumption in RS training is that it will eventually lead to increases in stride length during normal unresisted running, by increasing the power and strength (3,4,14). Although RS training is commonly practiced across a wide variety of sports, inspection of the literature shows limited scientific evidence to support the use of RS training as a method of speed development (4). Various studies have focused on the effect of RS training on the kinematics of sprinting, but there is little research examining the longer-term effects of RS training program on strength and running speed performance.
Lockie et al. (14) investigated the effects of resisted sledge towing on sprint kinematics in field-sport athletes towing weighted sledge loads of 12.6 and 32.2% of their body mass for a 15-m distance. Kinematic analysis of horizontal hip velocity, stride length, ground contact time (CT), and angle of trunk lean showed decreases in mean stride length of approximately 10 and 24% for the 12.6 and 32.2% loads, respectively. Horizontal hip velocity (used to assess horizontal running speed) decreased with the increasing load. Ground CT for the second step also showed an increase of up to 22% compared with unresisted running. Lockie et al. (14) also found that hip extension was not significantly reduced under a resistance sprint condition. Significant changes in hip flexion occurred when a load of approximately 13% of body mass was added to the subject. The increase in hip flexion was assumed to signify an increase in activity of the hip flexors as they attempted to drive the leg forward during recovery. Although these results showed a deterioration in sprint technique and sprint performance with an increase in towing load, other evidence supports the use of RS as a prolonged training method.
Zafeiridis et al. (23) investigated the effects of RS sledge pulling on acceleration and maximum speed performance in running and found that RS with a 5-kg weight significantly improved running velocity from 0 to 10 m and from 0 to 20 m (p < 0.01) compared with unresisted sprint training (US); however, RS training had no effect on running time for the 10- to 20-m section of the sprint for both RS and US. The US training showed a significant improvement for 20- to 40-m sprinting. Zafeiridis et al. (23) conclude that the change for 20 m for the RS group was attributable to the increased running velocity in the 10-m running section and that the improvements in the strength of the knee and hip extensors, along with the gluteus maximus, were likely to have led to the improvement in acceleration performance. Tziortzis et al. (20) studied the effects of assisted and resisted training on mechanical power, maximum velocity, and the performance of 60- and 100-m sprints. They found that the RS group showed a significant increase for both maximum mechanical power (p < 0.001) and absolute cyclic leg velocity (p < 0.05) for the 60- and 100-m RSs after participation in a 12-week resisted training program. Their results also showed a significant decrease, posttraining, in times taken to complete the 60- (p < 0.001) and 100-m sprints (p < 0.001). Kafer and Adamson (12) examined the effects of assisted and resisted training on sprint times for 20, 40, and 60 m using a weighted sledge. The RS group recorded average improvements of 0.08 seconds (p < 0.01) and 0.35 seconds (p < 0.01) in sprint times for 20- and 60-m distances. It was suggested that improvements in performance were attributable to the increased resistance from the sledge training, which resulted in increased force production to develop and maintain velocity. It was speculated that this effect would increase muscle stiffness and vertical force at each ground contact.
Most of the research to date on RS training has focused on the short-term effects on joint kinematics, changes in running speed, and force production in sprinting. Although some studies have examined the longer-term training effects of RS training on sprint performance, few studies have examined the effects of RS training interventions on dynamic aspects of force production such as reactive strength index (RSI), rate of force development (RFD), and ground CTs during jumping. Harrison et al. (9) and Flanagan and Harrison (6) have demonstrated previously that these parameters can be measured effectively using a force sledge apparatus. Therefore, the aim of this investigation was to determine the effects of a 6-week RS training intervention on 30-m sprint times and maximal speed attained in 30-m sprints from a static start and flying 30-m sprints. Additionally, the investigation examined the effects of the RS training on the dynamics of force production in squat jumps (SJs), drop jumps (DJs), and rebound jumps (RBJs) performed on a force sledge apparatus.
Experimental Approach to the Problem
This investigation required subjects to perform speed and strength tests before and after a 6-week RS training intervention. Subjects were randomly selected into a control and intervention group. The control group participated in no form of RS training, and the intervention group performed a 6-week RS training intervention. In all other respects, all subjects completed identical training activities. In the pre- and posttesting, the subjects performed three single-legged SJs and DJ-RBJs on a specially constructed sledge and force platform apparatus. The jumps were done with the dominant leg, giving a total of six jumps. Subjects also performed six 30-m sprints from a static and flying start. The testing was completed in four testing sessions on different days (two testing sessions before and two sessions postintervention). Session one consisted of a 30-m sprint test and sledge jump familiarization session, and session two consisted of a test of DJs, RBJs, and SJs. Dependent variables included times to 5, 10, and 30 m, and maximum velocity in the 30-m sprint and flying 30-m sprint.
Fifteen male subjects, aged 20.5 ± 2.8 years and weight 87 ± 10.5 kg, completed this study. All subjects were professional or semiprofessional rugby players, recruited from a local rugby academy. All subjects had experience of RS training and speed training and were proficient with the techniques of the SJ and DJ. Subjects were all participating in strength and power training at the time as part of their normal training, and the testing was completed during the players' midcompetition period. Ethical approval for this study was granted from the university research ethics committee, and written consent was obtained from all subjects. Before participation, the subjects completed a pretest physical activity questionnaire and were given a volunteer information leaflet describing the study and outlining potential benefits and risks. After consideration of this information, all subjects volunteered and provided informed consent. In total, 24 subjects were recruited for the study, but, because of illness or injury, only 15 subjects (7 control subjects, 8 intervention subjects) completed all parts of the testing and intervention program. The results of the study are based on the data obtained from these 15 subjects.
All sprint trials were performed on a synthetic indoor running track, and a laser measurement device (LAVEG, Jenoptik, Sport) was used to obtain linear distance measures during all trials at a sampling frequency of 100 Hz. This device has been found to produce valid and reliable measures of distance in running trials, with intraclass correlation coefficients (ICCs) ranging from 0.96 to 0.99 (7). The specific procedures used with the laser were consistent with the recommendations of Harrison et al. (7). All raw distance data were filtered using a Butterworth zero-lag, low-pass filter at a cutoff frequency of 3 Hz.
All jumps were performed on a specially built force sledge apparatus. This consisted of a sledge frame, sliding chair, and force platform, and it was similar in design to those used in previous studies (1,8,11). This system has been shown to produce reliable results for jumping activities, with ICCs ranging from 0.95 to 0.98 (9).
All subjects completed a standard warm-up that consisted of 3 minutes of low-intensity jogging and static stretching, which included one exercise for each of the quadriceps, hamstrings, triceps surae, gluteals, and hip adductors, with stretches held for 15 seconds. Subjects then stretched dynamically using squats, leg bounds, and other specific exercises and drills typically used by sprint runners. They also completed two 30-m sprints at 50 and 70% effort with unloaded sledges to complete the warm-up using a sledge towing device as illustrated by Cronin and Hansen (4) (see Figure 1).
Force Sledge Measurements
The testing protocols for the force sledge movements were explained to each subject before the testing procedure. In all jumps, the subjects were instructed to keep their arms across their chests at all times to minimize any contribution to jump impulse by the upper body. Each subject performed a practice trial for each of the movements before performing the test trials. The SJs were performed starting from a 90° knee angle position, and no drop or countermovement was permitted. If any countermovement was detected on force-time display, the subject was required to repeat that trial. In the DJ and RBJ trials, subjects were instructed to minimize CT on the force platform and maximize jump height. The force-time traces for the SJs were analyzed to obtain four dependent variables-namely, maximum RFD (max RFD), time to max RFD, flight time, and starting strength. The start of concentric contraction was defined as the point where the force readings were 10 N greater than the average of the force readings when the subject was static in the SJ starting position (also see Figure 2A). Max RFD was calculated by examining the steepest gradient of the force-time graph in the SJ (19). It was calculated as the greatest rise in force during 5-millisecond periods from the start of the concentric contraction. The difference in the force readings for every 5 milliseconds was found, and this was divided by the time difference between these two readings (t = 0.005 seconds). The 5-millisecond periods were calculated every millisecond from the start of the concentric contraction in the SJ until the instant of peak ground reaction force (GRF). The start of the concentric contraction was defined as the point at which the force readings were 10 N greater than the average of the force readings when the subject was static in the SJ starting position.
Time to max RFD was simply obtained by finding the time difference between the start of the contraction and the beginning of the max RFD. Similarly, time to peak GRF is calculated by finding the difference in time between the point of the first contraction and peak GRF on the force-time curve for the SJ. Ground CT and height jumped (HJ) were calculated for the DJ, and HJ and RSI were calculated for the RBJs. The jumps were performed as one movement sequence of one DJ, followed immediately by three RBJs, but, for analysis, the data were divided into discrete DJ and RBJ phases. The RSI was calculated by dividing the HJ by the ground CT (21). The HJ was calculated from the flight time (FT) for each jump using the equation HJ = (9.81 × FT2)/16 (see Figure 2B).
For all sprinting trials, the laser was mounted on a tripod at a height of 1.37 m and was located 2 m behind the start position. Subjects were required to wear a close-fitting white T-shirt to ensure good reflection of the laser signals. The laser was directed at each subject's lower back. Subjects completed two different types of sprint protocols. In the first protocol, each subject performed three 30-m trials of maximal effort sprints from a static crouch start position. In the second protocol, subjects performed two flying 30-m sprints. In these trials, the sprint distance was a total of 40 m, and the subjects had a 10-m rolling start to allow them to develop acceleration. In all trials, the time to 10, 20, and 30 m was calculated from the filtered 30-m sprint data set. Maximum velocity was obtained from the 30-m instantaneous velocity data set using Microsoft Excel. The distance corresponding to the instant of maximal velocity was also determined.
The Training Protocol
The subjects began with the same standardized warm-up each day before the training session, as described previously. Predetermined loads were added to the sledge at approximately 13% of body mass (BM). Loads on the sledge were determined based on Lockie et al.'s (14) recommendations of 12.6-13% BM, using the following equation: load = ([body mass × body mass percentage] - sledge weight). After the warm-up, subjects performed six 20-m maximal sprints using the weighted sledge, with 4 minutes of passive rest between each sprint. The subjects completed two training sessions per week for a period of 6 weeks, totaling 12 training sessions altogether. After this period, subjects repeated the test procedures. The control group completed both pre- and posttests. They did not perform any type of RS training for the 6 weeks of the intervention period.
SPSS for Windows (release 11.0.1) was used for all statistical analyses. A general linear model analysis of variance (ANOVA) was used to analyze the pre- to posttest differences for each testing session. The general linear model ANOVA had two within-subjects factors-namely, time with two levels (pre and post), and trial (number of trials conducted per testing session). The ANOVA model also had one between-subjects factor-namely, group with two levels (control and intervention). The interaction term group × time was determined, and this was used to indicate whether a significant training effect over time was found in the intervention group compared with the control group. The ANOVA model was applied to all dependent variables, with correction for type I error rates. Effect sizes were estimated using Cohen's dz, which is recommended for repeated-measures designs (2). Interpretation of these effect sizes was based on the recommendations of Hopkins (10).
The results for the DJ are presented in Figure 3. The HJ was considered the performance measure, and statistical analysis indicated a significant interaction effect for group × time on HJ (p = 0.018; see Table 1). These data show that the intervention group increased their HJ, whereas the control group decreased in HJ scores. No statistically significant interaction effect was found for group × time on CT; however, an overall decrease was found in the main effect of time (i.e., pre- to posttest change) for CT across both the intervention and control groups (p = 0.008), indicating that the total experimental population significantly shortened their ground CT scores between pre- and posttests.
Figures 4, 5, 6, and 7 show the results for the SJ analysis on max RFD, time to max RFD, starting strength, and flight time, respectively. Table 2 reports the results of the hypothesis tests on these variables. The statistical analyses revealed no difference in the main effect for time or the group × time interaction effects for max RFD, time to max RFD, or flight time. Starting strength showed no statistically significant main effect for time, but the group × time interaction effect was significant (p = 0.004). Supplementary analyses showed that the mean max RFD of the forwards was greater than that of the backs.
Figure 8 illustrates the results for the RBJ analysis on the HJ for the first RBJ (HJ1). Table 3 reports the results of the hypothesis test on the RSIs for each of the three RBJ trials-RSI1, RSI2, and RSI3-and Table 4 reports the results of the hypothesis tests on HJ1, HJ2, and HJ3. These data reveal a significant interaction effect on HJ in the first trial HJ1. No statistically significant differences were found for the group × time interaction effects for HJ2 and HJ3 or for any of the RSI indices.
Figures 9 and 10 show the pre- and posttest results for intervention and control groups for the 5- and 10-m times in the 30-m sprint from a static start. The p values for time to 5, 10, and 30 m for the 30-m sprint are reported in Table 5. A significant group × time interaction effect was found for the 5-m sprint time. These data indicate that the intervention group subjects significantly improved their 5-m sprint times, with no improvements in the control group. The data also show that similar mean improvements were found for 10- and 30-m times, but these were not statistically significant. Table 6 shows that there were no significant group × time interaction effects for maximum velocity or time to maximum velocity in the 30-m sprint from a static start. Analysis of the results for times to 5, 10, and 30 m for the flying 30-m sprint are displayed in Tables 7 and 8. Whereas significant improvements were observed between pre- and posttests for all subjects for 5-m time, 10-m time, 30-m time, and maximum velocity, no significant group × time interaction effects were found, indicating that the training intervention provided no significant improvements on the flying 30-m sprint scores compared with the control group.
The results of the 30-m sprint tests from a static start showed a statistically significant group × time interaction effect for time to 5 m (p = 0.020). The effect size classified this as a large effect size (Cohen's dz = 1.21). The results therefore, reveal that start and acceleration capabilities during the first 5 m were improved by the RS training intervention. The results show no improvement in maximum velocity from pre- to posttests between groups, indicating no training effect overall for a 30-m sprint; therefore, the RS intervention was more effective than normal rugby training in producing start and acceleration capabilities but not for improving maximum speed. The improvements in 5- and 10-m times and in maximum velocity in the RS group from pre- to posttest showed that most subjects improved, whereas only some of the control group subjects did not improve. Improvement in control group subjects could occur because subjects in both groups were involved in speed training drills once weekly as part of their normal training. This suggests that different forms of training can cause improvements in sprinting, although the evidence of this study indicates that RS training provided additional benefits to the training undertaken by the control group. These training effects are important for rugby players because research has shown that rugby players typically perform sprints of 10-20 m in a match (5). Forward players commence sprints from a standing start most frequently (41%), with back players evenly distributed between standing, walking, and jogging starts. An increase in this initial acceleration may provide the player sufficient power to break through tackles and make territorial gains in a match situation.
The results of the flying 30-m tests show that both groups improved from pre- to posttest in maximum velocity, 5-m time, 10-m time, and 30-m time, but the lack of significance in group × time interaction effect indicates that the RS training group did not gain any additional benefit compared with the control group. The comparison of flying 30-m sprint with the static 30-m sprint results further confirms that the benefits of RS training were on the acceleration capabilities of athletes. These results agree with those of Spinks et al. (18), who found that RS training improved acceleration mechanics and recruitment of the hip and knee extensor muscles.
The results of the SJ analysis indicate a significant group × time interaction effect for starting strength (p = 0.004), and the data show that the RS training group improved starting strength postintervention, whereas the control group got worse (see Figure 6). Starting strength is considered essential in sports where great initial speed is necessary for optimal performance, such as boxing, fencing, and karate (16). The validity of isolating starting strength has been corroborated by electromyographic research and confirms the suggestion that starting strength is, in part, determined by the innate qualities of the neuromuscular apparatus, particularly the ratio of fast- to slow-twitch fibers in the muscles (17). Young (21) has commented that starting strength is regarded as a measure of very fast force production capabilities and found that the initial acceleration phase (0-2.5 m) is highly correlated (r = 0.86) with the force applied in a concentric-only SJ. Specialization of the neuromuscular system to develop starting strength is determined chiefly by the magnitude of external resistance (17). Research has shown a correlation between starting strength and initial acceleration (22), and the results of this study provide further evidence of the link between starting strength and improved acceleration in the early part of the sprint.
The results of the DJ analysis show significant reductions in CT in both groups pre- to posttest (p = 0.008), but there was no significant group × time interaction effect. These data show that RS training provided no additional improvement in CT compared with the control group training. In contrast, a significant group × time interaction effect was found for HJ in the DJ test, and these data show that the intervention group increased HJ, whereas the control group decreased HJ. Although this indicates a training effect attributable to the intervention, caution should be exercised when interpreting the results. Figure 3 shows that the significant interaction effect was attributable more to the decrease in HJ in the control group as opposed to an increase in HJ for the intervention group. The results suggest that the RS training provided a stimulus for maintaining stretch-shortening cycle (SSC) functioning throughout and postintervention in comparison with the control group.
The RBJs were assessed to evaluate the effect of RS training on rhythmic SSCs. Komi and Gollhofer (13) have argued that natural movements such as hopping and running represent better models of an SSC; therefore, RBJs on the force sledge apparatus are functionally more closely related to running than SJ or DJ. The results of the RBJ analysis show a possible training effect on initial RBJ performance (p = 0.035), but RS training did not seem to have a positive effect subsequent RBJs. The RBJ test results show no significant group × time interaction effect on RSI. The RSI has been used by practitioners (15) and researchers (21) to gauge an individual's SSC abilities. McClymont (15) has commented that RSI testing provides an effective and useful tool in the preparation of elite athletes. The RSI is determined by the ratio of HJ/CT; therefore, it is likely that RSI is functionally related to leg-spring stiffness because HJ is dependent on impulse generated and CT is directly related to the magnitude of the countermovement in an SSC activity. Increased leg-spring stiffness has been linked to sprint performance (9), so it is reasonable to assume that RSI would also be linked with sprinting performance. From the RBJ test results, it seems that RS training provided few additional benefits compared with the control group training. These results are consistent with the sprint test results because they suggest that RBJ performance is related more with maximum running speed than starting ability or acceleration.
Spinks et al. (18) have reported that an 8-week RS training program (a) significantly improved acceleration and leg power performance but was no more effective than an 8-week nonresistance sprint training program, (b) significantly improved reactive strength, and (c) had minimal impact on gait and upper- and lower-body kinematics during acceleration performance. Spinks et al. (18) conclude that RS training would not adversely affect acceleration kinematics in running. The results of this investigation broadly support the findings of Spinks et al. (18) and provide evidence that RS training improves the initial acceleration phase of sprinting from a stationary position. In contrast to the findings of Spinks et al. (18), the results of this study suggest that RS training provides an additional benefit compared with nonresistance training. It is not clear whether the improvements observed in the RS training group were merely attributable to the additional training load performed by this group or to a specific adaptation to the RS exercises. To evaluate this, a third experimental group would have been required; however, practical constraints did not permit this. Further investigation comparing training volume-matched groups is recommended.
The results of this study provide a good rationale for the use of RS training as part of a structured conditioning program for rugby players. The study has shown that RS training can improve initial acceleration in sprinting from a static start. Improvements in time to 5 and 10 m in field-based tests and starting strength in laboratory-based tests can be attributed to the RS training intervention, confirming a significant increase in muscle force development. The findings support the use of a training load of approximately 13% of body mass for a running distance of 15-20 m, as previously recommended by Lockie et al. (14) and Zafeiridis et al. (23). The use of this load seems to ensure that an athlete's acceleration mechanics are sufficiently overloaded to develop increases in acceleration performance during a 30-m sprint from a static start. The ability to accelerate quickly from a stationary position will provide a competitive advantage for rugby players as well as athletes in many other sports.
1. Avela, J, and Komi, PV. Interaction between muscle stiffness and stretch-reflex sensitivity after long-term stretch-shortening cycle exercise. Muscle Nerve
21: 1224-1227, 1998.
2. Cohen, J. Statistical Power Analysis for the Behavioral Sciences. New York: Academic Press, 1977.
3. Costello, F. Training for speed using resisted and assisted methods. Natl Strength Cond Assoc J
7: 74-75, 1985.
4. Cronin, J and Hansen, KT. Resisted sprint training for the acceleration phase of sprinting. Strength Cond J
28(4): 42-51, 2006.
5. Duthie, GM, Pyne, DB, Marsh, DJ, and Hooper, SL. Sprint patterns in rugby union players during competition. J Strength Cond Res
20: 208-214, 2006.
6. Flanagan, E and Harrison, AJ. Muscle dynamics differences between legs in healthy adults. J Strength Cond Res
21: 67-72, 2007.
7. Harrison, AJ, Donoghue, O, and Jensen, RL. A comparison of the reliability of video and laser
techniques for determining displacement and velocity during running. Meas Eval Phys Educ Exerc Sci
9: 219-231, 2005.
8. Harrison, AJ and Gaffney, SD. Effects of muscle damage on stretch-shortening cycle function and muscle stiffness control. J Strength Cond Res
18: 771-776, 2004.
9. Harrison, AJ, Keane, SP, and Coglan, J. Force-velocity relationship and stretch-shortening cycle function in sprint and endurance athletes. J Strength Cond Res
18: 473-479, 2004.
10. Hopkins, WG. A scale of magnitudes for effect statistics. In: A New View of Statistics
. Available at: www.newstats.org/effectmag.html
. Accessed April 20, 2007.
11. Horita, T, Komi, PV, Nicol, C, and Kyröläinen, H. Effect of exhausting stretch-shortening cycle exercise on the time course of mechanical behaviour in the drop jump: possible role of muscle damage. Eur J Appl Physiol
79: 160-167, 1999.
12. Kafer, R, and Adamson, G. Methods for maximising speed development. Strength Cond Coach
2: 9-11, 1994.
13. Komi, PV and Gollhofer, A. Stretch reflexes can have an important role in force enhancement during SSC exercise. J Appl Biomech
13: 451-460, 1997.
14. Lockie, RG, Murphy, AJ, and Spinks, CD. Effects of resisted sledge towing on sprint kinematics in field-sport athletes. J Strength Cond Res
17: 760-767, 2003.
15. McClymont, D. 2003. Use of the reactive strength index (RSI) as a plyometric monitoring tool. Available at: http://www.coachesinfo.com/index.php?option=com_content&view=article&id=146:rugby-rsi&catid=47:rugby-general&Itemid=77
. Accessed November 5, 2008.
16. Schmidtbleicher, D. Training for power events. In: Strength and Power in Sports
. P.V. Komi, ed. Oxford: Blackwell, 1992. pp. 381-395.
17. Siff, M. Biomechanical foundations of strength and power. In: Biomechanics in Sport
. V. Zatsiorsky, ed. London: Blackwell Scientific, 2000. pp. 103-139.
18. Spinks, CD, Murphy, AJ, Spinks, WL, and Lockie, RG. The effects of resisted sprint training on acceleration performance and kinematics in soccer, rugby union, and Australian football players. J Strength Cond Res
21: 77-85, 2007.
19. Tidow, G. Aspects of strength training in athletics. New Stud Athletics
1: 93-110, 1990.
20. Tziortzis, S, and Paradisis, GP. The effects of sprint resisted training of the peak anaerobic power and 60 m sprint performance. In: First Annual Congress, Frontiers in Sport Science, The European Perspective
. P. Marconnet, ed. Nice, European College of Sport Science, May 28-31, 1996. pp. 88-89.
21. Young, W. Laboratory strength assessment of athletes. New Stud Athletics
10: 88-96, 1995.
22. Young, W, McLean, B, and Ardagna, J. Relationship between strength qualities and sprinting performance. J Sports Med Phys Fitness
35: 13-19, 1995.
23. Zafeiridis, A, Saraslanidis, P, Manou, V, Ioakimidis, P, Dipla, K, and Kellis, S. The effects of resisted sledge-pulling sprint training on acceleration and maximum speed performance. J Sports Med Phys Fitness
45: 284-290, 2005.