The ability to efficiently accelerate and reach maximum running velocity is essential for athletic success. Training for enhanced speed is a key element in most strength and conditioning programs and typically develops 2 essential components of speed: acceleration and velocity. Acceleration is defined as the rate of change in velocity and is often measured by assessing sprint performance over short distances, such as 5 or 10 yards (16). Velocity refers to the rate of movement over a specified distance and is commonly measured by 40-yard sprint performance (6,16).
Improved acceleration and velocity is achieved by increasing the physical, metabolic, and neurological components associated with sprinting (9). Specific techniques used to enhanced speed include resisted or assisted running (5,9,10,24). Assisted methods include high-speed treadmill training, towing, and downhill sprinting (9,10) and are thought to increase running velocity by improving stride length or stride frequency (11,14,18,21).
Anecdotal observations and research suggest that assisted methods such as downhill sprinting help produce acute supramaximal running velocity (4,9,10,12,13,15,17,19), defined as the velocity greater than that which can be achieved without assistance at a 0° slope (10). Anecdote as well as evidence also suggest that training at a supramaximal running velocity may improve chronic speed (1,9,10,21-23). These studies have typically used downhill slopes of approximately 3°. For example, in one study, subjects were filmed and timed during a 40-meter sprint on a flat land as well as at a 3° uphill and downhill slope. Results indicate that downhill running yielded approximately 5.4% faster sprint times, compared to flatland running, lending support to the idea that downhill running has a positive acute effect. According to this study, this improvement was accrued as a result of greater stride length (13). Theoretically, it is possible that downhill running results in improved acute running velocity as a function of gravity. Questions remain as to whether or not this acute improvement is manifested chronically and in nondownhill conditions. Some evidence suggests that training in downhill conditions improves sprinting performance, suggesting that the benefit of this form of training is not related to gravity-mediated improvements in velocity, since stride rate improves with this type of training (21,22). Training studies have evaluated aspects of downhill sprinting, using slopes of 3.0° or 3.3°. Of these studies, 2 examined the effect of a combination of uphill and downhill sprinting at slopes of 3°, with subjects demonstrating improved running velocity compared to controls (1,20). Paradisis and Cooke (21) indicated that subjects who trained with a combination of uphill and downhill as well as downhill running on a slope of 3° resulted in statistically significant improvement in posttest 35-meter sprint performance compared to those who trained on flat land. While the evidence suggests that combined uphill-downhill running or downhill running alone may improve speed, no evidence exists to determine the optimal slope to use for such training.
The majority of literature on downhill sprinting has focused on velocity, with little information addressing the effects of such training on acceleration, despite the belief that acceleration, rather than maximal speed, is more important for successful athletic performance (5). Consequently, training should enhance both acceleration and velocity (16). Therefore, the purpose of this study was to determine the optimal slope for acceleration and velocity and to determine the point of diminishing return at which the slope becomes suboptimal.
Experimental Approach to the Problem
A randomized repeated-measures research design was used in order to assess a variety of hill slopes to determine the optimal downhill slope for acceleration and velocity. This study tests the hypothesis that a downhill slope exists for optimal acute acceleration and running velocity, and that that slope is greater than the slopes of approximately 3°, which were previously described in the literature. Independent variables included 5 different hill slopes that were believed to be within the optimal range, based on the literature and pilot testing. Dependent variables included electronically measured 10- and 40-yard sprint times.
Subjects included male athletes (n = 44) with a mean age of 19.8 ± 1.4 years. The majority of subjects participated in sports, such as soccer, football, or track, at the high school or NCAA division III level. The subjects were informed of the test procedures prior to participation in the study and provided written informed consent for the study. The study was approved by the institution's office of research compliance.
The test slopes were measured using a calibrated #4-21 digital slope indicator (R&B Manufacturing, Riverside, MO). According to manufacturer specifications, the digital level is accurate within 0.1°. The level was attached to a 6-foot long 2 times 4 board. A variety of hillslopes were assessed in order to find those within the range desired for this study. The digital level attached to the 6-foot board was placed on the hillslope, and slope measurements were recorded for every 6-foot increment for the entire 40 yards of measured slope, resulting in twenty 6-foot measurements. Each slope was then determined by the average of 18 total measurements, since the highest and lowest measurements were omitted to account for variances in surface consistency. This process was repeated until the following slopes were found. As a result of these measurements, slopes of 2.1°, 3.3°, 4.7°, 5.8°, and 6.9° were used to test the subjects, since they represented a variety of increments over a range of slopes thought to be optimal according to the literature (1,10,15).
All subjects were familiarized with the test procedures during a pretest orientation session. Prior to testing, subjects were warmed up and performed running-specific dynamic stretching exercises. Subjects then completed 2 submaximal sprints at approximately 75% and 90% of maximal speed. Subjects rested for 5 minutes and then ran 7 sprints, with the order determined by a random integer generator with 5 minutes rest between each sprint. These sprints included 2 sprints at 0°, averaged to determine baseline sprint times, and 1 sprint on each of the 5 test slopes. All sprints were performed during calm weather with wind conditions of less than 5 miles per hour on grass that was cut to a standardized 2.0 inches.
All experimental sprint distances were exactly 40 yards and determined using a fiberglass measuring tape. Sprint time was determined using a Brower Timing System Speedtrap II (Brower, Salt Lake City, UT) (Figure 1). This timing system is an infrared laser timing device that is accurate within 0.01 seconds, according to manufacturer specifications. The timing system was triggered when the subject removed his fingers from the pressure pad at the beginning of the sprint and stopped when the athlete triggered the final laser sensor at 40 yards. Ten-yard split times were also recorded with electronic timing. All sprint starts were volitional with the starting form standardized across subjects and slopes (Figure 2). Subjects were encouraged to put forth an equal effort for all trials.
The data from the investigation are presented as mean ± SD. The statistical analyses were undertaken with SPSS 15.0 for Windows (SPSS, Inc., Chicago, IL) using a one-way, repeated measures ANOVA to test for main effects. Assumptions for linearity of statistics were tested and met. Significant main effects were further analyzed with Bonferroni adjusted pairwise comparison of within-subjects differences in sprinting performance for each of the hillslopes. The criterion for significance was set at an alpha level of p ≤ 0.05.
Significant main effects were found for both the 10-yard and 40-yard sprints (p < 0.05). Post hoc analysis revealed a number of differences between the slopes for each distance assessed (Tables 1 and 2). Results indicate that the 5.8° slope yielded a 6.54% ± 1.56% decrease in 10-yard split time, representing the largest difference in acceleration compared to 0° (Figure 3). Results further indicate that the 5.8° slope yielded a 7.09% ± 3.66% decrease, representing the largest difference in 40-yard sprint times compared to 0° (Figure 4). Thus, 5.8° represents the optimal downhill slope of those assessed, for both acceleration and velocity. The 6.9° slope produced less optimal results compared to the 5.8° slope, as evidenced by a 0.09 ± 0.08 second greater 10-yard sprint time and a 0.30 ± 0.17 second greater 40-yard sprint time, compared to the 5.8° slope. These findings suggest that the 6.9° slope represents the point of diminishing return, where the slope becomes too steep and sprint times significantly increase.
The present study was the first to evaluate the optimal slope for acute downhill acceleration and velocity. The main finding of the study was that hill slopes of 3.4°, 4.0°, 4.8°, and 5.8° resulted in acute improvements in both 10- and 40-yard sprint times. The downhill slope of 5.8° yielded the optimal 10- and 40-yard sprint times compared to all other slopes assessed, suggesting it is the slope of choice for acute downhill sprint training. The 5.8° slope resulted in running times that were 7.09% faster than the normal condition of 0°. This finding indicates that supramaximal training may be effective at slightly greater intensities than previously recommended. Mero and Komi (15) reported that attempting to train at speeds above 106% of maximal velocity result in an uncontrollable increase in stride length, which increases the breaking phase of each ground contact, resulting in a decreased stride rate. In the present study, the 6.9° slope resulted in a significant increase in 10- and 40-yard sprint times, suggesting that the 6.9° slope is too steep for acute downhill sprint training, which may be the result of changes in technique and increased breaking forces to prevent falling, as previously proposed (3,10,15).
It is interesting to note that only one athlete demonstrated an optimal 40-yard sprint time on the 6.9° slope. This subject was an NCAA division III national qualifier as a sprinter. This case suggests the possibility that elite athletes may require a steeper slope (greater than 5.8°) for downhill speed training.
Previous research by Bissas et al. (1) indicates that acute speed is enhanced when subjects run downhill at a 3.0° slope, a finding which is partly confirmed by the present study. Similarly, training studies have supported the idea that downhill running may be useful for chronic adaptation as well (21), though this finding was not supported by Sullentrop (23). However, subjects in this study trained only twice a week with 6 downhill sprints per training session, suggesting that there may be a requisite volume threshold (23).
All pervious research examining downhill sprinting used slopes of approximately 3.0°, perhaps as a result of early reports (7,8), or research that indicated that 3.0° may be an effective slope (13), if not optimal. Previous research that assessed subject performance changes with hill slopes of 3° resulted in approximately 1.1% (21) to 4.6% (22) improvements in chronic sprint times. Results of the present study indicate that previous research and anecdotal recommendations for the use of approximately 3.0° hill slopes were less than ideal. Future research and training should be conducted at nearly double the slope that has been used historically.
In the present study, performance times were somewhat slower than anticipated, for all trials, possibly do to the length of the grass; the nature of the starting mechanics, including the use of a pressure pad sensor to activate the timing device; and the use of electronic timing, which typically results in times that are 0.22 seconds slower than handheld stopwatches (2). Nonetheless, the grass, wind direction (less than 5 miles per hour), and starting mechanics were uniform across all test conditions. Results of this study are most likely generalizable to athletes who are similar to the high school and college athletes who participated in this study. Subjects in this study were not specifically trained with downhill sprinting, and as athletes become acclimated to this form of training, slopes greater than 5.8° may become optimal.
The results of this study indicate that downhill sprint training and research training studies should be conducted on slopes of approximately 5.8°. Slopes of 3.4°, 4.0°, and 4.8° were also effective for acute downhill running, if not optimal. Downhill sprinting performed on a slope that is too shallow or steep results in a suboptimal acute training stimulus. Practitioners may want to increase the intensity of this form of training by progressing from slopes of approximately 3.4° to slopes of approximately 5.8°. Downhill sprint training should be individualized, and practitioners are advised to frequently time athletes' performance in order to determine if the optimal slope changes with increased experience with this form of training. Slopes can be determined using a commercially available slope level using the techniques described in the methods section. The results from the present study also should be used to guide the design and development of overspeed hills and platforms. Future research will help determine if training on the optimal hill slope will result in greater chronic adaptations as well as determining if athlete skill level and gender affect the optimal hill slope.
This study was funded by a Marquette University College of Health Science Faculty Development Research Grant. The authors would like to thank Jeff Groth for his technical assistance in determining hillslopes.
1. Bissas, A, Paradisis, G, and Cooke, C. The effects of uphill-downhill sprint running on maximum speed and acceleration rates. In: 7th Annual Congress of the Proceedings of the European College of Sports Science
. Koskolou, M, ed. Athens, Greece, July 24-28, 2002. pp. 24-28.
2. Brechue, W, Piper, F, and Mayhew, J. Differences between hand timed and electronically timed sprints in college football players. NSCA National Conference
, July 6-9, 2005, Las Vegas, Nevada. pp. 87.
3. Cissik, J. Means and methods of speed training, part one. Strength Cond J
26: 24-29, 2004.
4. Costello, F. Training for speed using resisted and assisted methods. Nat Strength Cond Assoc J
7: 74-75, 1985.
5. Cronin, J and Hansen, K. Strength and power predictors of sports speed. J Strength Cond Res
19: 349-357, 2005.
6. Ebben, W. A review of football testing and evaluation. Strength Cond J
20: 42-47, 1998.
7. Facciono, A. Resisted and assisted methods for speed development
: part one. Strength Cond Coach
1: 10-11, 1992.
8. Facciono, A. Resisted and assisted methods for speed development
: part two. Strength Cond Coach
1: 7-10, 1992.
9. Facciono, A. Resisted and assisted methods for speed development
. Strength Cond Coach
1: 10-11, 1993.
10. Facciono, A. Assisted and resisted methods for speed development
: part two-resisted speed methods. Modern Athl Coach
32: 3-6, 1994.
11. Hunter, J, Marchall, R, and McNair, P. Interaction of step length and step rate during sprint running. Med Sci Sports Exerc
36: 261-271, 2004.
12. Ito, A, Komi, P, Sjodin, B, Bosco, C, and Karlsson, J. Mechanical efficiency of positive work in running at different speeds. Med Sci Sports Exerc
15: 299-308, 1983.
13. Kunz, H and Kaufmann, D. Biomechanics of hill sprinting
. Track Tech
82: 2603-2605, 1981.
14. McFarlane, B. A look inside the biomechanics and dynamics of speed. Nat Strength Cond Assoc J
9: 35-41, 1987.
15. Mero, A, Komi, P, and Gregor, R. Biomechanics of sprint running: a review. Sports Med
13: 376-392, 1992.
16. Murphy, A, Lockie, R, and Coutts, J. Kinematic determinants of early acceleration in field sport athletes. J Sport Sci Med
2: 144-150, 2003.
17. Otason, J. Sprinting
ala Finland. Modern Athl Coach
13: 2-5, 1975.
18. Ozolin, N. How to improve speed. Track Tech
44: 1400, 1971.
19. Ozolin, N. Athlete's training system for competition. In: The role and sequence of using different training-load intensity. Bondarchuk, A, ed. Escondido: Sports Training, Inc., 1971. pp. 202-204.
20. Paradisis, G and Cooke, C. Kinematic and postural characteristics of sprint running on slope surfaces. J Sport Sci
19: 149-159, 2001.
21. Paradisis, GP and Cooke, C. The effects of sprint running on sloping surfaces. J Strength Conditioning Res
20: 767-777, 2006.
22. Paradisis, G, Cooke, C, Newton, J, and Bissas, A. Sprinting
performance and uphill-downhill training. In: XVth Congress of the International Society of Biomechanics
. Häkkinen, K, ed. Jyvaskyla, Finland, July 2-6, 1995. pp. 714-715.
23. Suellentrop, J. Variation of Russian downhill sprint training for selected college students. Microform Publications, Univeristy of Oregon. Eugene, OR. 1981.
24. Verhoshansky, Y. Recommended methods of speed development
in elite athletes. In: Sprints and Relays: 5th edition
. Javer, J, ed. Mountain View: TAFNEWS Press; 2000. pp. 79-82.
Keywords:© 2008 National Strength and Conditioning Association
assisted training; overspeed; speed development; sprinting