Here is some relevant research. Please note that the first two studies only assess differences in speed and not safety or training efficacy of running on different slopes.
The optimal downhill slope for acute overspeed running.Ebben WP.
Dept of Physical Therapy, Marquette University, Milwaukee, WI 53201, USA.
PURPOSE: This study evaluated a variety of downhill slopes in an effort to determine the optimal slope for overspeed running. METHODS: Thirteen NCAA Division III college athletes who participated in soccer, track, and football ran 40-yd (36.6-m) sprints, on downhill slopes of 2.1 degrees , 3.3 degrees , 4.7 degrees , 5.8 degrees , and 6.9 degrees in random order. All sprints were timed using the Brower Timing System Speedtrap II. Data were analyzed with SSPS 15.0. A 1-way repeated-measures analysis of variance revealed significant main effects for the test slopes (P = .000). Bonferroni-adjusted pairwise comparisons determined that there were a number of differences between the hill slopes. RESULTS: Analysis reveals that 40-yd sprints performed on hill slopes of approximately 5.8 degrees were optimal compared with flatland running and the other slopes assessed (P < .05). Sprinting on a 5.8 degrees slope increased the subjects’ maximal speed by an average of 0.35 s, resulting in a 6.5% +/- 4.0% decrease in 40-yd sprint time compared with flatland running. Compared with the 4.7 degrees slope, the 5.8 degrees slope yielded a 0.10-s faster 40-yd sprint time, resulting in a 1.9% increase in speed. CONCLUSIONS: Those who train athletes for speed should use or develop overspeed hills with slopes of approximately 5.8 degrees to maximize acute sprinting speed. The results of this study bring into question previous recommendations to use hills of 3 degrees downhill slope for this form of overspeed training.
The same study, but with more participants:
1: J Strength Cond Res. 2008 May;22(3):898-902.Links
Effect of the degree of hill slope on acute downhill running velocity and acceleration.Ebben WP, Davies JA, Clewien RW.
Department of Physical Therapy, Program in Exercise Science, Marquette University, Milwaukee, Wisconsin, USA. webben70@hotmail.com
This study analyzes the effects of hill slope on acute overspeed running. This study considers both acceleration and supramaximal velocity. Forty-four athletes ran 40-yard sprints, on 5 different hill slopes, ranging from 2.1 degrees to 6.9 degrees . Forty-yard sprint times and 10-yard split times were recorded using the Brower Timing System Speedtrap II. Analysis reveals that 40-yard and 10-yard sprints performed on hill slopes of approximately 5.8 degrees were optimal compared to flatland running and the other slopes assessed. Sprinting on a 5.8 degrees slope increased the subjects’ maximal speed by 7.09% +/- 3.66% and increased the subjects’ acceleration by 6.54% +/- 1.56%. Strength and conditioning professionals who train athletes for speed should develop and use overspeed hills or platforms with slopes of approximately 5.8 degrees in order to maximize acute sprinting velocity and acceleration.
Something on resisted sprints:
J Strength Cond Res. 2008 May;22(3):890-7.Links
Effects of three types of resisted sprint training devices on the kinematics of sprinting at maximum velocity.Alcaraz PE, Palao JM, Elvira JL, Linthorne NP.
Kinesiology and Biomechanics Laboratory, Department of Physical Activity and Sport Sciences, Universidad Católica San Antonio de Murcia, Guadalupe, Murcia, Spain. palcaraz@pdi.ucam.edu
Resisted sprint running is a common training method for improving sprint-specific strength. For maximum specificity of training, the athlete’s movement patterns during the training exercise should closely resemble those used when performing the sport. The purpose of this study was to compare the kinematics of sprinting at maximum velocity to the kinematics of sprinting when using three of types of resisted sprint training devices (sled, parachute, and weight belt). Eleven men and 7 women participated in the study. Flying sprints greater than 30 m were recorded by video and digitized with the use of biomechanical analysis software. The test conditions were compared using a 2-way analysis of variance with a post-hoc Tukey test of honestly significant differences. We found that the 3 types of resisted sprint training devices are appropriate devices for training the maximum velocity phase in sprinting. These devices exerted a substantial overload on the athlete, as indicated by reductions in stride length and running velocity, but induced only minor changes in the athlete’s running technique. When training with resisted sprint training devices, the coach should use a high resistance so that the athlete experiences a large training stimulus, but not so high that the device induces substantial changes in sprinting technique. We recommend using a video overlay system to visually compare the movement patterns of the athlete in unloaded sprinting to sprinting with the training device. In particular, the coach should look for changes in the athlete’s forward lean and changes in the angles of the support leg during the ground
A study arguing for contrast training:
J Strength Cond Res. 2006 Nov;20(4):767-77.Links
The effects of sprint running training on sloping surfaces.Paradisis GP, Cooke CB.
Track and Field Unit, Department of Sport and Exercise Science, University of Athens, Athens, Greece. gparadi@phed.uoa.gr
The aim of this study was to examine the effects of sprint running training on sloping surfaces (3 degrees ) on selected kinematic and physiological variables. Thirty-five sport and physical education students were randomized into 4 training groups (uphill-downhill, downhill, uphill, and horizontal) and a control group, with 7 participants in each group. Pre- and posttraining tests were performed to examine the effects of 6 weeks of training on the maximum running speed at 35 m, step rate, step length, step time, contact time, eccentric and concentric phase of contact time, flight time, selected posture characteristics of the step cycle, and peak anaerobic power performance. Maximum running speed and step rate were increased significantly (p < 0.05) in a 35-m running test after training by 0.29 m.s(-1) (3.5%) and 0.14 Hz (3.4%) for the combined uphill-downhill group and by 0.09 m.s(-1) (1.1%) and 0.03 Hz (2.4%) for the downhill group, whereas flight time shortened only for the combined uphill-downhill training group by 6 milliseconds (4.3%). There were no significant changes in the horizontal and control groups. Overall, the posture characteristics and the peak anaerobic power performance did not change with training. It can be suggested that the novel combined uphill-downhill training method is significantly more effective in improving the maximum running velocity at 35 m and the associated horizontal kinematic characteristics of sprint running than the other training methods are.
A study arguing against the exclusive use of resisted sprints or overspeed:
J Strength Cond Res. 2006 Nov;20(4):833-7.Links
Velocity specificity in early-phase sprint training.Kristensen GO, van den Tillaar R, Ettema GJ.
Human Movement Sciences Programme, SVT, NTNU, Norway.
A comparison of resistance running, normal sprint running, and supramaximal running was performed. Nineteen young, generally well-trained subjects were divided into 3 training groups: resistance, normal, and supramaximal groups. Resistance and supramaximal training was done using a towing device, providing extra resistance or propulsion forces, resulting in running speed differences of about 3.3% (supramaximal) and 8.5% (resistance), compared to normal sprinting. The training period was 6 weeks, with 3 training sessions per week (5 sprint-runs over 22 m). Running times were measured using photocells, and average step length and cadence were recorded by digital video. A small (0.5%) but significant (p < 0.05) overall pre-post difference was found in running velocity, but the 3 groups changed differently over the running conditions. All individual subjects improved sprinting velocity most on the trained form, at 1-2% (p < 0.001), and thus, the principle of velocity specificity in sprint training was supported. This indicates that to obtain short-distance sprinting improvement in a short period of time, one may prefer normal sprinting over other training forms.
joel is thelad i coach lol won the 1 and 2, tbh with ya we do this session about once a month, we did last sat a session of 2-30m harness resistance followed by 2-60m light tow, followed by to flying 30m with 40 build ups then he had the week off with some strides on thursday, wasnt happy with his starts at the weekend in all his races but the times were pb’s so cant complain too much, he now got the u23 this weekend so did some starts tonight and only ligt warmup and strides on thursday