Same clock, different stride: how elite sprinters accelerate

The paper: Clark, K.P., Noble, T.J., Stewart, H.E., Meng, C.R., Skoufos, L., Ryan, L.J., & Chapman, R.F. (2026). Spatiotemporal acceleration strategies in elite American male and female sprinters: between-group similarities and individual differences. Sports Biomechanics. doi.org/10.1080/14763141.2026.2688484

This study measured how 24 of the fastest sprinters in the world accelerate over the first 10 metres. Its central finding is worth holding onto: at the elite level there is no single acceleration template. Athletes of near-identical ability reach the same times through measurably different mechanics.

The study

Twenty-four elite American sprinters — 12 men (100 m best 9.89 s) and 12 women (10.87 s), with 51 World and Olympic medals between them — were filmed accelerating from the blocks through a calibrated 10 m zone. A single high-speed camera recorded each run at 240 fps, and every step was digitised by hand to extract four spatiotemporal measures: step length (ground covered per step), step rate (steps per second), contact time (how long the foot stays on the ground) and flight time (time airborne between steps).

The authors set out to answer two questions: how elite men and women differ over the first 10 m, and whether athletes of the same standard accelerate alike.

What they found

The male–female difference is one of step length, not turnover. The men reached 10 m faster than the women, but the advantage came almost entirely from longer steps. Step rate, contact time and flight time were near-identical between the sexes.

Athletes of the same standard accelerate differently. Within each group, sprinters reached the same times through different combinations of the four measures, along two broad strategies. Step-length-dominant sprinters — striders — cover ground with long steps at a lower turnover, and spend marginally longer on each ground contact. Step-rate-dominant sprinters — spinners — take shorter steps at a higher frequency, with more time spent airborne.

Illustrative of the study’s finding: two elite sprinters can reach 10 m at the same instant through opposite mechanics. Numbers are representative of the ranges the study reports, not two specific named athletes.

Because running speed is the product of step length and step rate, both strategies arrive at the same velocity. Across the sample, athletes occupied the full range between the two. The one variable that held constant from athlete to athlete was timing: contact and flight times varied remarkably little.

Strengths and limitations

Strengths

• A genuinely elite sample — 51 World and Olympic medals is a cohort rarely available to researchers.
• The entire 0–10 m acceleration captured step by step, for both sexes.

Limitations

• One ruler for the whole frame. Every step is scaled by a single pixels-per-metre figure, derived from two markers — one at 0 m, one at 10 m. A side-on camera does not image the run at a constant scale: steps toward the edges of the frame, and any stride that drifts off the calibrated line, are foreshortened by perspective and distorted by the lens. A single scale factor cannot correct for this — it applies the wrong conversion to those steps, and does so silently. Two points on a line calibrate a line, not a whole image.
• The reported “0.02 m accuracy” is flattering. It reflects only how repeatably those two markers can be re-clicked, not the true error on a step measured elsewhere in the frame, which is larger.
• Twelve athletes per sex is a small sample, and each athlete’s strategy is defined relative to the other 23 — a different roster would shift the boundaries.
• The data come from a single practice session rather than competition, so they may not represent an athlete’s best or their race-day pattern.
• The study is descriptive, not experimental: no athlete altered their strategy, so it establishes that different approaches exist at the top, not that changing one’s own would help.

Practical takeaways

• There is no single correct acceleration profile. A step-rate-dominant sprinter is not a deficient step-length-dominant one; both are valid solutions to the same problem.
• Running speed is step length × step rate. An athlete can develop either lever, so coach the profile the athlete has rather than imposing a preferred one.
• Timing is shared; stride is individual. Contact time, flight time and step rate stay consistent even among the best, so the meaningful individual variation sits in step length versus step rate. That is where assessment should focus.
• Normative data is a reference range, not a target. It cannot identify a given athlete’s profile — only measuring their own step length, step rate and contact time can. Tracking those across a season shows whether training is developing an athlete’s strengths or eroding them.