IAAF 100m study


Changes in the step width, step length, and step frequency
of the world’s top sprinters during a 100 m race
Ito, A.1), Ishikawa, M.2), Isolehto, J. 2), and Komi, P.V. 2)

  1. Osaka University of Health and Sport Sciences, JAPAN
  2. Neuromuscular Research Center, Department of Biology of Physical Activity,
    University of Jyväskylä, FINLAND

The coaches usually instruct that the narrow step width and the high step
frequency are needed to obtain high performance in the sprint start. However, the
experimental data regarding the step width during a 100 m race has not been well
documented. The purpose of this study was to clarify the changes in the step width, step
length, and step frequency of the world’s top sprinters during the sprint acceleration
phase and for the phase when the full stride length had been achieved. It was thought
that this information would be valuable for the coaches and athletes alike, because, the
measurements were taken during the men 100 m finals of the IAAF world
championships in 2005.
The subjects were 18 male sprinters who participated in the heats of the 10th
World Championships in Athletics held in Helsinki, Finland (2005). They were divided
into two groups based on race times: the high performance group (HG; 10.12–10.32 s, 9
sprinters) and the lower performance group (LG; 10.40–10.9 s, 9 sprinters). Two video
cameras were set up on the spectator stand. One of them focused was placed so that it
recorded the foot contacts with the track surface covering the initial acceleration phase
of the 0 to 30m. The second camera was set the cover middle phase of the race so that it
gave the measures of the longest stride length achieve the race. It was assumed that
60m center point for the camera with the optical range from 40 to 80m would provide
this information. Two-dimensional Direct Linear Transformation method was applied to
the x-y coordinates of the runners’ toe during each visible foot contact period.
Consequently, the analysis provided the accurate measures of the step length, step
frequency and step width (Fig.1). These values were then used to compare the two
groups of the sprinters.


As expected, the step length increased gradually during the early phase of the
sprint acceleration (Fig. 2). The patterns of the increase were similar in both groups.
However, the better sprinter group (HG) had longer step length during this acceleration
phase. The same was true for the full stride length that was, on the average, 0.12 ± 0.03
m, longer (p<0.003) in HG as compared to LG. The step frequency was maintained at
almost the same level (4.56 ± 0.16 steps /s; Fig.3) in the starting dash and the full stride
phases, and no difference was observed between the groups. While the sprint running
velocity increased during the acceleration phase, the distance (60m) when the full stride
length had been achieved in both groups demonstrated that the HG group had already
reached higher velocity (see Fig. 4). The step width demonstrated no difference between
groups, but it decreased in all runners from 0.39 ± 0.07 m in the 1st step after the start
to 0.17 ± 0.04 m in the phase of the full stride length (p < 0.001; Fig 5).
These results indicate that the wide step width may be suited for developing the
driving force during the long foot contact period under the acceleration phase such as
the sprint start. On the other hand, the narrow step width may be suited for developing
the driving force during the short foot contact period under the fast velocity condition
such as the full stride sprinting. However, the full mechanisms of the optimal step
changes need further clarification.
Suggestions for coaching
The results obtained could suggest the following advices (Fig. 6). Coaches should
advice sprinters 1) to concentrate not only to reach a higher step frequency in the sprint
start, but 2) to utilize the longer step already from the beginning of the start; and finally
3) it could be advisable that the step width will be maximized during the first steps and
then gradually decreased from the about 0.4 m (1st step) to about 0.17 m (the full stride).

In one of the charts they show that all the sprinters hit the max number of strides right out of the blocks. Is that correct? Does anyone have a number for stride frequency number for 10m out of the blocks. I would like to compare to the number strides that the study has to someones real life experience.


They’re saying stride rate is almost the same in starting accel and top speed. Then again, almost only counts in horseshoes and hand grenades not biomechanical analysis:)

Do you think stride rate would start out much lower then increase more? Maybe it would in lower level athletes (10.90+).

The problem with this study is that it doesn’t measure the most important kinetic variables. Stride lengths and frequency are a product of the ground impulse & forces. Research findings on these topics would provide more meaning insight. Knowing the cause of step length/frequency variations is the key.

The findings of this study have little practical applications to coaches. Advising athletes to lengthen step lengths is counterproductive. Biomechanical analysis is a vulnerable tool. However, the applications of research can be incorrect.

My is guess that it would be because of a lack of strength or elastic strength. But on the other hand weaker athletes cant drive and accelerate for too long so they may hit top speed/top stride rate quickly too.

I would have rather seen the differences in flexibility, weight room strength and elastic strength between all the athletes. I think that’s the reason the stride widths are different.

I have found for 10m to be done in 6-8 strides.