Sports Medicine and Orthopaedic


The influence of swing leg motion on maximum running speed

Maximal running speed in humans demonstrates great variability. The speeds of over 12 m/s have been recorded by elite male athletes in competition, which is significantly faster than the reported values for the majority of physically active men and women. This prompts the question of what allows some humans to run so much faster than others.

Clear differences exist between the techniques of faster and slower sprinters, particularly in relation to the swing leg. The hip angle of the swing limb at touchdown of the contralateral leg has been shown to differentiate between elite and sub-elite sprinters and to be the largest single kinematic predictor of maximum running speed, despite exhibiting no differences in angular displacement and angular velocity throughout the stride. Trained sprinters exhibit a more flexed hip throughout the first half of swing, resulting in the thigh of the swing leg being closer to vertical at touchdown of the contralateral leg and closer to horizontal at contralateral toe-off. Furthermore, as running speed increases, the mechanical demands on the muscles of the swing leg, mainly at the hip, substantially increase. It has been shown that muscle volumes of the hip flexors and extensors are extremely large in sprinters when compared with controls. These results indicate that, although the time taken to reposition the limb at maximum speed is similar in faster and slower runners, the kinematics and kinetics of the limb during this portion of the stride are different.

Despite the observed differences in technique between faster and slower runners, there has been limited consideration of the effect swing leg motion might have on maximum running speed. The aim of this study was to use computer simulation to investigate the influence of swing leg motion on maximum achievable running speed.

A computer model was developed to simulate the stance phase of sprinting. Optimisation was used in two ways: to evaluate the capacity of the model to match performance data collected on a sprinter; then to maximise the running speed of the model using two different swing leg techniques, one representative of an elite athlete (SLTE), and the other of a sub-elite athlete (SLTSE).

A male athlete (28 years, 91.1 kg, 1.86 m, 100 m personal best time: 10.50 s) sprinted at 9.7 m/s on an instrumented treadmill recording three-dimensional ground reaction forces (2000 Hz). Sixty-five retroreflective markers were placed on the participant, and their positions were recorded (250 Hz) using sixteen infrared cameras, synchronised with the force data. A custom Vicon BodyBuilder model, composed of the head and trunk, upper and lower arm, thigh, shank, rearfoot, and toes, was written to extract segment and joint angles.

To determine the effect of different swing leg techniques, a publicly available video of an elite athlete and a sub-elite athlete running at their respective top speeds was used, as it illustrated typical differences in technique, with the hip joint of the elite athlete more flexed throughout stance.

A seven-segment planar computer model was constructed to simulate the stance phase of sprinting. The model was evaluated by minimising the difference between simulation output and experimental data collected on the participant. Viscoelastic parameters obtained from this evaluation process were used in all subsequent optimisations.

Optimisations were carried out to establish the maximum running speed achievable by the model whilst angle-driving the swing leg using joint angle time histories obtained by digitisation of each of the two techniques from video data.

The maximum speed of the model when using SLTSE was 9.3 m/s compared with 10.2 m/s when using SLTE. Stance time was slightly shorter and flight time and vertical impulse slightly larger using SLTE. The vertical whole-body CoM displacement did not differ between the two techniques, but the horizontal displacement was greater with SLTE (0.861 m vs 0.814 m).

Using SLTE, the model produced an average of 51 N (0.06 bodyweights) more vertical ground force during stance. This increase was due to higher forces in the first half of stance, with similar forces during the second half giving a net positive effect. When using SLTE, the initial vertical momentum of the stance limb was more negative (-32.0 kg.m.s-1 vs -28.6 kg.m.s-1) which led to a larger vertical impact force peak. Average extensor torques at the knee and ankle were higher using SLTE, whereas the average extensor torque at the hip was lower.

The results showed that using the swing leg technique an elite sprinter can augment the vertical ground force passive impact peak by increasing the negative vertical momentum of the stance leg at touchdown, putting the joints of the stance leg in faster eccentric conditions, placing them closer to their optimum angles for torque production, and allowing more torque and vertical ground force to be produced in the early portion of stance, increase the horizontal displacement of the mass centre during stance, and ultimately increase the maximum running speed. Athletes aiming to increase their maximum running speed should devote time to training associated with the swing phase of running.


Rottier TD and Allen SJ (2021) The influence of swing leg motion on maximum running speed. Journal of Biomechanics.

Link to Paper:


Share this post with your friends

Share on facebook
Share on twitter
Share on linkedin
Share on pinterest