Arm swing remains a rather controversial topic relative to the mechanics of speed. Some researchers believe that arms provide lift and help the runner maintain a more constant horizontal velocity. However, the same researchers point out that, at least relative to their research on distance runners, the arms do not provide any direct forward propulsion.

Further, the point out that there does not appear to be any advantage in the ‘classical style’ of swinging the arms directly forward and backward.

The following clip gives an interesting analysis of what is really going on in the sprint cycle relative to arm mechanics. In this clip, the sprinter is running over nine meters per second. His contact time is actually less than eleven-hundredths of a second. During that time, tracings at the elbow show that the arm moves forward less than 32 cm, about twelve inches. However, the athlete rotates over his grounded foot a full meter, and that is consistent with Dr. Weyand’s findings on step length for all the subjects in his JAP speed paper. The speed of the arm and the speed of the forward rotating trunk are the same: 9.14 meters per second.

If the rate data is indeed correct, for arms to assist in ‘propulsion,’ which really doesn't occur once athletes are up to speed, they must do so in a very brief period of time (.09-.11), over a very short amplitude (32 cm), and at a speed equal to that of the body rotating forward over the grounded foot (step length).

Perhaps this explain why Mann and Hermann concluded that arms are simply used for balance and do not play a significant role in dictating the quality of running performance. This is certainly not a very popular concept among the today's elite speed coaches.