STRIDE ENTROPY McGregor’s most intriguing study so far is one in which he measured and compared entropy levels in the strides of trained and untrained runners. Entropy is essentially a measure of the predictability of the behavior of a system. In the case of the running stride, entropy has to do with how much each individual stride looks like the one before and the one after. McGregor speculated that entropy would decrease as runners approached exhaustionthat reduced entropy would prove to be a signal of fatigue. He anticipated finding that the stride would become more predictable and less varied near exhaustion because such predictability in any system would indicate that system was €œconstrained. In the case of running, it stands to reason that the stride becomes constrained when some component of the stride runs up against a performance limit. To oversimplify, suppose that after sustaining a certain speed for a certain period of time, a runner’s left soleus muscle begins to lose contractile force. This limit becomes a constraint on the whole system of the runner’s stride, making each stride look more like every other stride than when the runner is not fatigued and his stride is not thus constrained, so that it has a little more €œplay. Research has shown that certain individual muscles do fatigue faster than others during running, and that the brain responds to local muscle fatigue by tweaking the entire stride to ensure that it remains properly coordinated within the limit set by that one tired muscle.3 It’s sort of like putting the most tired sled dog at the head of the team to keep the whole unit working together.