Recently it seems there has been some passionate discussions with “experts” in the field of sprinting (Yessis vs bearpowered @supertraining list). With access to literature both past and present (and some in progress) I feel compelled to enter a new discussion about new sprint training ideas rather than argue about past theories with devoted theoreticists that won’t listen. For that, this site is the best forum. I’d welcome non-board members to join our discussion. An important point is to recognise that literature is the building block to further study and solutions. No one should be shy about putting their ideas forward as we will all grow from the contribution.
Firstly it seems we need to update our sprint modelling. The spring-mass model is a great place to start but the last time I threw a spring on the ground it did not continue to move perpetually in a vertical or preferrably a horizontal direction as close to 100m as possible. More likely the model we need is a bicycle wheel. Throw a bicycle wheel to the ground and it will bounce and might roll forward a little. Spin the bicycle wheel before throwing it to the ground and it will bounce and roll forward considerably and at speed. Take away the rim and the bicycle wheel may not continue to roll effectively. This will depend on whether the spokes are too rigid as some yielding is necessary to allow rotational force to act against the ground. There is an optimum stiffness and yielding for best bicycle wheel modelling performance.
Farley and Gonzalez (1996) showed that there is an optimum stride frequency where stiffness is compromised for force production. At higher stride frequency stiffness increases but force production decreases. At lower stride frequencies stiffness is reduced but force production did not reduce so much. This could suggest that greater stiffness is not required for better sprint performance. In fact, studies have shown that increasing strength (implying capacity for stiffness) does not result in improved sprint performance.
Delecluse et al. (1995) demonstrated that a strength program significantly enhanced strength (p<0.0001) but did not improve 100m performance; 12.45s before training, 12 42s after the training program. However in the same study, plyometric training improved 100-m performance from 12.51s to 12.3s although when taking into account the standard deviation it cannot be considered conclusive. What must be recognised is that the researchers found this improvement occurred within the first 10-m of the race. Which means...
there is still another 90-m of racing that we can devise methods to improve sprinting! In this regard, the Jamaicans are leading the way. Thanks to the 21st century You-tube, information highways etc., we've been able to see and read of some of what they do. One word comes to mind; innovation!
High knees up a slope at high frequency - hip flexor training, resistance training to move the body up the hill, short ground contact time. (Is this regarded as a sprint drill?!)
Rapid PNF stretching (more likely resisting) - conditioning the muscles to resist the stretch and fight back accordingly.
Innovation in the starting blocks - a technique change which must have been deliberately taught to them.
Sprint specific strength training such as the single leg squat - although stated that the weight held in front is for safety, this facilitates hip extensor strengthening and knee rotator strengthening. If you're going to strengthen your muscles then you should strengthen those muscles that rotate you about the knee rather than extend the knee. Watch yourself on video or in the mirror - are you extending the knee or rotating about it when sprinting?