To keep this somewhat on topic, I am going to argue Nick’s side here although I think both Nick and Carl are correct and incorrect as the goal of positive acceleration is to attain a higher velocity.
To achieve a higher velocity requires an overload on both components of velocity with respect to step length and rate. Stiffness and shin angles are all fine and dandy, but joint torsional stiffness in transfer applications has to take into account COM alterations by the added mass of the sled pull. The added mass in front because of application of force although the mass is behind. The friction of the surface you pull on is another factor that increases the load and the height of above the load also affect the magnitude of the load you are pulling. I don’t believe I am going out on a limb, but shin angle isn’t going to mean much with a front shifted COM as whole body vertical stiffness is reduced. Application of force is through COM to the center of pressure. The temporal-spatial problems associated with this type of training should be obvious as you may get correct shin angles, but you get incorrect feedback loops that if repeated enough present a problem of the athlete having proper shin angles but improper application of force which is now a learned habit.
For a proper acceleration to occur with a normal human gait pattern. Both the stride length and stride rate should increase with each successive step in initial acceleration (fast component). Once the fast component of acceleration is reached step rate drops and acceleration continues as long as stride length increases depending on stiffness (slow component). I don’t see this with Carl’s sprinter in terms of time durations of this happening although I would need to download the video and put it under intense scrutiny.
Refer to James Hay’s works on stride length and stride rate.
DBandre,
Could you post some video for us to see the difference. Most of us are visual and it would be nice to see what you do with the above details.