Ok, so first an observation I've made regarding how to serve properly. I've often heard it said that the shoulders should be square with the net at the same time that the arm reaches full extension. The idea is that alignining the timing of the shoulder/chest turn with the timing of the arm extension allows for most efficient power transfer to the ball. One of the posters here (jollyroger) even posted side by side comparison photos of his serve vs. another poster's serve which illustrated this beautifully. This makes intuitive sense, however it seems (at least on the face of it) to contradict a principle of whip cracking. One of the things that allows the kinetic chain to manifest such high speeds is the conservation of angular momentum. Suppose you have three straight pieces of wood, and you connect them together end by end, but allow them to rotate within a certain range of motion. They can rotate 180 degrees in one direction, but once they align up with each other, they cannot rotate any more. See picture below: The dashed lines indicate the limits of the "joint angles". Now suppose the black portion weighs more than the blue portion, and the blue portion weighs more than the red portion. We can now crack this whip as follows: We simply rotate the heavy black handle, and then allow the momentum to channel to the next link (the blue one). If the black handle stops moving, all that momentum gets channeled to the blue link. But because the blue link is lighter, it will move faster than the black one did, due to conservation of momentum (mass * velocity). By the time the red one moves independently (due to the black and blue one having reached their joint limits), it's much faster than the original black one. This principle can be used in cracking the whip of the arm, but my question is whether it should also be used to channel momentum from the torso to the arms. If so, then wouldn't it make sense to have the shoulders and chest square with the net/target before the arm starts to crack? So long as you prevent the torso from overrotating, and "apply the brakes", then that momentum will be channeled into the arm. Why doesn't serve instruction encourage this delayed cracking of the whip principle between torso and arms? p.s. here is link to a good physics paper on the physics of whips: http://math.arizona.edu/~goriely/Papers/2003-PhysD(whip-waves).pdf I've only skimmed it, so my understanding is limited.