I understand the evolution of relay passing. The push is a modified downsweep. The elbow articulates in two directions so I don’t recognize the push as being significantly different, at least from the standpoint of the incoming runner.
Mach advocated a relatively low, straight arm for the outgoing athlete. Effectively he was simply reversing the hand position of the outgoing runner and changing the elbow articulation of the incoming runner. This, in turn, created an extremely “hard” target. The natural position of the hand when grabbing an object is facing downward. Anatomically it’s a very basic concept that is overlooked. Additionally the low are position of the outgoing runner presents a poor target for the incoming runner. There is no differentiation between the hand and wrist. Finally as stated in the previous post, the exchange window is now reduced and cannot take place anywhere but behind the outgoing runner.
US teams, whether they are collegiate or National have attempted to mitigate some of these issues by forcing the outgoing runner to raise his exchange arm to an unnaturally high position. This does give a better target, but it causes a series of other issues. The target window is reduced even further and the acceleration pattern of the outgoing runner is disrupted, often severely (ie: Patton).
The upsweep does have the issue of baton management (how’s that for a euphemism 
). Interesting so did the US yesterday so I don’t think that the downsweep/push completely eliminates this issue. As for practice time required I am not at all sure that I would agree. What I would say is that since most nations have switched to the downsweep/push there have been far more drops and other fumbles.
Free distance is a false construct. It was created as a means of justifying the downsweep/push. The quest for this has ignored the primary goal of baton speed, which is after all the prime objective.