Thanks for all your feedback ; I’ve been thinking about all this.
I agree that returning 2nd serves is currently a bit too easy, and this has actually been an issue since TE2013. However, there is no simple and clean solution to it.
I don’t want to artificially forbid accelerations on 2nd-serve returns, because in real tennis it is perfectly possible to attack them quite regularly. The difficulty is getting the right proportion: "all the time" would be unrealistic, but "almost never" would be unrealistic too.
It’s also difficult to solve through the general strike physics, because anything that makes it much harder to hit strongly with short preparation time can also make baseline rallies too difficult.
The AutoPos will stay in one form or another. It is needed for nearly everybody except the very best online players (and around 95% of TE4 players don't play or barely play online at all). It also effectively simulates the split-step on serve returns, as in real tennis you don’t simply wait for the ball to leave the racket before starting to react.
That said, one thing I’m going to test is reducing the AutoPos speed specifically on serve returns. On 1st serves, this may not change much, because players normally have to take manual control very quickly anyway. On 2nd serves, though, especially when returning very close to the baseline or slightly inside it, it may make a bigger difference. Currently, AutoPos can sometimes be enough to move the player to the ball without interrupting the strike preparation (but it might not happen often, except maybe if the player stands within the baseline ?).
Anyway, it should lead to a few more off-center contacts, making the serve a bit more effective in general.
I’m not sure yet how strong the effect will be, but this seems like the cleanest first thing to test. I don't know when I'll get to it, though.
