Dani21 & Jirafalox,
I'm still having a hard time understanding your explanation, because the ball trajectory before the bounce hasn't changed at all, and the AutoPos already predicts the trajectory after the bounce to position the returner correctly, both before and after the update.
So I don't see why the old twist would have been harder to read, or why the new one would automatically lead to better returns. Are you starting to press a strike button after the bounce ?
I actually tested the old and new bounces side by side, alternating between them on every serve, and my observations are quite different from yours.
For example, on the backhand side, the old twist behaved more like a sliced serve, with the ball coming towards the returner's body. Now, the ball moves away from the body after the bounce, so the returner has to position himself differently from what the initial curve might suggest.
And a wide twist serve from the advantage court, with a right-handed server against a right-handed returner, pushes him outside even more !
Of course, some serve placements have become less effective, while others have become more effective. That's the logical consequence of changing the bounce direction, and it doesn't mean the twist has been nerfed overall.
As for the bounce height, it didn't change. The maximum height is the same; only its location on the court has shifted slightly.
Dani21BTW, I checked the first 2 games of the Bug Report you sent me yesterday, and the only twisted 2nd serve I saw was actually a service winner against a ~2300 Elo player. Of course, that's just one example, but it doesn't exactly support the idea that the twist has become "basically useless".

So I'm really struggling to understand what you're experiencing. Maybe you're still using your usual serve placements, which no longer produce exactly the same results ?
Regarding the updates, there's no single overarching goal of improving the online experience (reminder : online players are a tiny minority of the player base). My goal is to improve the game bit by bit, as I've been doing for many years now.
Jirafalox,
Your explanation of the twist becoming easier to read is also difficult for me to follow, since the trajectory before the bounce is exactly the same as before. In fact, the new bounce can make the ball move in a direction that is less consistent with its initial curve, not more.
And please, let's keep the discussion focused on the twist serve. If you want to discuss baseline rallies, their speed or their balance, feel free to open a separate topic. Mixing several unrelated gameplay issues into the same discussion makes it much harder to analyze them properly. 🙂
Dani21 & Jirafalox,
That said, I'm considering making twist serves a bit more difficult to return by increasing the aiming imprecision caused by their spin : this would make returning a twist more demanding without changing the bounce physics or preventing aggressive returns.
However, I'm not convinced this is necessarily a good idea yet. It could make the twist by far the best 2nd serve, especially compared with the flat serve (the sliced serve already combines lateral displacement with an aiming penalty).