Thank you for the detailed comment!! I honestly think you were onto something…
The main question I take from this is whether the repetition rule should favor the attacking or the defending player. I’ll have to think about that a bit more if I keep working on the game. It does need some kind of repetition rule though, because otherwise the AI can get stuck in long back-and-forth loops.
By the way, yesterday I tested the AI at a search depth of 7 plies (one move took like five minute) because I had the same suspicion about the 3-stack strategy. My impression was that it loses against a stronger player because you overextend on one side, allowing the opponent to counterattack. But in my (chess-heavy) experience BCC has some pretty unintuitive movement patterns , so the right moves can be hard to spot.
You’re also completely right about the undo function! I’ll look into it. :) and maybe add some better feedback about who won haha
Thanks again! Comments like yours make designing a game fun!