Ultimate Tic Tac Toe is played on nine tic-tac-toe boards arranged in a 3 × 3 grid.
Rules: Playing on a spot inside a board, determines the next board in which the opponent must play their next move. The goal is to win a three in a row boards. You can play your next move at any board if you are directed to play in a full board or a board that has been won.
If the last rule is excluded then the game has been solved and has a winning strategy for the first player (paper, implementation). That is to say, if a player is directed to play at a board that has been won but is not full yet, she does not get a wildcard move, instead she must play at that already-won board.
With the rule that directing to a won board allows the opponent to play at any board, the game is currently unsolved. This was mentioned on the site some years ago.
I known of computer implementations of the game that use Monte Carlo or similar methods. Are there known strategies that a human can apply and use? Something like favorable or recommended openings for 1st player, and responses for 2nd player?
- Super-Human agent (AlphaZero-like) at https://www.uttt.ai/