Ultimate Tic-Tac-Toe takes the game you already know and makes it genuinely hard to think about. Instead of one 3x3 board, there are nine of them arranged in a larger 3x3 grid. You win a small board by getting three in a row. You win the game by getting three small boards in a row. The catch is that every move you make determines which board your opponent must play in next.
That single rule changes everything. Sending your opponent to a board that is almost full gives them a free choice. Sending them somewhere you are close to winning hands them the chance to block you. Every cell on every board has consequences two levels deep.
The game spread online around 2013 and has appeared in maths classrooms and puzzle communities as an example of nested strategy. It is significantly harder to solve than regular tic-tac-toe, and no complete optimal strategy is publicly known.
The game is sometimes called Super Tic-Tac-Toe or Meta Tic-Tac-Toe. It became widely known after Ben Orlin's 2013 article popularised it online, and has since been used in computer science courses to teach search algorithms and game trees. Regular tic-tac-toe has 255,168 possible games and was fully solved decades ago. Ultimate Tic-Tac-Toe has not been solved, and belongs to a different category of difficulty entirely.
On your first turn you can play anywhere. After that, the cell you play within a small board determines which small board your opponent must play in next. The position of your cell maps directly to a board in the larger grid. If your opponent sends you to a board that is already won or full, you get to play anywhere.
Win three small boards in a row, horizontally, vertically, or diagonally, to win the match. A small board is won by getting three of your symbols in a row within it, just like regular tic-tac-toe.
The strongest opening is the centre cell of the centre board. It sends your opponent to the centre board, where the middle cell is the most valuable, and you have already taken it. Most beginners ignore this entirely.
Avoid sending your opponent to a board where they can win immediately. Before placing your piece, check where your move will send them and what they can do there. Boards that are nearly full give your opponent a free turn, so avoid filling boards without winning them.
Think two moves ahead: not just what you are doing, but where you are sending your opponent, and where they will send you back.
Corner boards are safer to fill than edge boards. Controlling a corner contributes to two diagonal macro lines, and opponents frequently underestimate diagonal threats while focusing on rows and columns. Winning all three boards in a diagonal is often the fastest route to victory.
Keep track of which boards are nearly full. If a board has only one or two cells left, it is about to become a free-turn board. Sending your opponent there gives them a free turn. Being aware of which boards are about to run out lets you plan where you want your opponent to end up.
If the small board you are sent to has already been won or is completely full, you get a free turn and can play on any open cell on any open board. Free turns are powerful because they let you choose the most strategically valuable position on the entire grid rather than being restricted to one small board.
Yes. If all nine small boards are decided and neither player has three in a row on the macro board, the player who won more small boards wins. If both players won the same number of small boards, the game is a draw. Draws are uncommon but possible, particularly when both players play defensively.
The centre cell of the centre board is widely considered the strongest opening. It sends your opponent to the centre board, where you have already claimed the most valuable cell, and gives you strong macro positioning from the start. Many experienced players open here automatically.
In regular tic-tac-toe, each move affects only one board and one level of play. In Ultimate Tic-Tac-Toe, every move controls where your opponent plays next, which affects an entirely separate board, which in turn affects the macro result. The consequences of each decision extend two levels deep, which makes planning significantly harder and intuition much less reliable.
The Hard AI uses a heuristic scoring function that evaluates each legal move for positional value, macro threats, where the move sends the opponent, and whether the opponent can win a small board on their next turn. It is strong enough to challenge most players but can be beaten with careful, deliberate play.