Score
0
Best
-
The classic combinatorial game Domineering on a 5×5 grid. You place dominoes standing upright, the AI places them lying sideways, and whoever has no room left for their next domino loses. No luck, no draws — a pure fight over space.
The board is a 5×5 grid of empty squares. Each domino covers exactly two neighbouring squares. You only ever place your dominoes upright, covering one square and the square directly above or below it; they are orange. The AI only ever places its dominoes sideways, covering two squares next to each other in a row; they are blue with stripes. You always move first, and pieces never move once placed.
To place a domino, press on the board in the column you want. A dashed preview of your domino appears under your finger: press on the upper half of a square to pair it with the square above, or on the lower half to pair it with the square below. Keep your finger down and slide to move the preview, then lift to place it. Slide off the side of the board before lifting to cancel. The AI's most recent domino is outlined in white so you can see what it just did.
The two numbers next to "You" and "AI" count how many places each side could still put a domino right now. Squares that neither side can ever use again are shaded with stripes, so the open space is easy to read.
On your turn you must place a domino if you can. The first player who has no room left for a domino of their own direction loses immediately — there is no passing and the game can never be drawn. If the AI has no sideways room after your move, you win; if you have no upright room after the AI's move, you lose.
Beat the AI and the next game starts with your win streak growing; your score is that streak. Every win also makes the AI more careful — it makes fewer loose moves the longer your streak runs. Lose once and the streak is over; you can save it to the Hall of Fame from the result card.
Domineering is a race for room, so count it. Every upright domino you place covers squares in two rows at once, which can wipe out up to four of the AI's sideways spots, while it only uses up one or two of your own. Before each move, look at the two counters and prefer moves that cut the AI's number the most while costing yours the least — the counters update after every move so you can check the effect.
The most valuable spots are the ones the opponent can never take away. An upright pair of squares is safe for you when neither square can be part of any sideways domino — for example because the squares beside it are already filled or it sits against the edge with a blocked neighbour. Such a spot is a guaranteed future move. Do not rush to fill your safe spots: save them for later, and spend your turns on contested areas where your domino also blocks the AI.
Think about the edges and about splitting the board. A narrow vertical strip one column wide is perfect for you and useless for the AI, while a horizontal strip one row high is the opposite. Placing dominoes so that the empty space breaks into separate pockets lets you count each pocket on its own: how many moves you get there, how many the AI gets. When the totals favour you, you have won even if the board still looks open.
Domineering was invented by Göran Andersson in the early 1970s and became famous through Martin Gardner, who wrote about it under the name Crosscram; it is one of the central examples in the theory of combinatorial games. On a 5×5 board the second player wins with perfect play, whichever direction moves first, and this AI does solve the position — so as the first player, you win only by exploiting its loose moves, and those become rarer with every game in your streak.