CheckerGames

Is Checkers Solved?

Last updated: June 2026

Yes — checkers is solved. In 2007 a team of computer scientists proved that with perfect play by both sides, a game of standard checkers always ends in a draw. It was, at the time, the most complex game ever solved. Here's what that actually means, how it was done, and why it doesn't ruin the game for the rest of us.

Is checkers solved?

Yes. Standard 8×8 checkers (English draughts) was weakly solved in 2007 by a research team at the University of Alberta led by Jonathan Schaeffer, using a program called Chinook. The result: from the starting position, perfect play by both players leads to a draw. Neither side can force a win against flawless defense.

What does "solved" mean?

A game is "solved" when the perfect outcome is known and can be proven from the starting position. Checkers is "weakly solved," meaning the result with best play (a draw) is proven and a strategy exists to achieve at least that result — but not every possible position has been individually catalogued. In short: nobody, human or machine, can ever beat perfect checkers play; the best you can force is a draw.

How was checkers solved?

The Chinook project ran for nearly two decades. The team combined enormous endgame databases — every position with up to ten pieces, worked out exhaustively — with a search that connected the opening to those solved endings. Checkers has roughly 5×10²⁰ (500 billion billion) possible positions, so the achievement required clever methods to avoid examining them all, plus years of computation across many machines. The proof was published in the journal Science in 2007.

Does solving checkers mean a computer can't lose?

With perfect play, a solved checkers engine can never lose — it can only win (if you err) or draw (if you play perfectly). Chinook famously became unbeatable. Against a perfect opponent it draws every game; against a fallible one it draws or wins, never loses.

Does being solved ruin checkers?

No — and this is the part people get wrong. "Solved" describes perfect play by machines, not the game humans actually play. No person can memorize the solution, so for real players checkers remains a deep, decisive contest where mistakes are punished and skill clearly tells. Knowing the theoretical result is a draw changes nothing about a game between two humans, just as knowing tic-tac-toe is a draw doesn't stop two careless players from losing it.

Is chess solved too?

No. Chess has not been solved and likely won't be for the foreseeable future — it has so many more possible positions than checkers that a full solution is far beyond current technology. Checkers being solved while chess remains open is a good illustration of the complexity gap between the two games. (More in Checkers vs Chess.)

Are the bigger variants solved?

No. The 2007 result applies to standard 8×8 checkers. Larger variants like International draughts (10×10) and Canadian draughts (12×12) have vastly more positions and have not been solved. So the deepest forms of the game still hold genuine mystery even for computers.

Why it still matters

The checkers solution was a landmark in artificial intelligence and game theory — the largest game solved to that point, and a milestone on the road from early game-playing programs to modern AI. It also settled a centuries-old question about the game's perfect result. And for you, sitting down to play? It means checkers is, provably, 100% a game of skill — every outcome is earned.

Play a (provably skill-based) game

Put your own skill to the test against the computer — play now. Want to see why perfect play is so hard to reach? Read the Strategy guide.

Frequently asked questions

Is checkers solved?

Yes. Standard 8×8 checkers was weakly solved in 2007 by the University of Alberta's Chinook team. With perfect play by both sides, the game is a draw — no one can force a win against flawless defense.

Who solved checkers and when?

A research team at the University of Alberta led by Jonathan Schaeffer solved checkers in 2007, using the Chinook program after nearly two decades of work. The proof was published in the journal Science.

Does solving checkers make it boring?

No. "Solved" refers to perfect machine play, which no human can reproduce. Between real players, checkers remains a deep, skill-based contest where mistakes are punished and the better player usually wins.

Is chess solved like checkers?

No. Chess has far more possible positions than checkers and has not been solved; a full solution is considered effectively impossible with current technology.

How many possible positions does checkers have?

About 5×10²⁰ — roughly 500 billion billion. Solving the game despite that number took the Chinook project nearly twenty years of computation.