Brilliant vs Great Move in Chess
Both labels mark a standout move, but they test different things. In our review a brilliant move is a sacrifice that works, and a great move is the only good move. No rulebook defines either, so each tool sets its own.
There is no shared definition
“Brilliant” and “great” are names that review tools give to moves. Chess itself does not define them. Chess.com’s help center describes its own version: “Brilliant Moves are always the best or nearly best move in the position, but they are also special in some way.” It adds that a brilliant move is “when you find a good piece sacrifice”. For a great move it says these are moves that were “critical to the outcome of the game”, such as “finding the only good move in a position” (Chess.com help center, read 28 October 2026).
That is one site’s wording. We do not know how another tool decides, and we do not claim to match any of them. Below are our own conditions, which you can check in the code.
Our conditions side by side
Win% is your winning chance from the engine’s score. The “deep pass” is the second, slower look at key moves.
| Label | What must be true |
|---|---|
| Brilliant | Deep pass checked it. Best move, or it loses under 2 Win% points. A real sacrifice. Win% after the move is 30 to 97. Win% before the move is at least 15. |
| Great | Deep pass checked it. It is the engine’s best move. The second-best move is at least 15 Win% points worse. Not a recapture and not an answer to a check. |
| Best | It is the engine’s best move, or it loses under 1 Win% point. No sacrifice test and no second-best test. |
Brilliant is tested first, then great, then the rest. So a move that fits both is called brilliant. A sacrifice here means material you could lose by exchange, and the opponent cannot simply hand it back.
How rare are they?
We ran 12 games, 584 half-moves in total, through the engine. Two moves were brilliant and 31 were great. Both brilliant moves came from historic games. This is a tiny sample, so do not read it as a statistic about chess. The point is that brilliant should be rare, and great showed up more often in our set.
A brilliant move: Lasker against Bauer, 1889
17.Bxg7 is a bishop sacrifice. The engine rates it best, and White's winning chance goes from 81.4% to 82.4%. White gives up the bishop without losing any winning chance, and the engine does not see a forced win yet, so the move stays inside our 30 to 97 range.
A great move: Anderssen against Dufresne, 1852
13.Qa4 is quiet, with no sacrifice. White goes from 55.1% to 56.5%. The label comes from the alternative: the runner-up move leaves White far worse, as the note under the board shows.
Open the line chip under each board and play it on the board. Each example came from the same engine you would use on your own game.
How to use the labels
Do not chase them. A brilliant or great label marks a move where the engine saw something that stands out, and it is a prompt to look at the position yourself. Your mistakes are the other side of the review, so read the blunder checker page as well. To see the labels on your own game, paste it into the game review, or run the leak report over many games.
Questions
Is a great move better than a brilliant move?
Not in our rule. A brilliant move is best or nearly best, and must also be a sacrifice. A great move is the engine’s best move, with the next-best move at least 15 Win% points worse.
Why is a queen sacrifice that mates not brilliant?
When a position is already a forced mate, Win% is 100 before and after the move. Our brilliant rule needs the Win% after the move to be 97 or lower, so such a move is labelled great instead. More on that case.
Will I see these labels in my own games?
Sometimes great, rarely brilliant. In the 12 games we ran, 2 moves were brilliant and 31 were great. That is a small sample, so treat it as a rough feel only.
Can a move be great in one tool and not in another?
Yes. Each tool sets its own conditions, so compare labels inside one tool, not between tools.