Brilliant, Great, Best: What Chess Move Classifications Actually Mean

Move classifications look like the engine is handing out grades for style. They aren't. Every label comes from a simple mechanical question: how much did the evaluation change between the position before your move and the position after it?

That change is measured in centipawns, where 100 centipawns is roughly the value of one pawn. Engines convert that to a win probability, because losing half a pawn when you are completely winning barely matters, while losing half a pawn in a level endgame can decide the game. Classifications are based on that swing in winning chances, not the raw number.

Best simply means you found the engine's top choice. Excellent and Good mean you picked something that gave up so little it makes no practical difference. Book means the position is still in known opening theory, so the engine isn't grading you at all - a database is.

Inaccuracy, Mistake and Blunder are the same idea at increasing severity: a small, a meaningful, and a serious drop in your winning chances. A blunder is not necessarily a hanging piece. Quietly allowing your opponent a strong plan can drop your evaluation just as hard as dropping a rook, and it will be labelled just as harshly.

Miss is the most useful label most players ignore. It doesn't mean you played a bad move - it means a genuinely winning opportunity existed and you walked past it. Your position may still be fine, which is exactly why these are so easy to overlook and so valuable to review.

Brilliant is the one everyone argues about. It is not awarded for any good sacrifice. Broadly, it requires that you gave up material, that the sacrifice is genuinely sound, and critically that a plausible non-sacrificial alternative existed. If taking the piece back was already forced, or the sacrifice was the only legal way to avoid losing, there was no real decision to make - so no brilliancy. This is why two visually identical sacrifices can get different labels: one was a choice, the other was forced.

Great move sits just below that: a clearly best move in a position where getting it wrong would have cost you significantly, usually the only move that holds a position together or the one that finds a decisive tactic.

The practical takeaway is to spend your review time on Mistakes, Blunders and especially Misses. Brilliancies are fun to collect, but they don't tell you anything you need to fix. The moves that cost you points are the ones worth understanding, and understanding why the engine preferred something else is where the improvement actually comes from.