How a Move Gets Labelled Brilliant
The brilliant label is not the engine's opinion. It is a rule about sacrifices that a site applies afterwards, and knowing the rule explains every strange one you have seen.
· 3 min read
Stockfish has no concept of brilliance. It produces a move and a number, and that is the entire output. The gold exclamation mark comes from a layer built on top, applying rules that the site chose and can change whenever it likes.
Knowing those rules removes most of the mystery, including why your best move of the year got nothing.
The usual conditions
Implementations differ, but the family resemblance is strong. A move is typically labelled brilliant when all of these hold:
- It sacrifices material, giving up a piece or an exchange that could be taken.
- It is the best or nearly the best move available.
- The position remains good for you afterwards, so it is not a desperate throw.
- The material is not immediately recovered by force.
- You were not already winning easily, since giving up a rook while up a queen proves nothing.
Some versions add a further condition: that the natural alternative would have lost a meaningful part of your advantage. That makes the sacrifice necessary rather than merely available, which is closer to what the word means.
Why obvious brilliancies get nothing
| Your sacrifice | Why it was skipped |
|---|---|
| Wins the material back next move | Nothing was given up |
| One of two moves that both work | Not required |
| Played from a winning position | Threshold not met |
| Second best, though it wins | Not the top move |
| A quiet move, no material given | Sacrifices only |
That last row catches the largest category. Many of the finest moves in chess are quiet: a king step, a rook lift, a pawn that costs nothing and wins twenty moves later. Those are frequently harder to find than a sacrifice, and no common rule set will ever mark one, because material has to change hands.
Why odd moves get the label
The reverse happens too. A sacrifice you played on instinct, without seeing more than two moves, gets a gold badge because the rule does not ask whether you understood it. In a scramble, giving up an exchange to open a file can satisfy every condition while being, honestly, a guess.
The label describes the move. It does not describe you.
What it is worth
Treat it as a bookmark rather than a grade. A brilliant marker points at a position where material was given up and the engine approved, which makes it a reasonable place to stop and work out why. That is genuinely useful, and it is the whole of the value.
If you want to know which of your moves were hardest to find, the honest measure is different: look for moves that were best by a wide margin over the second choice, whether or not anything was sacrificed. Those are the positions where finding it mattered, and no badge will point you to them.
The labels around it
Brilliant sits in a family of positive markers, and the others are worth knowing because they are frequently confused with it.
Great move usually means the move was best by a clear margin over the second choice, with no sacrifice required. This is often the more meaningful award: it marks a position where finding the right move actually mattered, which is precisely what brilliant does not measure.
Best move means it matched the engine's first choice. In a position with one legal recapture, everybody gets one.
Book move means the position is still in the opening database, so no judgement is being offered at all.
A game full of best-move markers and no great ones was a game with no difficult decisions. That is worth knowing about your own play, and it is visible in the summary without opening a single position.
A test you can run
Take a game where you got a brilliant marker and set the position up on an analysis board. Remove the sacrifice from consideration by looking at the second and third choices instead.
If those moves keep most of the evaluation, the sacrifice was optional and the marker is decoration. If they throw the advantage away, the sacrifice was forced by the position and you found the only path, which is the version worth being pleased about.
The report will not distinguish those two cases for you. Two minutes on the board will.
Common questions
- What makes a chess move brilliant?
- Most implementations require that the move gives up material, that it is the best or near-best move, and that the position stays good afterwards. Some also require that a plausible alternative would have thrown the advantage away, so the sacrifice was necessary rather than decorative.
- Why did I not get a brilliant for a good sacrifice?
- Usually because the sacrifice was not required, another move kept the same evaluation, or the material came straight back within a move or two so nothing was genuinely given up.
Try it on your own position
The analysis board runs the same engine in your browser, free. The app adds what a web page cannot: reading a position straight from a photograph, saved boards, and analysis with no connection.
