I was playing a tournament on Chess.com and received a "Brilliant Move" in the post game analysis. I haven't received this before and was curious of the following.

[FEN ""]

1. d4 d5 2. c4 e6 3. Nf3 c5 4. Nc3 Nc6 5. g3 dxc4 6. e3 Nf6 7. Bd2 a6 8. Ne5 cxd4 9. Nxc6 bxc6 10. exd4 Rb8 11. Bxc4 Rxb2 12. Bb3 Bb4 13. Qc1 Rxd2 14. Qxd2 Ne4 15. Qd3 Bxc3+ 16. Ke2 Qxd4 17. Rhf1 Bxa1 18. f3 Nc3+ 19. Kd2 Nb1+ 20. Rxb1 Qf2+ 21. Kc1 Bf6 22. Ba4 O-O 23. Bxc6 Rd8 24. Rb8 Qc5+ 25. Qc2 Qg1+ 26. Qd1 Qxd1# 
  • What is considered a "Brilliant Move" ?
  • What about the 12. Bb4 move constitutes it being "Brilliant"
    • My logic was to maintain pressure , develop the queen towards the center , trade out a rook when I felt like there was no way I was going to be able to keep it back there. I am quite low ranked (very new to chess outside of the odd game with friends) . I don't see how this is considered a brilliant move as it was just placing pressure , buying little time for future development.

My Thought Process;

    1. Rxb2 - Free Pawn remember hearing about Rooks on the 7th , lets give it a crack
    1. Bb3 - That didn't work out so great, rooks trapped in there, lets hope he moves his black bishop to try and take it so I can keep the attacks up and check him.
    1. Rxd2 - Going to lose the rook to the queen, may as well try and capture a minor piece.
    1. Ne4 - Maintain Pressure on the Queen , can't take my knight without losing his queen, can probably capture his knight put a check in, when in doubt check it out.

Literally nothing about that felt "Brilliant" anyone able to assist with why it was perceived to be good?

  • Just because theres been a little "shade" casted about the He/His etc. I know the other player personally. – David Bateman Aug 28 '20 at 5:18
  • Well, you are locking down both minor pieces with the threat of capture of the other w/ check. You are negating the threat of Ra2 by the tempo gaining Bxb3. The pawn on d4 is hard to defend against Qxd4 and then there is trouble for white. Your rook seems pretty safe, though not invincible on the 7th rank. – Stian Yttervik Aug 28 '20 at 7:58
  • Sorry for the question here - I am puzzled by 16...Qxd4 and 17 Rhf1 - am I missing something obvious, why wouldn't white exchange queens? – Zizy Archer Aug 28 '20 at 9:55

To be honest, brilliant seems like an overstatement for a fairly obvious move. But to answer your question, I think the "brilliancy" is that you actually found the only way to not only save your rook, but even end up winning material.

The point is that after 12..Bb4 there is the threat of 13..Rxd2 after which Ne4 will win the knight on c3.

The only moves stopping black from executing this threat consist of defending the e4 square by playing 13.f3 for example. But then you play 13..Qxd4 after which the c3 knight is under attack. Note that this is the difference between this situation and the situation in which you played Qxd4 right away.

After 13..Qxd4 the knight has to be defended by 14.Rc1. Now black simply castles to bring the other rook into play and there is nothing white can do because of the awkward positioning of its pieces.

Instead of 13.f3 white could also play 13.Bc2 to defend the e4 square, but in that case 13..Qxd4 is crushing as well.

To summarize: I wouldn't say it's a brilliant move (11..Rxb2 would be a more serious candidate as this move requires black to have found the idea of 12..Bb4 and everything after already), but it's definitely strong.

  • yeah i dint think it was brilliant tbh – David Bateman Aug 27 '20 at 8:47
  • @DavidBateman you were facing a low level computer, no? i believe the brilliancy of a move is related to the level of the computer, see my answer. – eps Aug 27 '20 at 18:34
  • @eps It was the the inbuilt chess.com CAPS system i think it has a default depth of 18 – David Bateman Aug 28 '20 at 0:17

Engines are far better at spotting mistakes than at recognizing "brilliancies". You made a move that is much better than all the alternatives which isn't something obvious like a recapture or a fork in one move, so that's probably what the engine identified as "brilliant". Just don't pay too much attention to it and congratulations for this beautiful win!


What is considered a "Brilliant Move" ?

Whenever you move, the computer compares your move to all the moves that are possible. If you made the same decision as the computer would have, you made the best move. However, the computer also examines these moves retroactively. If your move ends up giving you better position than you would have if you made the "best move", it is reclassified as brilliant.

I believe this is all in relation to the level you set the computer to. A low level computer obviously considers more moves brilliant than the highest setting.

  • 3
    That classification of moves seems unlikely, because it could backfire easily. Suppose I capture a piece, but hang my queen in the process. However, my opponent doesn't notice, so on the next turn, I can save my queen: winning material, and doing better than the engine's safer "best move", which did not make the bad capture to begin with. Does that mean my queen-hanging move should be classified as "brilliant"? – Misha Lavrov Aug 27 '20 at 21:04
  • queens weren't hung at any point – David Bateman Aug 28 '20 at 7:56
  • 1
    @MishaLavrov umm, no, I think you misunderstood. The evaluation, not the board, decides the best move. So if you did a horrible move, then followed by a blunder by the opponent, your move wouldn't be scored better than the engine best move. If you did a move the engine did not evaluate (i.e did not think was best) - and it turned out - on recalculating after the fact - that it was better = brilliant. – Stian Yttervik Aug 28 '20 at 8:03
  • @StianYttervik But in that case, what new information would the engine get to recalculate and assign the move a different score? Given the same inputs, the evaluation algorithm will always give the same outputs. – Misha Lavrov Aug 28 '20 at 13:40
  • 2
    @MishaLavrov: The missing piece, I think, is that engines aggressively prune the game tree when they are searching for good moves. For example, if one possible move would hang the queen, and get nothing for it, the engine might only search a few moves after that before giving up and assuming it's a "bad" move. This helps it devote more processing power to more promising moves. So when the engine calls something the "best move," this is a bit of a lie. It's the "best out of all the stuff I looked at," not the "best out of all the stuff I could have looked at." – Kevin Aug 28 '20 at 17:12

The 'brilliant move' indication shows that the engine has missed out on some move, and would not have found this. So you had a general approach that the engine could not evaluate. So consequently, it was evaluated as a brilliant move.

enter image description here

  • 11
    That seems unlikely. Do you have a source for this? – BlueRaja - Danny Pflughoeft Aug 27 '20 at 17:54
  • 3
    Chess.com engine evaluation. – user24344 Aug 27 '20 at 22:18

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.