Why is this move "Brilliant"?

From a weak player's perspective, it's way easier for me to see that there is a pin on the g7 bishop and that my e4 pawn is weak. I would be much more likely to play d5 to temporarily protect e4 and to allow me to eye the pinned bishop next turn. Just a guess, please dont bite my head off if that was a dumb comment.

Thanks but I'm already a weak player, so your input is not needed  Just kidding.

By the way I checked your profile and you're on an 11 game win streak! Nice job! Keep it up

Not dumb at all. But I think you underestimated the strength of the White attack after 22.Rxf7! kxf7 23.Rf1+ 

Haha thanks I've been trying to learn how to play better, weak players have to stick together! Either way nice move whether it was "brilliant" or not. 

It is quite a good attack. After 23. ...Kg8, the best move was dxc5, threatening mate with a tempo on black's d pawn, but I instead went d5?, closing the board and making the position even.... I guess I thought I could take black's pawn en passant after e6, without realizing my queen was hanging. Then I went on to lose the game.

Me too! The first time I looked at the position I thought d5(?) was killing. Luckily, I looked again!

I probably am underestimating the attack, I usually do. I ran the position through stockfish11 after playing d5 and it said Qc8 was the best move for black, then Qg5 for white, but that's as far as i looked into it. But the "not dumb" comment made my day so thank you for that!

True, d4 is the first move that comes to mind. But as it turns out, it loses to e5 which neutralizes the white bishop. This is exactly what happens later in move 24. With Rxf7 on the other hand, white gets a strong attack and black has to give material to hold. This becomes clear only after several moves. If Rxf7 only traded material it would not be called brilliant by the engine. Pretty cool that it detected that.

I annotated the game here. Have fun.
https://www.chess.com/a/2S2diD5K4STfk

I didn't see how easy it was to shut down the bishop again. The annotation was very helpful, explaining the thought process. Thinking about this position and trying to find a weaker than brilliant move was actually very instructive for me.

This might not work either just curious, what if white takes the e5 pawn on e6 en passant, attacking the rook?

Then he loses his queen on g5.

Lol I completely missed that, but thanks for entertaining the question.

I guess the logic can easily produce apparent false positives since it triggers whenever the engine realizes that a move is better than it originally thought. This can for example be caused if it suddenly sees an additional move combination that occurs many moves in the future and wins additional material. Of course, the immediate benefit of the move (i.e. taking a piece) is unchanged. Usually, whenever the engine marks a move as "brilliant" it is a good idea to take a deeper look at the variant the engine computed.

In ordinary human speech, what meaning does the word "brilliant" convey to you, when applied to a chess move?

It doesn't really matter. Obviously the chess engine can't actually determine if a move was "brilliant" in the sense you're talking about. But it's just a title.

Thank you for annotating this game! And thanks for allowing me to view the analysis with a depth of 30. I thought my move 19. fxg6 was a nice move, as did the analysis in depth 20, but your deep analysis showed quite the opposite--turns out it was a mistake. Moreover, the best move your deep analysis suggested was actually an inaccuracy according to depth 20.

You are welcome. I am glad the analysis was useful. The higher depth values are a recent addition of chess.com and finally make more accurate analysis possible. If you do not have premium (but have a lot of time) you can always do a "self-analysis" using your own machine.

I have Lucas Chess downloaded, but any depth beyond 15 takes an unreasonably large amount of time. I have no idea how the difference in time is so big between it and chess.com's analysis. Is there a chess analysis engine available for download that can analyse games in great depth in a reasonable amount of time?

"Is there a chess analysis engine available for download that can analyse games in great depth in a reasonable amount of time?"

You bet! Stockfish 11 is available for free download now. https://stockfishchess.org/