2 brilliant (!!) moves in the same game

I think I've never had a brilliant move before (In the games I've analyzed so far) and here I had to in 2 the same game : 

https://www.chess.com/analysis/game/live/6441078127?tab=report

If someone could explain to me the "brilliancy" in these moves would be great, especially 25.Nc3  where I just saved my knight from a certain death 

Anyway so sad I managed to blunder despite all my brilliancy 

Especially weird since the evaluation does not change much when you play the "briliant" moves.

I think the evaluation isn't supposed to change if you play the best move. On my end I don't see it changing after playing these moves.

The evaluation shouldn't change if you play the best move and the computer foresaw that move.  I thought "brilliant" meant so good even the computer didn't see the move, in which case it should change the evaluation, I would think.

wow

I dont see why knight to  e 5 would be considered brilliant i think most people woudl have seen that opportunity.  but hey whatever the computer thinks is brilliant lol who are we mere mortals to argue, unless there was a follow up to that move that no one could have possibly seen.

No, it wasn't a brilliant move, you're right. Just a natural move. I wonder who they pay to write these programs.

i should have 1500 brilliant moves by now if thats brilliant lol

Yeah, same here. Also, I didn't find Brilliant Move number 2.

still isnt brilliant though its a good move but brilliant hell no lol

Wait what happened? The analysis is not the same now it only shows 1 "brilliant" move and its not even one that was showed in yesterday's analysis !

A brilliant move in the computer analysis means a move that the engine does not consider best at a low depth, but it realizes that it's the best move at the depth you have selected. For example, if you set the depth to 18, a brilliant move will be considered a move that the engine evaluates as not best at d<=17 (17 is just an example, I don't know where the cutoff is), but it realizes that it's best at d=18. If you analyse the same game at a higher depth, a move that was marked as an inacuracy at d=18 may be marked as a brilliancy at d=30. This means that a lot of the time, moves that are considered brilliances might be very easy to see for a human, it's just that the computer needs a higher depth to correctly evaluate them.

ie one move further than the engine's horizon.

I don't know why, but there is only 1 brilliant move in my analysis and the one that appeared (Ne5) is not really hard to find because it attacks the queen and centralizes the knight so...

Huh

thats cool

*eats frozen child*

Congrats PawnToBthree, I've played hundreds of bot games looking for the elusive brilliant move.  I didn't think it was possible.

Yep I don't know why the analysis changed completely 

But with this analysis showing an extra brilliant move, can we now say that I had 3 brilliant moves in the same game? That must be a new record, I hope chess.com has a special prize for me 

Worst Brilliant moves I have ever seen

They were not brilliant at all.
Chess.com's stockfish evaluation on how good a move is sucks, honestly.

It may affirm that a genial sacrifice made by a super GM is a blunder while considering so common things such as capturing a piece or returning an attacked knight brilliant.