# Analyse a blunder

Black (me) to move.

I took the knight at f3 with my knight at e1. However, Lichess analysis shows that this move is a blunder, and suggests to take the bishop at g2 instead.

IMO, I am winning no matter I exchange the knight or bishop, but I feel hard to understand why exchanging the knight f3 is a blunder?

``````[FEN "r1bq1rk1/pp3ppp/4p3/2Pp4/2P5/P1B2NP1/3NPPBP/R3nRK1 b - - 0 1"]

1...Nxf3? (1...Nxg2!)
``````
• A link to the lichess game might be useful. That said 1) I wouldn't worry too much, the difference in the captures is between very, very winning and very, very, very winning 2) That said I would have taken the bishop without thinking too hard - bishops are usually a little stronger than knights, especially when you possess the bishop pair, and given the choice I would always take the bishop unless there is a good reason to do otherwise, and here the white squared bishop is potentially about White's best piece. Jan 15 at 13:07
• Did Lichess actually say "blunder" or is that your interpretation? Jan 15 at 21:33
• @user21820 This is a blunder considered by Lichess. Please see the link: lichess.org/GoIUYiwE/black#28.
– null
Jan 16 at 20:32
• Thank you for the link; it seems that the Lichess SF engine used in that analysis had searched to a depth where it saw a way to reach much clearer winning position for `Nxg2` than for `Nxf3`, and this yielded −6.7 instead of −3.2, which was significant enough that made it classify it as a blunder. From the engine's viewpoint, −6.7 is very much better than −3.2; although it will almost always win as Black given either, the likelihood that it will fail is much lower for −6.7 than −3.2. Jan 17 at 13:14
• It's a matter of opinion whether you should consider it a blunder or not, because the bishop is really somewhat annoying. To understand the difference, simply try playing against maximum-level SF after `Nxf3` (via Menu > Continue from here), and I would be interested to see the game that you end up with. Jan 17 at 13:25