Computer recommends piece sacrifice, but I'm not sure what the plan is afterwards - Chess Forums

Oct 17, 2019 #1

Here is a fragment of a game that I played yesterday (I was Black). I had a lot of fun with this game. But when going through it afterwards I puzzled over the position after 11. b4.

During the game, I thought very hard about taking with the knight 11. ... Nxb4. But after 12. cxb4 Bxb4+, I wasn't really sure what the plan would be. It looks like White can just block the check with 13. Nfd2, and then try to get his king castled.

There's a lot of open space around White's king, and his pieces aren't doing much good yet. But on the other hand, the black pieces don't look like they have any easy way to get into the action either. I consulted the computer oracle about this, and indeed, it claims that trading the knight for two pawns is the way to go. What do you guys think? I don't mind sacrificing a piece, but in this case I just don't see it. Maybe the idea is that Black should play to push the queen-side pawns? What would Black's plan be?

Oct 17, 2019 #2

lund0682 wrote:

Here is a fragment of a game that I played yesterday (I was Black). I had a lot of fun with this game. But when going through it afterwards I puzzled over the position after 11. b4.

During the game, I thought very hard about taking with the knight 11. ... Nxb4. But after 12. cxb4 Bxb4+, I wasn't really sure what the plan would be. It looks like White can just block the check with 13. Nfd2, and then try to get his king castled.

There's a lot of open space around White's king, and his pieces aren't doing much good yet. But on the other hand, the black pieces don't look like they have any easy way to get into the action either. I consulted the computer oracle about this, and indeed, it claims that trading the knight for two pawns is the way to go. What do you guys think? I don't mind sacrificing a piece, but in this case I just don't see it. Maybe the idea is that Black should play to push the queen-side pawns? What would Black's plan be?

The compensation is pretty clear. Wouldn't it have been fun to try it?

Oct 17, 2019 #3

11...Nxb4 is a positional sacrifice, the most difficult type of sacs. While the white king is not under immediate threat, it is several tempi from castling to safety, a fact that black hopes will bring him enough compensation.

White might be able to hold with precise play, but practically it’s much easier to attack than defend. As an example: 12. cxb4 Bxb4 13. Nbd2 (Objectively speaking, 13. Nfd2 Bg4 14. Qd3 Ne4 15. Be2 Bxe2 16. Qxe2 Rae8 17. O-O might be better, when black will regain the piece and enjoy his extra pawn) 13... Ne4 and it’s already impossible to parry the threat of ...Bg4 and ...g5.

For positional sacrifices, intuition is more important than calculation. Attacking moves will come as natural — the knight hops to e4 where it’s really active, the LSB goes to g4, the a8-rook goes to e8 and all your pieces are aiming at the white king. Don’t worry that your opponent might defend like Stockfish. Only the engine itself can do that!

Oct 18, 2019 #4

I like the first response about sacrifing “just for the fun of it”. I really like the idea of fun being ample compensation. It IS tremendous fun! On the other hand, I’ve lost countless games this way, which does take away some from the fun.

But inkspirit, your comments about intuition make a lot of sense. During the game, I wasn’t confident in my ability to calculate far enough to see Black regaining the material (it was just a 10 minute game, so not much time to think). But I did see that White was far from castling, and I had a feeling that as Black I would probably be able to stir up some kind of action. I guess learning to trust those hunches comes with experience (hopefully).

Oct 18, 2019 #5

How much calculation skills one needs to see that after 13.Nfd2 Ne4 white can barely move something, and Black has several nasty threats like Bg4, Qf6, etc?

Oct 18, 2019 #6

There is always principal variaiton of computer move (main line) which is usually on the top. Pause the analysis.

If you cant visualize the engine variation properly, Right click on the engine output window

Click visual variation board.

Now you will see additional visual board.

Click left and right arrows on the bottom of additional visual variation board.