Positional sacrifice - how to evaluate?

Consider the following position:



Black has just moved 15. ... h6 and white's knight is en prise. At this point, should white sacrifice the knight with 16. Nxf7?! I can see this leading to
16. ... Kxf7 17. fxg6 Ke8
Now black cannot castle, and white's f rook has a nice open file. Is this enough compensation for the knight? How does one evaluate positions like this?

Personally I reckon a waste of a knight
besides he can castle kingside threatening your knight and no matter where you flee you lose the knight without compromising his ability to castle

Originally posted by Mctayto
Personally I reckon a waste of a knight
besides he can castle kingside threatening your knight and no matter where you flee you lose the knight without compromising his ability to castle

after castling - knight captures h6+ - position is blown open, I think it's good.

Originally posted by onyx2006
after castling - knight captures h6+ - position is blown open, I think it's good.

What? Did you even look at the position and intended moves?

The sacrifice has no real followup and just leaves white down material. Bad.

Originally posted by XanthosNZ
What? Did you even look at the position and intended moves?

The sacrifice has no real followup and just leaves white down material. Bad.

It's only a bloody knight... ;-)

Originally posted by XanthosNZ

The sacrifice has no real followup and just leaves white down material. Bad.

After 16... Nxf7 17. Kxf7 fxg6 18. Ke8 white has two pawns, one passed pawn that he might be able to hold on to, one open file and a king to harass. Is this enough compensation?

In terms of actual continuations, I was thinking of something like:

18. e5! dxe5 19. g7 Rg8 ( here 19 ... Bxg7? 20. Qg6+ Kd8 21. Qxg7
and white has regained the material and black is in trouble.)
20. Bxh6 and white has three pawns for the knight.

18. ... Bxe5 has similar results.

Also,
16. Nxf7 0-0 17. Nxh6 and black is two pawns down with nothing to show for it.
16. Nxf7 Kxf7 17. fxg6 Kxg6 18. e5+ ... white regains material and black king is exposed.

16.Nxf7 is no good, White simply doesn't have enough pieces to launch an effective attack. 16.e5 might be worth a shot though.

Originally posted by ngkabra
After 16... Nxf7 17. Kxf7 fxg6 18. Ke8 white has two pawns, one passed pawn that he might be able to hold on to, one open file and a king to harass. Is this enough compensation?

After 16. Nxf7 Kxf7 17. fxg6, I reckon black will play Kg7 instead of Ke8. And while the king is exposed, black has pretty good defensive options - rooks onto the f-file, Be8, e5, that sort of thing.

I don't think it's worth it, personally.

Here's a position from one of my games:

Game 2086397



It's Black to play. I looked at all his pieces and saw that his dark squared Bishop was his most active piece, preventing me from playing moves like Rd8. Well I felt a sacrifice was in order and played Rd8, hoping to trade his most important piece for an inactive rook. Well he did and the rest of the game White suddered from lack of space and did not have any control over the dark squares. It was probably far from sound but was fun to play 😛.

Originally posted by ngkabra
[fen]r3k2r/pp1bpp2/2qp1bpp/2p2PN1/2P1P3/2PQ3P/PP4P1/R1B2RK1 w kq - 0 16[/fen]

Black has just moved 15. ... h6 and white's knight is en prise. At this point, should white sacrifice the knight with 16. Nxf7?! I can see this leading to
16. ... Kxf7 17. fxg6 Ke8

After 17...Kg7, Black is quite safe. And up on material.

Thanks for all the responses. I do now have a better idea of what to look for when evaluating a sacrifice.

Originally posted by ngkabra
Thanks for all the responses. I do now have a better idea of what to look for when evaluating a sacrifice.

It's not really about the next move but what comes after? Will you have an attack or chances of compensation? It doesn't look like it to me.

If white had more of his pieces pointed at the kingside it might be justified but blacks defences look to hold as is. After further build up of force, the kingside would definitely be the place to start an attack though.