7

According to Wikipedia, "any chess piece except the king may be sacrificed". However, I am wondering,if in some rare situations, whether endangering the king can also be a good strategy. Here is what I mean.

In a position, I make a move after which my opponent has a checkmate in one. But this checkmate is very counterintuitive and is difficult to find, especially if my opponent is not so strong and/or they are in time trouble. If they miss this checkmate, my position is actually better.

Are there any examples of the above-mentioned scenario? Can endangering the king be a good idea occasionally, or should it always be avoided?

6
  • 18
    I don't think "sacrifice" would be the right word for that. A sacrifice, as the article also explains, means deliberately offering up material for some other compensation. Here, you don't really want to give up your king and you pray that the opponent doesn't even see the offer. I'd rather call that "hope chess".
    – Annatar
    Nov 9, 2021 at 15:56
  • 17
    I'm not sure why you'd call this a "sacrifice" - there is no intent to give up a piece, and the advantage is only gained if your opponent fails to recognize the opportunity. A sacrifice is an intentional move to give up material for some tactical advantage, but you can't give up the king without losing - this is just a risky move. In general, though, non-optimal play can be better (faster) than optimal play against an opponent who does not play optimally, but underestimating your opponent's skills can lead to disaster. Nov 9, 2021 at 15:57
  • 14
    When playing against Wookies, I prefer to sacrifice my King than my arms.
    – dotancohen
    Nov 10, 2021 at 13:12
  • All the answers are assuming you mean "risk", not "sacrifice", so I'd suggest editing to ask that (maybe with just a note at the bottom that the question originally said "sacrifice"). Actually directly sacrificing your king is usually prevented by rules that prevent moving into check. But there's an even simpler mechanism to give up your king: resign. Nov 11, 2021 at 2:51
  • Some kind of gambit maybe. Nov 11, 2021 at 13:22

5 Answers 5

26

Whereas "sacrifice" is a bad word, as already was pointed out ("endangering" maybe), from the standpoint of game theory this strategy may be completely rational...if the players are fallible.

I abstract your example a bit. Assume the position is already at a -3 value with objective play value, i.e. it is lost. You have moves A and B. All opponent moves on A give a -3 position. On B the opponent can mate immediately (-infinity) or blunder away (0). The minimax value thus is -3, and a computer plays A since -infinity<-3. With "hope chess" you play B - you don't have to lose anything since you will "normally" lose anyway. But since you wanted a position:

[FEN "2k5/2b5/N1KPN3/8/7B/8/8/8 b - - 0 1"]
[startflipped ""]

Black to move. "Best" is 1...Bxd6, and White only has to mate with BNN, not that hard. But why not shorten the torture by 1...Bd8? Maybe White is a patzer and triumphantly captures the B...stalemate.

4
  • 4
    What is "BNN"? I don't see a mating move following Bxd6. (Admittedly I'm bad at chess, and not familiar with the terms.) Nov 10, 2021 at 12:59
  • 7
    @SodAlmighty just refers to mating with bishop (B) + knight (N) + knight (N). like a lot of other endgame(?) mates it's a strategy, not a mate in one
    – somebody
    Nov 10, 2021 at 13:04
  • 3
    Oh I see. I guess the phrase "shorten the torture" in the answer should have been a clue. Thanks for the clarification. Nov 10, 2021 at 13:13
  • Search on YouTube for < Eric Rosen stalemate trap >. One he uses often is like this. (White, losing, seeks a draw.) Kb1 & piece-c2 hoping for ...Qxc2+. Then Ka1 hoping for a random move. Black blundered a win as Qd2 would've avoided stalemating White.
    – Rosie F
    Nov 11, 2021 at 6:59
9

What checkmate in 1 might be tricky to spot? One possibility: promoting a pawn to a knight. This might work better online than over the board: the software your opponent uses might make it easier to promote to queen than to another piece, and indeed some players have "auto-queen" turned on.

5
  • 1
    Thank you for the answer! For some examples of tricky mates in one move, see thechessworld.com/articles/problems/…
    – Zuriel
    Nov 10, 2021 at 13:12
  • 1
    @Zuriel Are those really examples of mates in one that are tricky to spot over the board? In problem 1, the mating move seems like absolutely the first move you think of, in response to the opponent's previous move. Problem 2 relies on the board being presented the wrong way around. In Problem 3, the challenge is not to mate. Nov 10, 2021 at 14:08
  • @JamesMartin, how about these ones? lichess.org/study/6zUr5NJY
    – Zuriel
    Nov 10, 2021 at 15:07
  • 1
    Those look like good puzzles! I think they'd be fun for a solving battle (even though they are nothing like positions that would occur in an actual game). Nov 10, 2021 at 15:15
  • 3
    A beginner's perspective: sometimes mates in 1 can be very difficult to spot if you don't know there is a mate in 1 available. If you know it exists though, it's not a difficult problem.
    – Chuu
    Nov 10, 2021 at 21:19
4

A maxim in chess is not to play a move while thinking "I hope he doesn't see it". This doesn't mean that your opponent will always spot the threat - and if they don't then congratulations on the easy win - but the point is that if opponent spots the threat, your move should still improve your position in some way.

Example:

[FEN ""]
1. e4 e5 2. Nf3 Nc6 3. Bc4 Nd4? 4. Nxe5?? Qg5 {Black wins material thanks to the dual threats of 5...Qxe5 and 5...Qxg2 6. Rf1 Qxe4+ 7. Be2 Nf3# if White rescues the knight with 5. Nxf7.}

So 3...Nd4 wins material if opponent doesn't see the threat. It's still not a good move however, because if it's parried (4. Nxd4 or 4. O-O) White gets a strong advantage.

You are describing something even worse than a simple bad move, since the consequence of opponent seeing it is you lose the game. Now if you are going to lose the game anyway, then it's a reasonable gambit to try, but in all other situations you should never intentionally make a move that can lose your the game if your opponent spots the refutation.

4
  • The comment on 4...Qg5 isn't quite right: 7...Nf3 is not mate (and indeed would be a gross blunder) while wN is still on e5. That line works after 5 Nxf7. Black must work harder to win after 5 Nf3 Qxg2 6 Nxd4 Qxh1+ 7 Bf1 (clear exchange up plus a pawn, but not checkmate) or better yet 5 f4. Nov 10, 2021 at 15:08
  • Indeed, will fix.
    – Allure
    Nov 10, 2021 at 15:21
  • Wouldn't 5. Ng4 be a save?
    – justhalf
    Nov 11, 2021 at 12:02
  • 1
    @justhalf good idea, but 5...d5 is a double attack that still wins a piece.
    – Allure
    Nov 11, 2021 at 13:11
4

Yes!

It is absolutely a valid strategy in online play with fast time controls. Here (youtube link) is an example in online play between two international masters:

[FEN "K/8/1R/8/8/5k/8/4qqqq w - - 0 1"]

1. Ka7 Qxb6 2. Ka8 Qef2?

After Qxb6, White has two moves, to take the queen which is mate in 3, or move to the corner which is mate in one. In the game, White played into the corner, which black did not expect, leading black to blunder stalemate before realizing what had happened.

1
  • 6
    I'm guessing Black pre-moved 2...Qef2. Indeed, this is a cunning strategy in these types of situations.
    – Durmus
    Nov 11, 2021 at 7:42
1

Further to existing answers, another sort of checkmate in 1 that might be tricky to spot is checkmate by discovered check.

2
  • The mate he is missing is not by discovered check. In fact it's the discovered check lines that distract him from the mate! Nov 11, 2021 at 14:50
  • @JamesMartin I stand corrected. So I've removed that example.
    – Rosie F
    Nov 11, 2021 at 20:12

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.