The usual case one often sees specially in blitz play is when one side (usually the winning) mis-plays and then stalemate occurs with the losing side dodging a loss. But are there positions where stalemate is a forced (maybe planned) outcome no matter what the opponent does?

5 Answers 5


Oh sure, here are two examples (annotated):

This is a neat puzzle from the chess.com tactics, with white to play!

 [title "White to play!"]
 [fen "1q6/2b2ppb/4p1k1/7p/2Np1p1P/3P1Q2/6PK/8 w - - 0 1"]

 1.Ne5+ Bxe5 {black has to take else white either checkmates or wins the queen.} 2.Qg3+ fxg3+ {Pawn takes queen is forced else white checkmates} 3.Kh3 {and black has no way to avoid the imminent stalemate, any move draws!}

Sometimes stalemates are exploited strategically or tactically to force a draw, e.g. in the game below between So and Karjakin, where black may have had the stalemate setup in mind already when the white pawns got to g6 and h7.

 [title "W. So vs S. Karjakin 2017, black to move"]
 [fen "4r2k/4P1pP/2pB2P1/2P2K2/8/8/R7/4r3 b - - 0 1"]

 1.R8xe7 {black switches to antichess style of play to force the draw.} Bxe7 2.Re5+ {and white has no way to avoid the stalemate, the rook will keep checking the king where ever it goes.} Kxe5
  • 7
    I would like to upvote this twice, once per example. Bravo.
    – thb
    Commented Feb 7, 2019 at 14:09
  • 1
    @thb why don't we let Ellie post 2 answers?
    – BCLC
    Commented Feb 13, 2021 at 14:20

Viswanathan Anand vs Vladimir Kramnik is a notable one:

[FEN ""]
[Event "World Championship Tournament"]
[Site "Mexico City MEX"]
[Date "2007.09.15"]
[EventDate "?"]
[Round "3"]
[Result "1/2-1/2"]
[White "Viswanathan Anand"]
[Black "Vladimir Kramnik"]
[ECO "C42"]
[WhiteElo "2803"]
[BlackElo "2811"]
[PlyCount "130"]
[StartPly "128"]

1.e4 e5 2.Nf3 Nf6 3.Nxe5 d6 4.Nf3 Nxe4 5.d4 d5 6.Bd3 Nc6 7.O-O
Be7 8.c4 Nb4 9.Be2 O-O 10.Nc3 Bf5 11.a3 Nxc3 12.bxc3 Nc6
13.Re1 Re8 14.cxd5 Qxd5 15.Bf4 Rac8 16.Qa4 Bd7 17.Qc2 Qf5
18.Qxf5 Bxf5 19.Bb5 Bd7 20.d5 Ne5 21.Bxd7 Nxd7 22.Bxc7 Rxc7
23.d6 Rxc3 24.dxe7 f6 25.Rad1 Rc7 26.Nd4 Ne5 27.f4 Nc6 28.Nxc6
bxc6 29.Rd6 c5 30.Ree6 c4 31.Rc6 Rexe7 32.Rxc4 Rxc4 33.Rxe7
Ra4 34.Rb7 h6 35.f5 Rxa3 36.Kf2 h5 37.g3 a5 38.Ra7 a4 39.h4
Ra2+ 40.Kf3 a3 41.Ke3 Ra1 42.Kf2 Kf8 43.Kg2 a2 44.Kh2 Ke8
45.Kg2 Kd8 46.Kh2 Kc8 47.Kg2 Kb8 48.Ra3 Kb7 49.Ra4 Kb6 50.Ra8
Kc5 51.Ra7 Kd5 52.Ra4 Ke5 53.Ra5+ Ke4 54.Kh2 Kf3 55.Ra3+ Kf2
56.Ra4 Kf1 57.Kh1 Ke1 58.Kg2 Kd1 59.Ra7 Rc1 60.Rxa2 Rc2+
61.Rxc2 Kxc2 62.Kf3 Kd3 63.g4 hxg4+ 64.Kxg4 Ke4 65.Kh5 Kxf5

Anand forces the draw with 65. Kh5, after which 65... Kxf5 is stalemate, while not taking the pawn allows white into a winning pawn endgame.


The most common is the situation with King + Pawn vs King, if it's a draw. Okay, the stalemate isn't forced, but everything else is a draw too and it's important that the stalemate exists:

[FEN "8/3k4/8/3KP3/8/8/8/8 w - - 0 1"]

1.e6+ Ke7 2.Ke5 Ke8 3.Kd6 Kd8 4.e7+ Ke8 5.Ke6 {Stalemate}

There is also a sideline in the so-called Saveedra position:

[FEN "8/8/1KP5/3r4/8/8/8/k7 w - - 0 1"]

1.c7 Rd6+ 2.Kb5 (2.Kc5? Rd1! {followed by Rc1(+), =}) Rd5+ 3.Kb4 Rd4+ 4.Kb3 Rd3+ 5.Kc2 Rd4! 6.c8=Q? (6.c8=R! Ra4 7.Kb3 {White wins}) Rc4+ 7.Qxc4 {Stalemate}

Instead 6.c8=R wins, because the stalemate isn't there.

  • Probably you mean 2. Kc5? Rd1! instead of 2. Kc5? Rc1! in the sideline of the second example?
    – wimi
    Commented Jun 19, 2020 at 17:44
  • @wimi I did, yes. Though Rc1 would also work if legal :-) Commented Jun 20, 2020 at 14:45

Here is a (rather simple) example of a position where stalemate is forced, no matter what either player does:

[FEN "8/p1p3pp/5ppk/6q1/5PPK/6P1/P1P3PP/8 w - - 0 1"]

1. fxg5+ fxg5+ 2. Kh3 a6 3. a3 a5 4. a4 c6 5. c3 c5 6. c4

Glorfindel’s answer shows an example of a forced mutual stalemate. On a similar note, in the below position the soon to be stalemated side must stalemate themselves or lose, and the opposing side has no choice but to allow it. Forced here is a bit of a combination of best move and must make a move due to the king being in check. I made the sequence absurdly long just for fun.

[Title "Me, chess.stackexchange.com 5/19/2019, Black Draws"]
[FEN "3Rbk1K/qqqqq1NQ/6Qq/6QQ/6Qq/2b3QQ/1B4Qq/q5R1 b - - 0 1"]
[startflipped ""]

1... Qexg7+ 2. Qxg7+ Qdxg7+ 3. Qxg7+ Qcxg7+ 4. Qxg7+ Qbxg7+ 5. Qxg7+ Qaxg7+ 6. Qxg7+ Bxg7+ 7. Bxg7+ Qaxg7+ 8. Rxg7 Qxh7+ 9. Qxh7  Qxh7+ 10. Qxh7 Qxh7+ 11. Kxh7
  • I've played around a lot with this kind of position in the context of dead reckoning. It's a lot easier to keep the train on the rails if both players are trying to optimize as you are here, than if there is help play (as is implicit in the DP rule). But what happens if Black begins Bxg7+ instead of Qxg7+? I don't find a line for White to win. With help play, there's a pattern I've observed which is that each queue can only contain pieces from one colour, otherwise the draw is avoidable. Should be provable
    – Laska
    Commented Feb 4, 2021 at 5:12
  • 3
    wow that was fun.
    – BCLC
    Commented Feb 13, 2021 at 14:22

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.