I'd like to study examples where a much weaker player succeeded against an IM or GM.


This game comes to mind:

It was a first game in a 2 game match of the first round knock out FIDE championship. Anand rated at 2770 was I believe the highest rated player in the event. Touzane, I believe, was not a professional player. Touzane won the first game. Despite the first game win for Touzane, Anand won the second, and then the tie-break. Nevertheless, Anand losing to a player rated 400+ below him was quite a shocker.

[FEN ""]
[Event "FIDE WCh KO"]
[Site "Moscow RUS"]
[Date "2001.01.02"]
[EventDate "2001.11.27"]
[Round "1.1"]
[Result "0-1"]
[White "Viswanathan Anand"]
[Black "Olivier Touzane"]
[ECO "C42"]
[WhiteElo "2770"]
[BlackElo "2368"]
[PlyCount "78"]

1. e4 e5 2. Nf3 Nf6 3. Nxe5 d6 4. Nf3 Nxe4 5. d4 d5 6. Bd3 Nc6
7. O-O Be7 8. c4 Nf6 9. Nc3 O-O 10. Re1 Be6 11. cxd5 Nxd5
12. a3 Bf6 13. Be4 h6 14. Bc2 Nxc3 15. bxc3 Bc4 16. Nd2 Bd5
17. Rb1 Bg5 18. c4 Bxd2 19. cxd5 Bxe1 20. dxc6 Ba5 21. Qd3 g6
22. cxb7 Rb8 23. Bxh6 Re8 24. Qf3 Re6 25. Bb3 Rf6 26. Qg4 Rb6
27. Bg5 Qe8 28. Qf3 R8xb7 29. h4 Qd7 30. g4 Rxb3 31. Rxb3 Rxb3
32. Qxb3 Qxg4+ 33. Kf1 Qxd4 34. Be3 Qa1+ 35. Kg2 Bb6 36. Bxb6
axb6 37. Qg3 Qc1 38. h5 Qc6+ 39. Kf1 Qh1+ 0-1

Anand - Touzane

  • 3
    I thought this was interesting, so I ran it through Phalanx-Scid. Anand has a slight pull and a nice attack through move 25. The exchange sacrifice got my attention! After 25. Be3 , however, Anand's advantage drops from 1.5 to 1.3 pts, which is probably still enough for a win against Touzane. Now things begin to come apart for Anand. 26. Qg4? misses that the rook on f6 is defending the g7 pawn against the Q, not the B on b3. When the queen moves, and leaves the rook on b1 and the pawn on b7 undefended, things get interesting. (more to come)
    – Tony Ennis
    Jun 16 '12 at 16:18
  • 1
    Now Anand's advantage is down to .3 points. Phalanx says, 26. ... Rb6 27. Ba2 R8xb7 28. Rc1 Qf6 29. Qc8+ Kh7 30.Qh3 Kg8 31. Qc8+, apparently considering a draw by repetition. Black of course can't allow the discovered check. Phalanx doesn't like 27. Bg5 stating it loses another half-pawn to 27. ... Qe8 28. h3 R8xb7 29. Qh4 c5 30. Rc1 Qd7 31. Rxc5 Rxb3 32. Rxa5 Rb1+ 33. Kh2 So what's come about has been related to black's excellent 26th move, which pinned the bishop on b3 and exploited Anand's weak back rank. (more to come)
    – Tony Ennis
    Jun 16 '12 at 16:50
  • 1
    Ok, I got curious and found out that Phalanx is a cool program but not in the top rated engines. I switched to Stockfish. After 27. Bg5 black is winning, if ever so slightly - about a quarter of a pawn. Then comes 27. ... Qe8 28. Qf3 [h3 - Stockfish] which costs a full point while still not addressing the back-rank weakness. Touzane finds the right move, and Anand plays 29. h4 [g4 - Stockfish] which doesn't help (or hurt) his game. (more to come)
    – Tony Ennis
    Jun 16 '12 at 17:38
  • 1
    What's interesting is that for the last few moves, Touzane is finding all the right moves according to Stockfish. Anand is not. Now, bear in mind Anand and Touzane are both more highly rated than this program on this computer... So this analysis is more of an assist than some gospel truth. After 30. g4 (keeping the back Q off of f6) Touzane exploits white's overloaded queen: 30. ... Rxb3 31. Rxb3 Rxb3 32. Qxb3 Qxg4+. Now white's d-pawn is forfeit, white pawns are all islands, and black has a passed pawn.
    – Tony Ennis
    Jun 16 '12 at 17:50
  • 1
    For all that, The computer rates Anand 1.25 pawns behind. Perhaps Anand can pull out the draw. Starting now, for brevity, I'm putting Stockfish's moves in brackets. But 33. Kf1 [Qg3] causes more erosion in Anand's position. Black finds the moves again while Anand does not with 33. ... Qxd4 34. Be3 [Qf3] Qa1+ 35. Kg2 [Ke2] Bb6 36. Bxb6 [Kg3] and white's position continues to erode. He's now down by over 2 points. Now 36. ... axb6 37. Qg3 Qc1 [c5] and now it's Touzane who gives some back. Anand's position looks lost just the same.
    – Tony Ennis
    Jun 16 '12 at 18:48

In 2009, GM Nurlan Ibrayev of Kazakhstan (rated 2407 at the time) lost a classical time control game to 9-year-old Hetul Shah of India, who then had a FIDE rating of 1817. Shah's current rating, now three years after that game, is 1979, not so much higher than it was at the time of his upset victory. So this game wasn't a case of Ibrayev running up against some seriously underrated prodigy whose rating was about to shoot up several hundred points. To the contrary, this seems to be exactly the sort of game you are after: a much weaker player having a good day and taking a grandmaster scalp, with the black pieces no less!

[FEN ""]
[Event "7th Parsvnath International Open"]
[Site "Delhi"]
[Date "2009.01.11"]
[EventDate "2009.01.11"]
[Round "1.36"]
[Result "0-1"]
[White "Nurlan Ibrayev"]
[Black "Hetul Shah"]
[WhiteElo "2407"]
[BlackElo "1817"]
[PlyCount "88"]

1.e4 e5 2.Nf3 Nc6 3.Bb5 a6 4.Ba4 Nf6 5.O-O Be7 6.Re1 b5 7.Bb3 d6 8.c3 O-O
9.h3 Na5 10.Bc2 c5 11.d4 Qc7 12.Nbd2 cxd4 13.cxd4 Bd7 14.Nf1 Rfc8 15.Ne3 Nc6
16.a3 Bf8 17.b3 Nxd4 18.Nxd4 Qc3 19.Ne2 Qxa1 20.Qd2 d5 21.b4 d4 22.Nd1 Rxc2
23.Qxc2 Rc8 24.Qd2 Nxe4 25.Qd3 Nd6 26.Bb2 Qa2 27.f4 Qc4 28.Qb1 d3 29.Ne3 Qe4
30.Nc3 Qxf4 31.Ncd5 Qg3 32.Nf1 Qg6 33.Bxe5 Nc4 34.Nf4 Qb6+ 35.Kh1 Nxe5 36.Rxe5
Bd6 37.Re4 Bc6 38.Qc1 Qb7 39.Re1 Bxg2+ 40.Nxg2 Rxc1 41.Rxc1 Bf4 42.Rc3 d2
43.Rd3 h6 44.Kg1 Qb6+ 0-1

Not very serious, but fun to watch: http://www.youtube.com/watch?NR=1&feature=endscreen&v=LMM2RV5-HVc

  • I learned the stalemate joke, but I'm afraid I don't find a blitz blunder to be of much interest (downvoting because the OP mentions “study”). Nov 18 '12 at 7:25

The 2017 Isle of Man Masters tournament provided a couple of these shock results.

First the fairy tale story of an aging, declining US GM overturning an almost 400 point rating difference to beat one of the pre-tournament favourites, ex world champion Vladimir Kramnik.

[fen ""]
[Event "Isle of Man Masters"]
[Site "Douglas IMN"]
[Date "2017.09.25"]
[EventDate "2017.09.23"]
[Round "3.28"]
[Result "1-0"]
[White "James Tarjan (2412)"]
[Black "Vladimir Kramnik (2803)"]
[ECO "A12"]
[WhiteElo "2412"]
[BlackElo "2803"]
[PlyCount "111"]

1. c4 Nf6 2. g3 c6 3. Nf3 d5 4. b3 Bg4 5. Bg2 e6 6. O-O Nbd7 7. Bb2 Bd6 8. d3 O-O 9. Nbd2 Re8 10. h3 Bh5 11. Re1 a5 12. a3 e5 13. cxd5 cxd5 14. Nh4 Nc5 15. Qc2 Ne6 16. Rac1 Nd4 17. Qd1 Nb5 18. Nb1 Qd7 19. Kh2 Ra6 20. Nf3 e4 21. dxe4 Nxe4 22. Rf1 Bb8 23. Nc3 Nbxc3 24. Bxc3 Rae6 25. Be1 h6 26. Rc2 Ba7 27. Qc1 Bb6 28. e3 Qb5 29. Nd4 Bxd4 30. exd4 Bf3 31. Bxf3 Nxg3 32. fxg3 Qxf1 33. Bf2 Qd3 34. Rc3 Qf5 35. Kg2 Rf6 36. Qc2 Qd7 37. g4 Rc6 38. Rc5 Rd8 39. Qf5 Rxc5 40. Qxd7 Rxd7 41. dxc5 d4 42. Kf1 d3 43. Ke1 d2+ 44. Kd1 Kf8 45. Bg3 Ke7 46. Bd6+ Ke6 47. Kxd2 b6 48. Ke3 bxc5 49. Bxc5 Rd8 50. b4 axb4 51. axb4 f5 52. b5 fxg4 53. hxg4 g6 54. b6 h5 55. g5 Kd7 56. b7 1-0

In the first round an obscure 70 year old English player took a draw by repetition in a winning position against a GM rated almost 700 points higher.

[fen ""]
[Event "Isle of Man Masters"]
[Site "Douglas IMN"]
[Date "2017.09.23"]
[EventDate "2017.09.23"]
[Round "1.14"]
[Result "1/2-1/2"]
[White "Zaki Harari (2027)"]
[Black "Maxim Rodshtein (2695)"]
[ECO "B21"]
[WhiteElo "2027"]
[BlackElo "2695"]
[PlyCount "147"]

1. e4 c5 2. d4 cxd4 3. c3 d3 4. Bxd3 Nc6 5. c4 g6 6. Nf3 Bg7 7. O-O d6 8. h3 Nf6 9. Nc3 O-O 10. Be3 Nd7 11. Rc1 Nde5 12. Be2 b6 13. Qd2 Bb7 14. Rfd1 Rc8 15. b3 Nxf3+ 16. Bxf3 Ne5 17. Be2 Nd7 18. Bh6 Nf6 19. Bxg7 Kxg7 20. Bf3 Qc7 21. h4 Qb8 22. h5 Rc5 23. h6+ Kg8 24. Nd5 Bxd5 25. exd5 Rfc8 26. Re1 R8c7 27. a4 Qf8 28. Qf4 a5 29. Qe3 Kh8 30. g4 Ng8 31. g5 f6 32. gxf6 exf6 33. Qf4 Re7 34. Kf1 Re5 35. Rxe5 dxe5 36. Qd2 Qd6 37. Rd1 Rc7 38. Bg2 g5 39. Bh3 Nxh6 40. Be6 Ng8 41. Qe2 Ne7 42. Qf3 Kg7 43. Qh5 Ng6 44. Bf5 Re7 45. Rd3 Nf8 46. Rh3 Kg8 47. Be4 Rg7 48. Bf5 Re7 49. Be4 Qd7 50. Rf3 Rf7 51. Bf5 Qe7 52. Rd3 Qd6 53. Rh3 Rg7 54. Rf3 Qd8 55. Rh3 Ra7 56. Rd3 Qd6 57. Rh3 Rf7 58. Be4 Qa3 59. Bf5 Re7 60. Qd1 Qd6 61. Qd3 Rg7 62. Rh6 g4 63. Qg3 Rg5 64. Be4 Kg7 65. Qh4 Qe7 66. d6 Qf7 67. d7 Qe7 68. Rxh7+ Nxh7 69. Qxh7+ Kf8 70. Qh8+ Rg8 71. Qh6+ Rg7 72. Qh8+ Rg8 73. Qh6+ Rg7 74. Qh8+ 1/2-1/2

Bd5 crying out to be played instead of Qh8.


Late to the party, but while Anand losing to a guy, who was rated 400 points below him was, clearly, a huge upset; an IM, who was over 2400 at one point, and 2368 when that game was played is definitely strong enough to take advantage if Anand makes an error.

This still probably beats all upsets I have ever heard of. In the first round, GM John Fedorowicz, rated 2479 at the time (I cannot find this to prove it, but I am recall his USCF rating being in the mid-2500's as well), lost to a 1714 player at the 1980 U.S. Open!

I still remember this because I was a really new player, and saw this game in "Chess Life and Review" (Later shortened to "Chess Life"), and I was amazed. Nevertheless, Fed brushed off the early setback, and tied for first.


[Title "W Stanley Davis-John Fedorowicz, 81st US Open, 8/3/1980"]
[FEN ""]

1.Nf3 e6 2.e3 b6 3.b3 Bb7 4.Bb2 c5 5.Be2 Nf6 6.d3 d5 7.Nbd2 Be7 8.Ne5 O-O 9.O-O Re8 10.f4 Bf8 11.Qe1 a5 12.g4 a4 13.Rc1 axb3 14.axb3 Ra2 15.Ba1 Na6 16.g5 Nd7 17.Bh5 Nxe5 18.Bxe5 Nb4 19.Qh4 Nxc2 20.Rf3 Nxe3 21.Rxe3 Rxd2 22.Rh3 c4 23.Bf6 Bc5+ 24.Kf1 Bf2 25.Bxf7+ Kf8 26.Qxh7 gxf6 27.Bg6 1-0

In a 1929 against Fritz Saemisch, a lesser master, former World Champion J. R. Capablanca blundered away a piece on the 9th move, and resigned some 50-odd moves later.


It was reported that Capablanca'ss wife or mistress (or possibly both) had just entered the room. In any event, the grandmaster LOST the game.

  • Giving this as an answer to this particular question doesn't seem quite fair to Saemisch. He was no Capablanca of course, but he was one of the top players in the world for many years. Chessmetrics, for instance, puts him at #14 in 1929 (while Capablanca was #2): chessmetrics.com/cm/DL/DL80.htm. That's an upset, sure, but one between world-class players, while this question asks for examples where "a much weaker player succeeded against an IM or GM."
    – ETD
    Oct 8 '12 at 21:12
  • 1
    Also I'm not sure he's much after piece blunders, especially if that's meant for study. Nov 18 '12 at 7:28

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

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