I ran the final position of the 1997 game Kasparov vs Deep Blue. Kasparov resigned the final position. But deep analysis with Deep Fritz 12 shows that after Qe3 Kasparov could have held the game despite being a bishop down! In the main variation Kasparov draws by perpetual check. Has this been pointed out before? Is Deep Fritz wrong?

Frederic Friedel of Chessbase apparently discovered what he thought was a sure draw, and showed it to Yuri Dokhoian the day after the game.  The two of them showed it to Garry at lunch.

Thx Riga, I guess I had forgotten about that.

Many years later I and a fellow other chess player used engine analysis to discover the game was NOT a draw. Both Frederic and Deep Blue were wrong. We posted our tandem work on forums and nobody could find any holes in it. I then sent the work to chessbase, but they refused to acknowledge we had proven everybody wrong about the draw and did nothing to post any news about it.

I'll go retieve the analysis from the forum I posted it on if you guys want to check it out.

"I'll go retieve the analysis from the forum I posted it on if you guys want to check it out."

* Draws up a sandbag *

I've found it, however I think I'm going to make the analysis into my next blog. There's some really fascinating stuff, and my computer is now several times more powerful than the original computer I used, yet the analysis still shows white being able to avoid the perpetual that everyone assumed Kasparov missed when he resigned.

I'll post here again when I've finished my article on it.

looking forward to it.

Blog completed and submitted. Complete with diagrams and move sequences. I want to state for the record that this was all work I did back in early 2007 (as part of a 10th anniversary piece). I recall I was even able to convince Kingcrusher (owner of another CC site) and he may have done some videos on this same subject after reading my original postings about this.

Edit: Somebody tried to scoop the blog right before mine was finished. However, their analysis was faulty. They missed the critical 51. Ra1 after black's Re8, preventing the perpetual.

The question was not whetehr or not Black had a perpetual (in case White tried to hold on to the material) but whether Kasparov had a draw. A deep position analysis by Deep Fritz 12 hows that with best play, White only had an advantage of about half a pawn, far from enough to win (about 2 pawns). Starting from your positino after 4.d6, although both the Queen and rook are attacked, Black has a check to free the Queen. Here is the analysis

Interesting.  From a 2007 post by Zahl on Chessninja:

"I have been running various tests with the new Rybka version 2.3.2a for the "Computer analysis of greatest ever" project and decided to see how it sees the famous Game 2 from the 1997 Kasparov - Deep Blue match.

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 h6 10. d4 Re8 11. Nbd2 Bf8 12. Nf1 Bd7 13. Ng3 Na5 14. Bc2 c5 15. b3 Nc6 16. d5 Ne7 17. Be3 Ng6 18. Qd2 Nh7 19. a4 Nh4 20. Nxh4 Qxh4 21. Qe2 Qd8 22. b4 Qc7 23. Rec1 c4 24. Ra3 Rec8 25. Rca1 Qd8 26. f4 Nf6 27. fxe5 dxe5 28. Qf1 Ne8 29. Qf2 Nd6 30. Bb6 Qe8 31. R3a2 Be7 32. Bc5 Bf8 33. Nf5 Bxf5 34. exf5 f6 35. Bxd6 Bxd6 36. axb5 axb5 37. Be4 Rxa2 38. Qxa2 Qd7 39. Qa7 Rc7 40. Qb6 Rb7 41. Ra8+ Kf7 42. Qa6 Qc7 43. Qc6 Qb6+ 44. Kf1 Rb8 45. Ra6 1-0

and here Kasparov resigned only to learn that after 45...Qe3 46.Qxd6 Re8 47.h4 h5 etc. the position is drawn by repetition.

However, Rybka says that's not all there is to it. After 45...Qe3 46.Qxd6 Re8 white has 47.Qd7+ Re7 48.Qc6 Qxe4 49.d6, hitting the rook and the queen at the same time. Black can try 49...Qd3+ 50.Kg1 Re8, but after 51.Ra1 there is no perpetual and the attack continues."

@ henri5:

No, I'm pointing out that there was no FORCED draw by perpetual as everyone claimed at the time the game was played. After my sequence, the game MIGHT still be a draw, but it is a very difficult battle for black to prove it if it is.

Edit: and Zahl is the other person I worked with in trying to find a solution for black. Honestly though, It still remains somewhat unclear.

Kingscrusher's video on youtube.

At first, I thought that there might be a draw with 48.Qc6 Qxe4 49.d6 Qxf5+ 50.Kg1 Rd7 51.Qxb5 Qe6 52.Qc5 Kg6 53.Rc6 e4 (smussbrighton on youtube) 54.b5 e3 55.Qxc4 Qe5 56.Qe2 (capabal on chessgames.com) 56...f5 (JamesMazur on chessgames.com)  57.c4 Ra7, but then JamesMazur came up with 54.Qd4 where it looks like white is winning again.

There is only a forced draw if White plays h4, which apparently was the move that Deep Blue was expected to play. It is not for Black to prove  that he has a draw, rather the onus is on White to show that he has a win, and he does not. Bottom line is that the game remained to be played with almost equal chances, and that it is still true that Kasparov resigned an equal game, forced draw or not.

And to make things even better, Stockfish (which apparently is stronger than Fritz12) finds that with best play the White advantage evaporates the position is a dead draw!And since this is an endgame position, Black can easily force a draw by exchanging a couple of pieces into a known  drawn endgame. Q.E.D.

It looks like no one ever answered MartinII's drawing line from the chessgames.com discusssion:

45...Qe3 46. Qxd6 Re8 47. Qd7+ Re7 48. Qc6 Qxe4 49. d6 Qd3+ 50. Kg1 Re8 51. Ra1 h5 52. Qxb5 Rd8 53. Kh2 h4 54. Qc6 Kg8 55. b5 Qg3+ 56. Kh1 Qxc3 57. Rd1 Qc2 58. Re1 Qd3 59. Re4 Qxd6 60. Qxc4+ Qd5 61. Rxh4 Qxc4 62. Rxc4 Rd1+ 63. Kh2 Rb1 =