Petrosian could have held!

by Frederic Friedel
5/6/2020 – The Fischer-Petrosian Candidates match of 1971 in Buenos Aires was the third and final round in the eliminations to determine the challenger for the World Championship. Former World Champion Tigran Petrosian was considered undefeatable, and the match started 2½-2½. Then in game six Fischer scored a superb win that turned the match. Since then chess experts have discussed the question: could Petrosian have defended the game to a fateful draw? No conclusive answer was reached. Now endgame expert GM Karsten Müller – together with our readers – made a new attempt – has made a new attempt. And: we have a solution!

Master Class Vol.1: Bobby Fischer Master Class Vol.1: Bobby Fischer

No other World Champion was more infamous both inside and outside the chess world than Bobby Fischer. On this DVD, a team of experts shows you the winning techniques and strategies employed by the 11th World Champion.

Grandmaster Dorian Rogozenco delves into Fischer’s openings, and retraces the development of his repertoire. What variations did Fischer play, and what sources did he use to arm himself against the best Soviet players? Mihail Marin explains Fischer’s particular style and his special strategic talent in annotated games against Spassky, Taimanov and other greats. Karsten Müller is not just a leading international endgame expert, but also a true Fischer connoisseur.


The background of the question was described in our article "Could Petrosian have held against Fischer?" The critical game took place in Buenos Aires, Argentina. As a candidate for the World Championship Bobby Fischer had demolished two of the world's strongest grandmasters in the quarter and semifinals: he defeated Mark Taimanov and then Bent Larsen with perfect 6-0 scores. Now he faced former World Champion Tigran Petrosian, whose defensive skills many though would be Fischer's stumbling block.

The match started badly for Fischer, who was probably suffering from a cold and could not sleep in the noisy hotel. After five games the score was 2½-2½, and Petrosian arrived for game six looking relaxed and confident, while Fischer was pale and exhausted. The game, a tense battle, was adjourned after move 41, with Petrosian sealing the move.

In the above image, left, Fischer has just played 41..Kc5-b5. In the image on the right (and on our front page in colour), Petrosian is in the process of sealing his move

After this victory Fischer went on to win games 7, 8 and 9, ending the match with 6½-2½ points. After winning the final game he briefly acknowledged the cheers of the audience, and then slipped through a back door to go bowling with his friend Miguel Angel Quinteros until 3:30 a.m.

Here we asked for your assistance: Leading endgame expert GM Karsten Müller asked: "Should we assume that the adjourned position before the sealed move is a draw (as most sources and analysts do)? Then where did Petrosian go wrong in the game – or did Fischer perhaps also go wrong later, and Petrosian then made another mistake? I personally thnik that there is exactly one mistake, and it is not the sealed move 42.Ne2. But this of course might be wrong and is very deep."

Readers were asked to send analysis to Karsten Müller, who would evaluate all contributions. The best analysis would receive a ChessBase software product, signed by at least one World Champion. Something like the one shown here – which was awarded in a previous contest. And Karsten would show us what his own analysis had revealed – so far.

It was a draw!

By GM Karsten Müller

This cooperative analytical enterprise went really well. I had had many pleasant email discussions and analysis duels with several readers, and the ChessBase "Let's Check" service was a place where one could follow these analysis duels. One reader was so motivated that he checked my claims a whole day long, and then "resigned."

This famous deep endgame is still a borderline case even for modern engines. I had come to the following conclusions:

  1. The amazing 42.f4!! draws. I cannot say who found the move first, but Kasparov's analysis in On my Great Predessors contains errors.
  2. Petrosian's sealed move 42.Ne2 draws as well.
  3. 46.f3?! is an inaccuracy, but still playable.
  4. Only 48.Ra2+? was the decisive error. White then loses coordination and can not regroup the knight to d3 in a good way anymore. This seems to be the real key not the pawn moves 42.f4!! and 46.f3?!. This is one reason for the difficulty of the ending.

Fortunately this was confirmed by the analysis of several readers. Many thanks for your feedback!

[Event "Buenos Aires Match 1971"] [Site "Buenos Aires"] [Date "1971.10.17"] [Round "6"] [White "Petrosian, Tigran V"] [Black "Fischer, Bobby"] [Result "0-1"] [ECO "A06"] [WhiteElo "2640"] [BlackElo "2760"] [Annotator "Karsten Müller"] [SetUp "1"] [FEN "8/r6p/P4p2/1k2p1p1/3pP1P1/2bK4/R4P1P/6N1 w - - 0 42"] [PlyCount "50"] [EventDate "1971.09.30"] [EventType "match"] [EventRounds "9"] [EventCountry "ARG"] [SourceVersionDate "2016.11.12"] {The old riddle of Petrosian vs Fischer White's fortress seems to be more difficult to break down then previously assumed. Probably Petrosian made only one real mistake and Bobby Fischer made of course no mistake in this famous Fischer endgame:} 42. Ne2 {Petrosian's sealed move should draw as well, but White's fortress is not easy to hold.} ({The amazing direct blow} 42. f4 $3 { (Matanovic, Kasparov) draws as well as I had claimed many times in my publications. When proofreading his excellent work Learn from the Legends I had a long analysis duel with GM Mihail Marin on this question. The critical position was:} gxf4 43. g5 fxg5 44. Nf3 Rxa6 45. Rxa6 Kxa6 46. Nxg5 Ba5 { Now at first I had given the greedy} 47. Nxh7 $2 {in my e-mail, which is also given by Kasparov in On my Great Predecessors IV on page 425. But it indeed loses as Marin proofed, e.g.} ({I was already considerung to resign the analysis duel, when during the night in my sleep I had the idea to keep the h-pawns to keep the kingside closed:} 47. Nf3 Bc7 48. h3 {By the way: Christian Glaría, Zaragoza (Spain) discovered this independently and published it at} Kb5 49. Nd2 Kb4 50. Kc2 $11 {Then Mihail Marin agreed with me that 42.f4!! draws and changed that in the new edition of Learn from the Legends. Of course there are many more lines and a lot of publications, but as no convincing winning proof has been published the position should indeed be a draw.}) 47... Bd8 48. Nf8 Bf6 49. Nd7 Bg7 50. h4 Kb5 51. h5 Kc6 52. h6 Bh8 53. Nf8 Kd6 54. Ng6 Bf6 55. h7 Ke6 56. Nh4 (56. h8=Q Bxh8 57. Nxh8 Kf6 $19) 56... Kf7 57. Nf3 Kg6 58. Ke2 Kxh7 59. Ne1 Kg6 60. Kf3 Kf7 61. Ke2 Ke6 62. Nf3 Bd8 63. Kd3 Bc7 64. Ke2 Kf6 65. Ne1 Kg5 66. Kf3 Kh4 67. Nd3 Kh3 68. Nc5 Bb8 69. Nd3 Bd6 70. Ne1 Be7 71. Nd3 Bf6 72. Ne1 Kh2 73. Nd3 Kg1 74. Ke2 Kg2 75. Ne1+ Kg3 $19) 42... Ba5 43. Rb2+ Kxa6 44. Rb1 $6 ({Vincent Keymer's suggestion} 44. Nc1 $5 Rc7 45. Rc2 $11 {looks more reliable.}) 44... Rc7 45. Rb2 Be1 46. f3 $6 {This is inaccurate and readers have mailed that Petrosian considered this a losing error, but I can not give a source.} (46. Rb1 {is more accurate.}) 46... Ka5 (46... Rc5 47. Ra2+ Kb5 48. Rb2+ Kc6 49. Rc2 $11) 47. Rc2 Rb7 48. Ra2+ $2 {This check seems to be the only real mistake. It wastes time and worsens White's coordination.} (48. Nc1 {seems to defend:} Bc3 (48... Bb4 49. Ke2 Ba3 50. Nd3 $11) (48... Kb5 49. Rb2+ Bb4 50. Ke2 Kc4 51. Nd3 $11) 49. Kc4 Rb4+ (49... Rc7+ 50. Kb3 Kb5 51. Nd3 $11) 50. Kc5 Ra4 (50... Rb2 51. Rxb2 Bxb2 52. Nd3 Bc3 (52... Ba3+ 53. Kc4 Ka4 54. h3 Be7 55. Nc1 Ka3 56. Kd3 Kb2 57. Kd2 Bb4+ 58. Kd1 Kc3 59. Ke2 $11) 53. Kc4 Ka4 54. Nf2 $1 (54. Nc5+ $2 Ka3 55. Na6 h6 56. Kd3 Bb4 57. Nc7 Kb2 58. Nd5 Ba5 59. Ne7 Kc1 60. Ng8 Kd1 61. Nxh6 Bd2 62. Nf5 Be1 63. Ne7 Bf2 64. Ng8 Ke1 65. Nxf6 Be3 $19) 54... Bd2 55. Kd3 Be3 (55... Be1 56. Nd1 Kb3 57. Ke2 Bh4 58. Kd2 $11) 56. Nh1 Bf4 57. Kc2 Bxh2 58. Nf2 Bg3 59. Nd3 Kb5 60. Kb3 Kc6 61. Kc2 Kd6 62. Kd1 Ke6 63. Nb4 Kf7 64. Nd5 Bf4 65. Kc2 $11) 51. Nd3 Bb4+ 52. Kd5 Ra3 53. Kc4 $11) 48... Kb5 $19 { Now Black should be winning, but matters are still deep and difficult:} 49. Rb2+ ({Now it is too late for} 49. Nc1 Bb4 50. Ke2 {due to} h5 51. gxh5 (51. Rb2 hxg4 52. fxg4 Rh7 $19) 51... Rh7 52. Kd3 (52. Nd3 Rxh5 53. Rb2 Rxh2+ 54. Kd1 Rxb2 55. Nxb2 $19) 52... Rxh5 53. Rb2 Rh3 54. Na2 Rxf3+ 55. Ke2 Re3+ 56. Kf2 Rxe4 57. Rxb4+ Kc5 58. Rb8 Rh4 59. Kg2 e4 $19 {and Black's pawns prevail.}) (49. h3 Bb4 50. Ra1 Rc7 51. Ra2 Be7 52. Ra8 Kb4 $19) (49. Ra1 Bb4 50. h3 Rc7 51. Ra2 Be7 52. Ra1 Kb4 $19) 49... Bb4 50. Ra2 Rc7 51. Ra1 Rc8 52. Ra7 $6 { Many readers thought that this is a losing mistake. But I could convince them that White is lost anyway. One even analysed this for one whole day and then "resigned" the analysis duel. When Black pushes hard enough White's house always collapses sooner or later. Mihail Marin had already given a proof of this in the first edition of Learn from the Legends.} Ba5 $1 53. Rd7 (53. Rxh7 Bb6 54. Kd2 Ra8 55. Rb7 Ra3 $19) (53. Rf7 Bb6 54. Rxf6 Ra8 $19) 53... Bb6 54. Rd5+ Bc5 55. Nc1 Ka4 56. Rd7 Bb4 57. Ne2 (57. Ra7+ Ba5 58. Ne2 Rb8 59. Kc2 Rb5 60. h3 h6 $1 {Zugzwang.} 61. Ra8 Ka3 62. Kd1 d3 63. Nc3 Rb2 64. Rxa5+ Kb4 $19) 57... Kb3 58. Rb7 Ra8 59. Rxh7 Ra1 60. Nxd4+ (60. Ng3 Ra2 61. Nf1 Rf2 $19) 60... exd4 61. Kxd4 Rd1+ 62. Ke3 Bc5+ 63. Ke2 Rh1 64. h4 $6 Kc4 65. h5 Rh2+ 66. Ke1 Kd3 {Many thanks to Alex Fishbein and Mihail Marin for checking a first version of my analysis!} 0-1

Chess Endgames 14 - The golden guidelines of endgame play

Rules of thumb are the key to everything when you are having to set the correct course in a complex endgame. In this final DVD of his series on the endgame, our endgame specialist introduces you to the most important of these rules of thumb.

You probably know that in our replay boards there are a large number of functions you can use to really appreciate the analysis. We recently published a comprehensive tutorial on how to get the most out of the live broadcast game viewer. Learn about all the powerful features and buttons that make the ChessBase's replay one of the best watching experiences around.

One big advantage is that you can start an engine (fan icon) that will help you to analyse. You can get multiple lines of analysis by clicking the + button to the right of the engine analysis window. The "!" key, incidentally, shows you the threat in any position, which is incredibly useful in the case of unclear moves. And of course you can move pieces on the graphic chessboard, trying different lines of play. Note that your analysis, in which you can delete, move or promote lines, is stored: in the end you will find the game with your analysis in the cloud. So nothing is ever lost.

There is one more thing you can do. It is a lot of fun, but also a serious challenge: Click on the rook icon below the notation window. This will allow you the play the current position against Fritz, at your level of playing strength (e.g. "Club Player"), right here on the news page. Unfortunately it does not work on mobile phones or tablets, where the hardware and screen space is not sufficient to support this function.

It is very good that Zoran Petronijevic basically came to the same conclusions independently. The following points are worth mentioning:

  1. His analysis of 42.f4!! is detailed and very instructional
  2. He bases his drawing proof on another line than me, which is good as now both would have to be refuted: 48.Nc1! Bc3 49.Ra2+ (I continue with 49.Kc4) 49...Kb5 50.h3. I can not break that.

[Event "Buenos Aires (candidates match)"] [Site "?"] [Date "1971.??.??"] [Round "6"] [White "Petrosian, Tigran V"] [Black "Fischer, Robert James"] [Result "0-1"] [ECO "A06"] [Annotator "Petronijevic,Zoran"] [PlyCount "132"] [EventDate "1971.??.??"] [SourceVersionDate "2016.11.12"] 1. Nf3 c5 2. b3 d5 3. Bb2 f6 4. c4 d4 5. d3 e5 6. e3 Ne7 7. Be2 Nec6 8. Nbd2 Be7 9. O-O O-O 10. e4 a6 11. Ne1 b5 12. Bg4 Bxg4 13. Qxg4 Qc8 14. Qe2 Nd7 15. Nc2 Rb8 16. Rfc1 Qe8 17. Ba3 Bd6 18. Ne1 g6 19. cxb5 axb5 20. Bb2 Nb6 21. Nef3 Ra8 22. a3 Na5 23. Qd1 Qf7 24. a4 bxa4 25. bxa4 c4 26. dxc4 Nbxc4 27. Nxc4 Nxc4 28. Qe2 Nxb2 29. Qxb2 Rfb8 30. Qa2 Bb4 31. Qxf7+ Kxf7 32. Rc7+ Ke6 33. g4 Bc3 34. Ra2 Rc8 35. Rxc8 Rxc8 36. a5 Ra8 37. a6 Ra7 38. Kf1 g5 39. Ke2 Kd6 40. Kd3 Kc5 41. Ng1 Kb5 42. Ne2 $1 {The game move doesn't lose and also deserves an exclamation mark.} (42. f4 {According Kasparov this move deserves "!!". However, after checking his analyzes, although this move leads to the draw (but not on way as Kasparov gave), the game move also leads to the draw, that's why i think athat f4 deserves only one exclamation mark.} gxf4 $1 43. g5 $1 fxg5 44. Nf3 {[#] The best practical try is} Rxa6 $1 (44... g4 {Matanovic's analyze from Chess informant. However, this move isnt danger fof White at all.} 45. Nxe5 h5 46. Ng6 f3 (46... Rxa6 47. Rxa6 Kxa6 48. Nxf4 h4 49. e5 Kb7 50. e6 Kc8 51. Ng6 g3 52. hxg3 hxg3 53. Nf4 $11) 47. e5 $1 Be1 (47... Rxa6 {Also leds to the draw} 48. Rxa6 Kxa6 49. e6 Be1 50. e7 f2 51. e8=Q f1=Q+ 52. Kxd4 $11) 48. e6 $1 f2 49. Ke2 Rxa6 ({In case of} 49... d3+ {White can make a draw after: } 50. Kf1 Rxa6 51. Rxa6 Kxa6 52. e7 d2 53. e8=Q d1=Q 54. Qe2+ {It isnt only move which makes a draw.} Qxe2+ 55. Kxe2 {transposing to a main line.}) 50. Rxa6 d3+ 51. Kf1 Kxa6 52. e7 d2 53. e8=Q d1=Q 54. Qe2+ ({Kasparov analyze. He gave "!" to his move, however, White can also make a draw after} 54. Qc6+ Ka5 55. Qe4 $11) 54... Qxe2+ 55. Kxe2 h4 56. Kf1 (56. Ne5 {Also gives draw} g3 57. hxg3 hxg3 58. Kf1 $11) (56. Nxh4 {also leads to the draw} f1=Q+ 57. Kxf1 Bxh4 58. Ke2 $1) 56... h3 (56... g3 57. h3 $1 $11) 57. Nf4 g3 58. Nxh3 gxh2 59. Nxf2 Bxf2 60. Kg2 $1 $11) (44... h6 {leads to relatively easy draw after} 45. Nxe5 Rxa6 46. Rxa6 Kxa6 47. Ng4 $1 f3 48. e5 Kb6 49. e6 Kc7 50. Ke4 Kd6 51. Kxf3 h5 52. Nf6 $1 Kxe6 53. Nxh5 $11) (44... h5 45. Nxg5 Ba5 46. Nf3 Bc7 47. h3 $11) 45. Rxa6 (45. Rg2 $2 {loses:} h6 46. Nxe5 Ra3 47. Nc4 Rb3 48. Rc2 Ba1+ 49. Kd2 Rh3 $19) 45... Kxa6 46. Nxg5 (46. Nxe5 $2 Kb5 $19) 46... Ba5 $1 ({After} 46... Bb4 $2 {White can attain easy draw after} 47. Nf7 $1 $11) 47. h3 $3 {Great idea of making a fortress.} (47. Nxh7 $2 {Suggested by Kasparov, loses due to} Bd8 $1 ({In case of} 47... f3 $2 {White can equaize after} 48. Nf6 f2 49. Ke2 Be1 50. Ng4 d3+ 51. Kf1 Kb5 (51... d2 52. Ne3 $1) 52. h4 (52. Nxf2 {also leads to draw.}) 52... Kc5 53. Nxf2 $11) 48. Nf8 Bf6 $1 49. Nd7 Bh8 50. Nc5+ (50. h4 Kb5 51. h5 Kc6 {Blacks idea is to catch the pawn.} 52. Nb8+ (52. Nf8 Bf6 53. h6 Kd6 54. Ng6 Ke6 55. h7 Kf7 56. Nh4 Kg7 57. Nf3 Kxh7 $19 {This position can appear in many lines and we should evaluate it properly. My evaluation is that this positon is lost for White. White cannot avoid the zugzwang (I have analyzed this type of the position on other place).}) 52... Kd6 53. h6 Ke6 54. Ke2 Kf6 55. Na6 Kg5 56. Kf3 Kxh6 {This position is lost. soon or later Black will put White in zugzwang.} 57. Nb4 {And now Blacks idea is to put opponent in Zugzwang.} Kg5 58. Nd3 Kh5 59. Nf2 Kh4 60. Kg2 (60. Nd3 {loses after} Kh3 61. Kf2 Kg4 $19) 60... Bg7 61. Kf3 Bf8 62. Kg2 Bd6 63. Kf3 Bc7 64. Kg2 Kg5 65. Kf3 Kf6 66. Ke2 Ke6 67. Nh3 Bb6 68. Kd3 (68. Ng5+ Kf6 69. Nf3 Bc7 70. Kd3 Kg6 $19) 68... Kf6 69. Ng1 Bc7 70. Nf3 Kg6 71. Ke2 Kh5 72. Ne1 Kg4 73. Kf2 d3 74. Nxd3 Bb6+ $19 {Lomonosov}) 50... Kb5 51. Nb3 f3 $1 {Black is winning. Let's see possible line:} (51... Kb4 $2 {Kasparov's move is a clear mistake and leads to the draw.} 52. Nd2 $1 Bf6 53. Nf3 Kb3 54. h3 Kb2 55. Kd2 $1 {thous making a fortress - Kasparov.} (55. Ne1 {also leads to draw} Kc1 56. Ke2 $11)) (51... Kc6 $2 52. Nd2 Kd6 53. h3 $11) 52. Nd2 f2 53. h3 (53. Ke2 Kb4 54. Kxf2 Kc3 55. Ke2 Kc2 56. Nc4 d3+ 57. Ke3 Bg7 58. Na3+ Kb3 59. Nb5 Kb4 60. Nc7 Kc4 61. Nd5 Bh6+ 62. Kf2 Bd2 ({Or} 62... Bf8 63. Ke1 Kd4 64. Nf6 Be7 65. Nd5 Bh4+ 66. Kd1 Kxe4 $19) 63. h3 Kb3 64. h4 Kc2 65. h5 Kc1 $19) 53... Bf6 54. Nf1 Bg5 55. Ke2 Kc4 56. Kxf2 Kc3 57. h4 Bxh4+ 58. Kf3 Bg5 $19 {Lomonosov}) ({Also possible is} 47. Nf3 Bc7 48. h3 $11 {transposing to the main line.}) 47... Kb5 48. Nf3 Bc7 {Although being a two pawns down, despite the fact that engine gives -2.77, this position is a draw! White cannot penetrate neither on the King side nor on the Queen side.} 49. Nd2 Kb4 (49... Kc5 50. Nf3 $11) 50. Kc2 $1 $11) 42... Ba5 43. Rb2+ ({The game move leads to draw. However, either} 43. Nc1 {leads to draw. White's main idea is to trade Rooks after which the position is easy draw due to already known fortress. Lets see possible lines:} Rxa6 ({After} 43... Kxa6 44. Nb3 {It is an easy draw:} Kb5 45. Ra1 Bb6 (45... Kb4 46. Rb1 Ka3 47. Nxa5 Rxa5 48. Rb6 $11) (45... Ra8 46. f3 $11) 46. Rxa7 Bxa7 47. Nc1 Bc5 48. Ke2 Kc4 49. Nd3 $11 {Fortress.}) 44. Nb3 ({I didnt find win for Black even after} 44. Ke2 {although it is hard to believe.} Bc3 ({After} 44... Rc6 45. Rb2+ Kc4 46. Nd3 Bc3 47. Rb7 h6 (47... Ra6 48. Rc7+ $11) 48. Ra7 Rb6 49. Rc7+ Kb3 50. Rc8 Ka4 51. Rc5 Ka3 52. Rc8 Ba5 53. Ra8 Ka4 54. Rc8 { In my opinion Black cannot break Whites fortress.}) (44... Bb6 45. Rxa6 Kxa6 46. Nd3 Kb5 47. f3 Ka4 48. Nf2 {In my opinion theres no way to break fortress:} Kb3 49. Nd3 Ba5 50. Nf2 Kc2 51. Nd3 Bc3 52. Nf2 Kb3 53. Nd3 Kc4 54. Nc1 Bb4 55. Nd3 Kc3 $11) 45. Rxa6 Kxa6 46. Nd3 {with already known fortress.}) 44... Kb4 ( 44... Bb4 {gives an easy draw} 45. Rxa6 Kxa6 46. Nc1 $11) 45. Nc1 Rd6 46. Ke2 Bd8 47. Nd3+ Kc3 48. Ra1 {Black cannot break the fortress.}) ({In my opinion} 43. Ng3 {should lose after} Rxa6 {the Knight is more important on c1 than on g3.}) 43... Kxa6 44. Rb1 ({Probably easiest was} 44. Nc1 {it is hard for Black to untangle position of his pieces on the Queen side.} Rc7 (44... Rf7 45. Ke2 Rc7 46. Nd3 Rc3 47. Rb8 h6 48. Rb1 Rc2+ 49. Kf3 Rc3 50. Ke2 {White should survive.}) 45. Rc2 Rxc2 (45... Rb7 46. Rc6+) 46. Kxc2 Kb5 47. Kd3 Be1 48. f3 $11) 44... Rc7 ({Interesting is} 44... Ra8 45. Nc1 Rc8 46. Na2 Rc6 47. Rb8 { and Black cannot penetrate.}) 45. Rb2 ({Although the game move gives a draw, easier was after} 45. Rc1 Rb7 (45... Rxc1 46. Nxc1 $11) 46. Rc6+ Ka7 47. Kc2 $11) 45... Be1 (45... Bb6 46. Rc2 Rf7 47. Nc1 Ba5 48. Kc4 {I dont see how Black can improve.}) 46. f3 $6 ({Easier was} 46. Rb1 $1 {after which the position is even:} Bxf2 (46... Ba5 47. Rc1 Rf7 (47... Rb7 48. Rc6+ Ka7 (48... Kb5 49. Rxf6 Ka4 50. Rd6 Rb3+ 51. Kc4 Rb2 52. Nxd4 exd4 53. e5 Rxf2 54. Rh6 Re2 55. Kxd4 Bc7 56. h4 gxh4 57. Rxh4 Bxe5+ 58. Kd3 $11) 49. Kc2 $11) 48. Rb1 Rf8 49. Nc1 Rh8 50. Nb3 Rc8 51. Rc1 $11) 47. Rf1 Be3 48. Rxf6+ Ka7 (48... Ka5 49. Re6 Rc5 50. h3 Rb5 (50... Bf4 51. Nxf4 gxf4 52. Re7 $11) 51. Kc4 Rc5+ 52. Kd3 Rb5 53. Kc4 Rb4+ 54. Kd3 Ra4 55. Rxe5+ Kb4 56. Nxd4 Ra3+ 57. Ke2 Bxd4 58. Rxg5 Rxh3 59. Rh5 Re3+ 60. Kd2 Rxe4 61. Rxh7 $11) 49. Re6 Rc5 50. Re7+ Kb6 51. Re6+ Ka5 52. h3 $11 Rb5 53. Kc4 Rb4+ 54. Kd3 Ra4 55. Rxe5+ Kb4 56. Nxd4 $11) 46... Ka5 47. Rc2 Rb7 48. Ra2+ $2 ({Decisive mistake. According to my analyze, White should hold after:} 48. Nc1 Bc3 49. Ra2+ Kb5 50. h3 $1 Rb8 (50... h5 51. gxh5 { leads to the draw} Rh7 52. Ra8 Rxh5 53. Rb8+ Kc5 54. Nb3+ Kc6 55. Rf8 Rxh3 56. Rxf6+ Kb5 57. Rf5 Kb4 58. Nd2 g4 59. Nc4 Rh2 60. Rxe5 gxf3 61. Re7 f2 62. Rb7+ Kc5 63. Rc7+ Kb5 64. Rb7+ Ka6 65. Rf7 Rh3+ 66. Ke2 Be1 67. Ne5 Re3+ 68. Kd1 d3 69. Ng4 Re2 70. Nxf2 Bxf2 71. Rd7 $11) (50... Bb4 51. Ke2 Bc3 52. Kd3) 51. Ke2 Rh8 (51... Kc4 52. Ra4+ Bb4 53. Nd3 Rb7 (53... Kb3 54. Ra6) 54. Ra1) 52. Nd3 h5 53. gxh5 Rxh5 54. Nf2 $11) 48... Kb5 $19 49. Rb2+ (49. Nc1 {also losses.} Bb4 50. Ne2 Rc7 51. Ra1 (51. Rc2 Ra7 $19) 51... Bd6 52. Rb1+ (52. Kd2 Be7 53. Ra2 ( 53. Kd3 Kb4) 53... Kb4 54. Rc2 (54. Kd3 Kb3 55. Ra1 Kb2 $19) 54... Ra7 55. Nc1 Kb5 56. Rc8 Ra3 57. Nd3 Bd6 58. h3 h6 59. Kc2 Ra2+ 60. Kd1 Rh2 61. Rc2 Rxh3 $19 ) (52. Ra8 Kb4 53. Ra6 Be7 54. Ra1 Kb3 55. Rb1+ Ka3 56. Rc1 Rb7 57. Rf1 Ka4 58. Ra1+ Ba3 $19) 52... Ka4 53. Rb6 Bc5 54. Ra6+ (54. Rxf6 Kb5 55. Rf5 Ra7 56. Rxe5 Ra3+ 57. Kd2 Kc4 $19) 54... Kb3 55. Ra1 Bb4 56. Rb1+ Ka3 57. Rc1 Rxc1 58. Nxc1 Kb2 59. Ne2 Be1 60. f4 (60. Ng1 Kc1 $19) 60... gxf4 61. Ng1 Ba5 62. Nf3 Bc7 63. Kd2 Kb1 64. Kd1 Ba5 $19) (49. Ra1 Bb4 50. Ra2 Rc7 51. Ra1 Bf8 52. Ra2 Kb4 53. Rb2+ Ka3 54. Rc2 Rb7 55. Nc1 Ka4 56. Ke2 Rb1 57. Nd3 Rh1 58. Nf2 Rxh2 $19) 49... Bb4 {With the idea of playing Ra7.} 50. Ra2 Rc7 51. Ra1 Rc8 52. Ra7 ({ Loses on the spot. More stubborn was} 52. Ra2) 52... Ba5 53. Rd7 (53. Rxh7 Bb6 {with next Ra8.}) 53... Bb6 54. Rd5+ Bc5 55. Nc1 Ka4 56. Rd7 Bb4 57. Ne2 Kb3 58. Rb7 (58. Rxh7 Ra8 59. Nxd4+ exd4 60. h4 gxh4 61. Kxd4 Rd8+ 62. Ke3 Bd2+ 63. Ke2 Bg5 $19) 58... Ra8 59. Rxh7 Ra1 60. Nxd4+ exd4 61. Kxd4 Rd1+ 62. Ke3 Bc5+ 63. Ke2 Rh1 64. h4 Kc4 65. h5 Rh2+ 66. Ke1 Kd3 0-1

There is one clear winner of the analysis competition: Zoran Petronijevic (ChessBase handle: zoranp). He basically came to the same conclusions I had reached, and supplied much of the analytical justification for them. I did not find a single error in his analysis.

In all other submissions I found at least one problem, and also most often not all relevant points were mentioned.

Congratulations, Zoran! We will send you your prize as soon as Covid-19 allows us to visit our office freely.

Editor-in-Chief emeritus of the ChessBase News page. Studied Philosophy and Linguistics at the University of Hamburg and Oxford, graduating with a thesis on speech act theory and moral language. He started a university career but switched to science journalism, producing documentaries for German TV. In 1986 he co-founded ChessBase.


Rules for reader comments


Not registered yet? Register

Karsten Müller Karsten Müller 5/16/2020 12:28
Christian Glaría has found the answer to my question on the results of Petrosian's adjournment analysis:
"I think you asked for sources on Petrosian comments after the 6th game.
In "Russians vs Fischer" (p.288), Petrosian is quoted saying: "Before
leaving for the adjournment session, I agreed to play f3, although
during my night-time analysis I had definitely established that it was
precisely this move that should not be made. I am the only one to blame
for the result.""
Robert Cuadros Robert Cuadros 5/14/2020 01:04
Congratulations to Zoran Petronijevic and again thank you GM Karsten Müller for giving us the opportunity to participate in this challenge.
catalanFischer catalanFischer 5/7/2020 09:21
44,,,,Kb4 draws in the spot.
I have been going over the Hort/Jansa analysis and 42.Ne2 is drawing as 42.f4. The difference is that f4 is more complicated and have more only moves, so Ne2 was a very reasonable choice.
White to draw in the spot had to exchange rooks - which Fischer avoided skillfully - or at least bring the rook to the defence, since it was clear was fighting for 2 results, draw or lose. The maneuver Ra7 followed by Rxh7 was simply bad strategy and black took advantage,
What impresses me of this endgame is the high accuracy of the players back in 1971, when engines didnt existed.
And congratulations Zoran for a well deserved draw.
Frits Fritschy Frits Fritschy 5/7/2020 01:31
I was wondering about 42 f4 gxf4 43 g5 fxg5 44 Nf3 Kb4!? which seemed quite logical: with the ‘wrong’ (because of the h-pawn) bishop and the possibility of a white fortress, you would like to conquer the a-pawn without exchanging rooks. And because white has to spend time regaining pawns, it looked white has time to attack the rook.
Now 45 Kc2 is too passive, after 45... d3+ 46 Kxd3 g4 47 Nd2 Bxd2 48 Kxd2 black just returns: 48... Kb5 with an easy win. And 45 Nxg5 Kb3 46 Rf2 (after 46 Rg2 Nf3 47 Nf3 Ra1 48 Nxe5 Re1, threatening mate, wins: 49 Nc4 f3 50 Rf2 Rd1+ 51 Nd2 Rxd2+ 52 Rxd2 Bxd2 53 Kxd2 f2 54 Ke2 Kc2) 46... Rxa6 47 Nf3 Ra1 48 Nxe5 Rd1+ 49 Ke2 Re1+ 50 Kf3 d3 51 Nxd3 Re3+ 52 Kxf4 Rxd3 should be a win.
That leaves 45 Nxe5 Kb3. I analysed both 46 Rg2 Rxa6 47 Rxg5 Ra2 48 Nf3 Rf2 49 Rb5+ Ka4 50 Nxd4 Bxd4 51 Rb7 Be3 52 Rxh7 Rd2+ 53 Kc3 f3 54 e5 f2 55 Rf7 Rd1 56 e6 Bc5! and 46 Rg2 Rxa6 47 Nf3 Ra1 48 Nxd4+ Bxd4 49 Kxd4 f3 50 Rf2 g4 51 Ke3 h5! 52 h3 (52 Kf4 Kc4 53 h3 Rg1 54 hxg4 hxg4 55 e5 Kd5 and wins) 52... g3! 53 Rxf3 h4 54 e5 Kc3 55 Ke4+ Kc4 56 Rf4 Kc5 57 Rg4 Ra4+ and wins.
But only then I discovered that after 45 Nxe5 Kb3 the ‘Zwischenzug’ 46 Nc6! Ra8 47 Rf2 (or Rg2) Rxa6 48 Nxd4+ Bxd4 49 Kxd4 secures the draw. Too bad, no place in heaven this time.
brian8871 brian8871 5/7/2020 10:46
Any chess game could have ended in a draw if someone hadn't made an error at some point.
chessbibliophile chessbibliophile 5/7/2020 03:25
Was waiting for this. Real progress on a position that had not given away all its secrets.
Keith Homeyard Keith Homeyard 5/7/2020 12:15
All of this presumably engine based analysis is very well but considering , even with adjournment in pre-engine times, how many players today would have fared better?
twamers twamers 5/6/2020 12:52
Well done to Zoran for winning the competition - and to Karsten thank you for running it.

Of course "drawn" endings like these are always so difficult to hold when playing to a time limit, when the nature of the game is so important (candidates match) and adds huge pressure and of course when your opponent is the mighty Bobby Fischer who you know will just play on and on (and at very little risk to his own position) to keep the pressure on.