Lasker, seeing the win slipping away, seemed to have lost his equanimity (a very rare occasion for him), also started slipping, and fell into a mating net.

"What's the difference between Schlechter and a pit bull?"

<INTRODUCTION>

Schlechter vs Lasker, 1910 is the <fifth game of the 1910 title match>.

Quantitative mapping of this game between these players is below. Figures in brackets immediately after each move are the corrected engine evaluations generated on the return slide, which smoothed out many, but not all fluctuation in the engine’s evaluations. The complexity of some variations was very likely too great to enable a fuller reconciliation from the initial reverse slide. <General methods used are described in the bio at User: Bridgeburner

The evaluation values in the opening come at the end of a full return slide back to the starting position from the last move of the game. Engine preferences are included throughout the game where they differ from players preferences.

<Summary>

The first four games of this match were drawn, and it looked initially as if this was to be like game 2, an uneventful draw, (unlike games 1, 3 and 4 which were hard fought).

This is a game of two parts. The first part sees White, instead of attempting to press for a further advantage to that provided by Black’s conservative opening strategy (Ruy Lopez Berlin Defense. Closed Bernstein Variation (C66)), simplifying quickly into a Q+R+7p ending/middlegame, in which both sides maneuver with equality, barely disturbing the seismometer, until about move 35, at which stage one would have expected the game to be called off as another draw.

In the second part of the game, maybe one of the players refused a draw offer at some stage, or perhaps fatigue from maneuvering in a difficult heavy piece ending kicked in (historians may know the answer), but at move 36, White’s play starts to become erratic, conceding ground and advantage. Then White blunders with <44. Qb4>, leaving Black with a winning advantage. However, Black is unable to force the difficult win in a heavy piece ending, making plausible but not decisive moves, firstly by allowing the win to slip from his fingers (by not fully securing the safety of his King with <47…Kd7> instead of the highly plausible <47…c4>), before inexplicably blundering with <54…Qc5> to allow White a mating attack.

<THE GAME>

<1. e4> ( 0.31) <1…e5> ( 0.31)

<2. Nf3> ( 0.31) <2…Nc6> ( 0.31)

<3. Bb5> ( 0.32) <3…Nf6> ( 0.31)

<Engine preference>: <3…a6> ( 0.32)

<4. 0-0> ( 0.31) <4…d6> ( 0.44)

<Engine preference>: <4…Nxe4> ( 0.31)

<5. d4> ( 0.44) <5…Bd7> ( 0.44)

<6. Nc3> ( 0.44) <6…Be7> ( 0.57)

<7. Bg5> (=0.25)

<First three engine preferences>: <7. d5> ( 0.57); <7. Be3> (( 0.43); <7. Be2> ( 0.43)

<7…0-0> ( 0.52)

<Engine preference>: <7…exd4> (=0.25); <7…h6> ( 0.36); <7…0-0> ( 0.42)

<8. dxe5> ( 0.24)

<Engine preferences>: <8. d5> ( 0.52); <8. Bxc6> ( 0.42)

<8…Nxe5> ( 0.24)

Capablanca commented that <Of course, if 8…dxe5 9.Bxc6 would win a pawn.>

<9. Bxd7> ( 0.11)

<Engine preference>: <9. Nxe5> ( 0.24)

<9…Nfxd7> (=0.11)

<10. Bxe7> (=0.11) <10…Nxf3+> (=0.11)

Capablanca: <If 10…Qe7 probably 11. Nxd4>

<11. Qxf3> (=0.11) <11…Qxe7> (=0.14)

<12. Nd5> (=-0.03)

<Engine preferences>: <12. Rad1> (=0.14); <12. Rfe1> (=0.14); <12. Qe3> (=0.06)

<12…Qd8> (=0.19)

<Engine preference>: <12…Qe5> (=-0.03)

<13. Rad1> (=0.19) <13…Re8> (=0.19)

<14. Rfe1> (=0.21) <14…Nb6> (=0.18)

<15. Qc3> (=0.18) <15…Nxd5> (=0.18)

<16. Rxd5> (=0.18) <16…Re6> (=0.09)

<17. Rd3> (=0.05)

<Engine preference>: <17. Rb5> (=0.09)

<17…Qe7> (=0.05)

<18. Rg3> (=0.00)

<Engine preference>: <18. f3> (=0.05)

<18…Rg6> (=0.14)

<Engine preferences>: <18…f6> (=0.00); <18. g6> (=0.00)

<19. Ree3> (=0.00) <19…Re8> (=0.00)

<20. h3> (-=0.01) <20…Kf8> (=0.21)

<First three engine preferences>: <20…Rxg3> (=-0.01); <20…c6> (=0.07); <20…Qd7> (=0.11)

<21. Rxg6> (=0.21) <21…hxg6> (=0.15)

<22. Qb4> (=0.11)

<First three engine preferences>: <22. Qa5> (=0.15); <22. b3> (=0.13); <22. Qb3> (=0.12)

<22…c6> (=0.11)

<23. Qa3> (=0.00)

<First three engine preferences>: <23. a3> (=0.11); <23. Kf1> (=0.06); <23. Qb3> (=0.02)

<23…a6> (=0.05)

<Engine preference>: <23…Qe5> (=0.00)

<24. Qb3> (=0.05) <24…Rd8> (=0.05)

<25. c4> (=0.03) <25…Rd7> (=0.02)

<26. Qd1> (=0.02) <26…Qe5> (=0.02)

<27. Qg4> (=-0.08)

<Engine preferences>: <27. Qd2> (=0.02; <27. b3> (=0.00)

<27…Ke8> (=0.00)

<28. Qe2> (=0.00) <28…Kd8> (=0.00)

<29. Qd2> (=0.00) <29…Kc7> (=0.01)

Capablanca: <White, apparently satisfied with his position has allowed the black king to move over to the queenside, without making any attempt to stop him. White now starts and advance of the queen side pawns, which is checked by black at once.>

<30. a3> (=0.00) <30…Re7> (=0.00)

<31. b4> (=0.01) <31…b5> (=0.00)

<32. cxb5> (=0.02) <32…axb5> (=0.03)

<Black’s strategy has been of a very high order. He has now a very fine game.> - Capablanca

<33. g3> (=0.00) <33…g5> (=0.01)

<34. Kg2> (=0.00) <34…Re8> ( 0.37)

<35. Qd1> (=0.00) <35…f6> ( 0.37)

<36. Qb3> ( 0.32)

<First three engine preferences>: <36. a4> ( 0.37); <36. Qd2> (=0.00); <36. Qd3> (=-0.03)

<35…Qe6> ( 0.32)

<37. Qd1> ( 0.32 ) <37…Rh8> (=-0.03 )

<Engine preference>: <37…Qc4> ( 0.32)

<38. g4> (=-0.03 ) <38…Qc4> (=-0.03)

<39. a4> ( 0.48)

<First three engine preferences>: <39. Qd2> (=-0.03); <39. f3> (=-0.03); <39. Kg3> (=-0.21)

<Black’s position was becoming too alarmingly strong. White prefers to give up a pawn and go for the Black king, rather than be boxed in without a chance to move anything.> - Capablanca

<39…Qxb4> (=-0.04)

<Engine preference> <39…Rb8> ( 0.48)

<39…bxa4, and if 40. Qxa4 Rb8, should be considered. It looks very much as though Black would win. The position is worth studying. By playing 40…Rb8, Black threatens to play 41…Rxb4 and then 42…Rb1, with a murderous check in sight at f1.> - Capablanca

<Note>: According to the engine, after <39...bxa4 40.Qxa4 Rb8 41.Qa5+ =0.00>

<40. axb5> (=-0.04) <40…Qb5> (=-0.04)

<41. Rb3> (=-0.18)

<Engine preference> <41. Qa1> (=-0.04)

<41…Qa6> (=-0.16)

<42. Qd4> (=-0.24)

<First three engine preferences>: <42. Qb1> (-=0.16); <42. Rb4> (=-0.18); <42. Rb2> (=-0.18)

<42…Re8> (=-0.24)

<If Black were satisfied with a draw, 42…Rb8 would have accomplished it. The champion’s play is better, as he could have drawn if he wished up to up to the fifty-fourth move.> - Capablanca

<43. Rb1> ( 0.79) <43…Re5> ( 0.79)

<44. Qb4> ( 1.68)

<First three engine preferences>: <44. Qc3> ( 0.78); <44. Rb2> ( 0.79); <44. Rc1> ( 0.89)

<44…Qb5> ( 1.68)

<45. Qe1> ( 1.68) <45…Qd3> ( 1.67)

<Engine preference>: <45…Qa4> ( 1.68)

<46. Rb4> ( 1.89)

<Engine preference>: <46. f3> ( 1.67)

<46…c5> ( 1.89)

<46…Rb5 gave Black some chance of winning.> - Capablanca

<47. Ra4> ( 1.89) <47…c4> ( 1.22)

<Engine preference>: <47…Kd7> ( 1.89) <<<47…c4>> is a blunder for the purposes of this analysis as it destroys Black’s winning advantage by causing the evaluation to shift across the 1.40 threshold. Accordingly, it is weighted at <<2.0>>, for an accumulated game weighting of <<4.5>>>.

<48. Qa1> ( 1.22) <48…Qe4+> ( 1.22)

<49. Kh2> ( 1.22) <49…Rb5> ( 1.09)

<Engine preferences>: <49…Qf4 and 49…Kd7> ( 1.22)

<50. Qa2> ( 1.09) <50…Qe5+> ( 1.04)

<Engine preference>: <50…Kd7>: ( 1.09)

<51. Kg1> ( 1.04) <51…Qe1+> ( 0.95)

<Engine preference>: <51…Qd5> ( 1.04)

<52. Kh2> ( 1..04) <52…d5> ( 0.95 )

<53. Ra8> ( 0.95) <53…Qb4> ( 0.71)

<Engine preference> <53…Qe5+> ( 0.95)

<He could easily have drawn the game by perpetual check, as White could not afford to play 54. Kg2 and 55. f3 because of 55…Qd4.> - Capablanca

<54. Kg2> ( 0.71) <54…Qc5> ( 2.81)

<First three engine preferences>: <54…c3> ( 0.71); <54…Rb8> and <54…Rb6> (=-0.02)

<The evaluation shift of 3.94 represents a blunder and a weighting of <<2.0>> for an accumulated game weighting of <<6.5>>>.

<A most rare case of chess blindness by Dr Lasker. 54…Rb8 would at least draw.> - Capablanca

<55. Qa6> ( 2.75) <55…Rb8> ( 4.11)

<First three engine preferences>: <55…c3> ( 2.75); <55…d4> ( 3.10); <55…Qb6> ( 3.24)

<It was bad now, but he might have tried 55…Qb6 56. Qc8+ Kd6 57. Ra6 Qa6 58. Qa6+ Kc5 and fight it out with the two passed pawns and his rook against the queen.> - Capablanca

<56. Ra7+> ( 4.11) <56…Kd8> ( 14.46)

<Engine preference>: <56…Qxa7> ( 4.11)

Evaluation values from this point on are transitional, constantly increasing with ply

<57. Rxg7> ( 14.46) <57…Qb6> ( 18.15)

<Engine preferences (20 ply)>: <57…Qe7> ( 14.46); <57…Qc7> ( 15.27)

<58. Qa3>( 18.15 ) <58…Kc8> (#3)

<ANALYSIS>

<Note> The fluctuations generated in the relatively low (16 minimum) ply initial reverse slide were smoothed out as far as possible in the return forward slide. The corrected evaluations extracted from the return forward slide are used in this analysis, as they are considered more reliable than the raw evaluations generated on the initial reverse slide. All moves have been evaluated on forward and return slide for completeness.

<Evaluation range>:

Between <#3> (equal to infinity) applying to <58…Kc8> - representing a forced mate in 3 by White - and < 1.88> in respect of move <46.Rb4> representing a winning advantage for Black.

<The largest evaluation shifts>:

- for White was – excluding the forced winning sequence after Black’s move 54 was 3.52 between <54. Kg2> (( 0.71), a moderate advantage for Black) and <54…Qc5> (( 2.81), a forced win for White.

- for Black was 0.89 from <43…Re5> ( 0.79), a moderate advantage for Black, to <44. Qb4> ( 1.68), a winning advantage for Black.

<Computer statistics>:

• 87.1% of the ply in this game (101/116) coincided with engine preferences 1, 2 or 3

• 77.6% of the ply in the game (90/116) coincided with engine preferences 1 or 2

• 55.2% of the ply in the game (64/116) coincided with the engine’s first preference

• 82.8% of Schlechter’s moves (48/58) coincided with the engine preferences 1, 2 and 3

• 91.4% of Lasker’s moves (53/58) coincided engine preferences 1, 2 and 3

• 75.9% of Schlechter’s moves (44/58) coincided with engine preferences 1 and 2

• 79.3% of Lasker’s moves (46/58) coincided with engine preferences 1 and 2

• 62.1% of Schlechter’s moves (36/58) coincided with the engine’s first preference

• 48.3% (28/58) of Lasker’s moves coincided with the engines first preference.

<Note>: There is no particular significance attached to these statistics except as a separate and general litmus of player accuracy. Clearly a second preference that leads to a forced mate as distinct from a first preference that is defensible or wins is a different creature from a fifth preference that can win, even if there are four more efficient alternatives. A player may make many moves that are not engine preferences and still win, even against a player whose every move bar one is an engine preference, if that exception is a gross blunder. The only way that engine preferences can be statistically relevant is with aggregated data accumulated from many games, however, even these must be secondary to analysis of evaluation shifts between moves.

<The engine evaluation of the final position>:

is <#3>.

<CONCLUSION>

<The game is weighted at <<6.5>>representing <<1 blunder (2.0) plus 1 dubious move (0.5)>> by Schlechter plus <<2 blunders (2 x 2.0 = 4.0)>> by Lasker.>

<Schlechter’s dubious move was <<36.Qb3>> conceding his slight advantage to Lasker. His blunder was <<44. Qb4>> and should have lost the game.

Lasker’s blunders were <<47…c4>> (conceding the win) and <<54…Qc5>> (conceding the game).>

Regarding <The first part sees White, instead of attempting to press for a further advantage to that provided by Black’s conservative opening strategy (Ruy Lopez Berlin Defense. Closed Bernstein Variation (C66)), simplifying quickly into a Q+R+7p ending/middlegame, in which both sides maneuver with equality, barely disturbing the seismometer, until about move 35> From my human chess eye’s point of view, Lasker’s decision to eventually transfer his King to the Queenside probably commencing on 20..Kf8 anticipating 21. Rxg6 hxg6 looks like a sign of his willingness to go for a win. He did not hunker down on the Kingside. He brilliantly conjured a plan of voluntarily marching his King to the opposite side of the board, increasing his winning chances.

So we now have this position <20…Kf8> (=0.21)

click for larger view

In spite of the fact that evals don’t seem to differ by much, with move <23…a6> (=0.05) Black’s pawn structure has changed.

click for larger view

Lasker has made his Queenside pawns immune to direct attacks by the White Queen by transferring them to light-colored squares in a pawn chain based on b7. He accepts the insignificant ‘weakening’ of his d-pawn, which isn’t really weak because it’s very easily well-defended, especially as he seems to have already planned on bringing his King to c7 anyway. Once his King gets on c7, he can now start operating on the center and Kingside freely. Looking at the evals alone, nothing of significance seems to be happening, yet with the changing Black pawn structure that allows for further serious action in the center, something clearly is.

By 29..Kc7, Black has already achieved one of the stop-gap positions he was aiming for, probably all the way when he first played 20..Kf8.

click for larger view

<30. a3> (=0.00) <30…Re7> (=0.00) <31. b4> (=0.01) <31…b5> (=0.00) <32. cxb5> (=0.02) <32…axb5> (=0.03)

So far nothing. Lasker’s position remains as solid as it can be. Further maneuvering did cost Schlechter a dubious move.

<33. g3> (=0.00) <33…g5> (=0.01) <34. Kg2> (=0.00) <34…Re8> ( 0.37) <35. Qd1> (=0.00) <35…f6> ( 0.37) <36. Qb3> ( 0.32)

<First three engine preferences>: <36. a4> ( 0.37); <36. Qd2> (=0.00); <36. Qd3> (=-0.03)

Note that the engine’s first preference <36. a4> ( 0.37) already consists of a pawn advance attempt to break through Black’s Queenside, something that Schlechter did three moves later.

<44. Qb4> ( 1.68)

<First three engine preferences>: <44. Qc3> ( 0.78); <44. Rb2> ( 0.79); <44. Rc1> ( 0.89) White’s move crosses the 1.40 threshold into a lost game for White and is therefore classified as a blunder.

<Weighting for the move is therefore <<2>>, accumulating to <<2.5>> for the game.>

As it turns out it is Schlechter who blunders first.

Yet, to be honest, it’s hard to determine if this move was a blunder without a computer! It’s pretty safe to say that in such open Queen positions, humans will almost always begin blundering sooner or later, and it’s the side with the more exposed King that would stand more chances of losing.

<47. Ra4> ( 1.89) <47…c4> ( 1.22)

<Engine preference>: <47…Kd7>

( 1.89) <<<47…c4>> is a blunder for the purposes of this analysis as it destroys Black’s winning advantage by causing the evaluation to shift across the 1.40 threshold. Accordingly, it is weighted at <<2.0>>, for an accumulated game weighting of <<4.5>>>.

Now it is Lasker who blunders, throwing away the win. The curse of open Queen positions.

<54. Kg2> ( 0.71) <54…Qc5> ( 2.81)

<First three engine preferences>: <54…c3> ( 0.71); <54…Rb8> and <54…Rb6> (=-0.02)

<The evaluation shift of 3.94 represents a blunder and a weighting of <<2.0>> for an accumulated game weighting of <<6.5>>>.

<A most rare case of chess blindness by Dr Lasker. 54…Rb8 would at least draw.> - Capablanca

I guess this means that evals began swinging more widely at around move 35. Although the computer sees it as approximately equal, I believe that most humans would very much prefer Black at this point. This includes Schlechter himself as he clearly began vigorous attempts to open up the game, even at the cost of a pawn sac. Ironically, he starts late. Instead of pushing it on the 36th move when it is not a pawn sac, he pushes it on the 39th, when it is now a pawn sac. He probably noticed his <dubious move 36. Qb3> and tried the a4 push in worse circumstances.

Yet, IMO it was courageous 39. a4 (although three moves late) that saved Schlechter. Perhaps a computer would be able to hold White's position passively, but no human could probably do the same for long. Schlechter in effect sacked a pawn in order to change the entire game structure from a closed to an open one, with his own King in a safer emplacement. Schlechter, being the talented and experienced master that he was knew of course that pawn advantages tend to become insignificant compared to the King's safety in open queen positions; and he played for it.

I don't know, but this could be one game where it was Lasker who got outfoxed in the psychological game. He should immediately have taken the draw, but got so used to playing for a win that he blundered and lost.

Frankly, I have great difficulties in understanding what goes on in open queen positions. I suspect that 44. Qb4 is a blunder because it loses tempo and ground to the opposing queen.

After this apparently Lasker, properly evaluating his position as winning, must have thought that the right plan was to quickly advance his pawns, So his first blunder was a direct result of this strategy. 47..c4. As you said he should have ensured his Kings's safety first. Instead he exposed it more.

These endings are very complicated. You'll notice that Schlechter's strategy for opening up the game was quite successful, not so much because Lasker blundered the game, but because Lasker consistently failed to find the best move during that period, and I think he sensed that the win was somehow falling away. Check out how he finds the second best moves at moves 47, 49, 50, 51, 52 and 53, progressively eroding his advantage, before he finally blunders the game with move 54. During the same sequence, Schlechter's play is impeccable.

Schlechter played quietly to start with and to give him his credit, his defense of his position after he opened up the queen side was exemplary.

He out-psyched Lasker, which was no mean feat!

<Check out how he finds the second best moves at moves 47, 49, 50, 51, 52 and 53, progressively eroding his advantage, before he finally blunders the game with move 54. During the same sequence, Schlechter's play is impeccable.>

---