Commentary - Dortmund 2013 round 6

The Kramnik-Fridman game from round 6 of the Dortmund tournament is an outstanding example of high-level master chess.  Some thoughts on it, in addition to the game annotations:

• Kramnik adopts a strategy in the English vs. Queen's Gambit Declined (QGD) setup that he is very familiar with. Although White's setup at first glance appears passive, in fact it contains significant kingside attacking prospects, once Black's early pressure in the center is dealt with. The fact that overall White has scored close to 60 percent from the inoffensive-looking position on move 8 is, I suspect, evidence that the White players who adopt this strategy simply understand the position to a much deeper level than their opponents.
• Tactical defense is used multiple times by Kramnik in this game to good effect, covering his weaknesses in a dynamic fashion and not tying down his pieces unnecessarily.
• Kramnik' sacrificial attacking idea that begins with 20. f6 is something born from deep positional understanding of the problems Black will face afterwards on the kingside.  It takes Houdini a few moves down its primary path before it is able to see in its evaluation function that White has full compensation for the material.  (Another good example of the pitfalls of computer analysis for the uninitiated.)
• The combination starting on move 29 is the highlight of the game and is worth looking at closely, especially because the initial knight sacrifice appears to come out of nowhere.

[Event "41st GM"] [Site "Dortmund GER"] [Date "2013.08.01"] [Round "6.5"] [White "Kramnik, Vladimir"] [Black "Fridman, Daniel"] [Result "1-0"] [ECO "A13"] [WhiteElo "2784"] [BlackElo "2629"] [Annotator "ChessAdmin/Houdini"] [PlyCount "67"] [EventDate "2013.07.26"] 1. c4 e6 2. Nf3 d5 3. b3 {this keeps the game in English territory, rather than transposing to the QGD.} Nf6 4. g3 {this would allow Black the ...d4 advance if desired. In practice, if Black has intended to play a QGD from the start, he probably will continue with standard moves as in the game.} Be7 5. Bg2 O-O 6. O-O c5 7. Bb2 {by this point the role reversal is clear, Black is going for a strong, classical center while White seeks to undermine it from the flanks and restrain further Black advances.} Nc6 8. e3 {from this innocuous-looking position, White has scored almost 60 percent.} b6 9. Nc3 dxc4 10. bxc4 Qd3 {the point of this move is to disrupt White's piece setup, rather than achieve anything directly for Black.} 11. Ne1 Qd7 12. Qe2 Bb7 13. f4 { White evaluates that his future strategic chances lie on the kingside, so starts the expanion process.} Rfd8 {until this point, we have been following the game Kosten-Hernando Rodrigo (given below) which contains a more normal attacking plan for White.} (13... Na5 14. d3 Bxg2 15. Qxg2 Rfd8 16. g4 Ne8 17. Rd1 Nd6 18. Nf3 Qb7 19. e4 b5 20. e5 Ndxc4 21. dxc4 Nxc4 22. Bc1 b4 23. Qe2 Rxd1 24. Nxd1 Nb6 25. Nf2 a5 26. Ne4 Qa6 27. Qg2 Qc4 28. Re1 a4 29. f5 Nd5 30. Nd6 Qd3 31. fxe6 fxe6 32. Qe2 Qxe2 33. Rxe2 c4 34. Nd4 b3 35. axb3 Bxd6 36. exd6 axb3 37. Nxe6 Nc3 38. Re1 Na2 39. Bb2 Nb4 40. d7 Nc6 41. Ng5 {1-0 (41) Kosten,A (2516)-Hernando Rodrigo,J (2381) Castelldefels 2004}) 14. Rd1 Nb4 15. d3 {the pawn will actually be stronger here, due to a tactical defense, and cover e4.} Bxg2 16. Nxg2 a6 (16... Nxd3 $2 17. Ne1 {and with the knight pinned and no new defenders coming to its rescue, the best Black can do is capture on f4 and get two pawns for the piece.}) 17. a3 Nc6 18. g4 {White's kingside attack now gets started.} Ne8 19. f5 Nd6 20. f6 {the start of a deep and unusual sacrificial idea by Kramnik. One wonders what he would have done in the event of 19...exf5.} Bxf6 21. Rxf6 {otherwise the previous move makes no sense.} gxf6 22. Nf4 Ne5 23. Nh5 Qe7 24. Rf1 {here Houdini's evaluation is that White has full compensation for the material.} Nd7 25. Qg2 (25. Qf2 { would be the more obvious follow-up.}) 25... h6 26. h4 Kf8 {this appears to be where Black starts going astray with his defense. Houdini recommends instead looking to start counterplay on the queenside.} (26... b5) 27. Qg3 Ke8 28. Qf4 Qf8 $2 {this appears to be a normal response to the threat against h6, but allows the following combination.} 29. Nd5 $1 {a very non-obvious move. White's pieces do not appear at first glance to be sufficiently coordinated for a combination and there are plenty of defenders. However, the Black king's very limited space and the weaknesses of f6 and g7 are the elements from which the combination is constructed.} exd5 30. Bxf6 {the immediate threat is Ng7+} Ne5 {to give the king an escape square on d7} (30... Nxf6 31. Nxf6+ Ke7 32. Qe5#) 31. Qxe5+ Kd7 32. Bg7 {the combination now morphs from mate threat to trapping Black's queen.} Re8 33. Nf6+ Kc7 34. Qa1 {the Bg7 is protected tactically by the knight check discovery White would have after ... Qxg7.} 1-0

Annotated Game #100: The Fun Milner-Barry Gambit

Inspired by a recent Brooklyn64 post on a swashbuckling Milner-Barry Gambit in the French Defense, here is my one and only tournament game featuring it.  This was played early in my scholastic career when I was still an e4 player and had a low Class C rating.

I had relatively little idea at the time how to most effectively conduct attacking play, but the gambit nevertheless showed promise against my significantly higher-rated opponent, a veteran of the local tournament scene.  For White players who hate facing the French, it's an fun way to take it on while avoiding typical position types.

[Event "?"] [Site "?"] [Date "????.??.??"] [Round "?"] [White "ChessAdmin"] [Black "Class B"] [Result "0-1"] [ECO "C02"] [Annotator "ChessAdmin/Fritz/Houdini"] [PlyCount "84"] [EventDate "1985.??.??"] {C02: French: Advance Variation} 1. e4 e6 2. d4 d5 3. e5 c5 4. c3 Nc6 5. Nf3 Qb6 6. Bd3 {the Milner-Barry Gambit. White is willing to give up the d-pawn (and sometimes the e-pawn too) for development and attacking prospects.} Bd7 7. O-O cxd4 8. cxd4 Nxd4 9. Nc3 (9. Nbd2 {is favored by the best White players of the line, allowing the knight to replace its brother on f3 after it is exchanged off. On c3, the knight can't assist with a kingside attack, although it does influence the key b5 square.}) 9... Ne7 {this position can also be reached from the c3 Sicilian, believe it or not.} 10. Be3 (10. Nxd4 Qxd4 11. Nb5 Qxe5 12. Re1 Qb8 13. Qf3 {would be the way to best exploit the Nc3 setup. White is now two pawns down, but has excellent attacking prospects and scores over 68 percent from this position.}) 10... Nxf3+ 11. Qxf3 Qc7 {a passive move. Houdini considers this position equal, with White having full compensation for the pawn.} (11... d4 12. Bxd4 Qxd4 13. Qxb7 $15) 12. Nb5 {a typical amateur move, starting operations before developing the rooks, although it is not a particularly bad error in this position. White hopes Black will take the e5 pawn, which would lead to disaster.} (12. Rac1 Qa5 $11 (12... Qxe5 $2 {is worthless, says Fritz, because of} 13. Nb5 Nc6 14. Bf4 $18)) 12... Bxb5 (12... Qxe5 $2 13. Bf4 Qf6 14. Nd6+ Kd8 15. Nxb7+ $18) 13. Bxb5+ Nc6 14. Rac1 a6 15. Bxc6+ {otherwise the bishop is going to be largely shut out of the game by all of Black's pawns on light squares.} bxc6 16. Rc2 (16. Qg3 {is Houdini's subtle choice. Black will have trouble getting his kingside in order now.}) 16... Rc8 (16... Qxe5 $5 17. Rxc6 Qxb2 $17 {was Fritz's brilliant deduction (/sarcasm) at the time, which in fact fails to Rc7 with a big plus for White. It was an early version of Fritz and was set to 20 seconds/move for the full game analysis, but it's still a good example of why not to blindly trust engine output.} (17... Be7 {and Black is fine, however.})) 17. Qe2 $11 {this targets the a6 pawn, but Qg3 is still a good idea here.} Qb7 18. f4 {White probably should followed up on the original plan of increasing pressure on the c-file. Instead, I push the f-pawn, because that was my idea of how to attack in this position. In practical terms it's not a bad idea, however, since it puts additional psychological pressure on my opponent.} (18. Rfc1 a5 $11) 18... Be7 $15 {Black is now getting his kingside developed, reducing White's compensation for the pawn.} 19. f5 c5 $2 {Black has sufficient defensive resources to parry White's attack, but instead chooses to completely ignore it. Not a good decision, because the opponent is right back in the game, says Fritz.} (19... exf5 $5 {is more easily played by an engine, who doesn't have a visual bias with its calculations.} 20. Rxf5 O-O $15 {and black will be able to help defend with his queen, for example ...Qb4 or ...Qd7.}) 20. fxe6 $14 { another typical amateur move. Rather than build up the tension and bring additional forces into the attack, the tension is prematurely lifted.} (20. Qg4 Rf8 21. Qxg7 exf5 22. Bh6 $18) 20... fxe6 {Black however plays an inaccurate defense, giving White the opportunity to regain his full advantage.} (20... O-O 21. exf7+ Rxf7 22. Rxf7 Kxf7 23. Qg4 $14) 21. Qg4 $16 Kd7 22. Bxc5 $4 {the losing move. I tried to get fancy with the tactics and failed to recognize that Black's 23rd move would come with check. Otherwise, White on move 24 could take on g7 with check and pick up the h8 rook.} (22. Rf7 Rhe8 23. Bg5 $16 ) 22... Rxc5 $19 23. Rxc5 Bxc5+ {d'oh!} 24. Kh1 g6 25. Rf7+ Be7 26. Qh4 Re8 { by this point White is a piece down and has no threats left that Black cannot easily parry.} 27. b4 Qc7 28. h3 Qxe5 29. Qf2 (29. Qxh7 {is no salvation, notes Fritz.} Kd8 30. Qxg6 Qa1+ 31. Kh2 Bd6+ 32. g3 Qxa2+ 33. Kh1 Qa1+ 34. Kg2 Qb2+ 35. Rf2 Qxb4 $19) 29... Rc8 30. Qa7+ Rc7 31. Qxa6 Rc1+ 32. Rf1 Rxf1+ 33. Qxf1 {White's lone queen can fight a guerrilla action, but in the end it's hopeless.} Bd6 34. Qf7+ Kc6 35. b5+ Kb6 36. Qf2+ Kxb5 37. a4+ Kxa4 38. Qc2+ Kb5 39. Qb3+ Kc6 40. Qa4+ Kc7 41. Qa7+ Kd8 42. Qa8+ Ke7 0-1

Annotated Game #99: Why it's bad to be positionally dominated

This final round tournament game saw Black achieve complete positional dominance by the early middlegame.  At the time, I thought I had played reasonably well against my higher-rated opponent until that point, but more objective analysis shows how I simply did not understand what was required in the position after playing some standard opening moves. I also had not learned much at all from my earlier round loss in the English where I mishandled the central pawn structure and failed to play e4 at a critical moment (see the recent Amateur Hour post for more on this).

Overall, it was bad to be White and good to be Black in this tournament (for me at least), since Black was the victor in all five of my games; I had Black twice.  Perhaps a contributing factor to this performance was a certain overconfidence in my experience with the English Opening and an unconscious assumption that the rest of the game should easily take care of itself.  As I was paired up by a significant margin each time I had White, this did not work out so well.

In this game, Black's central threats could have been nullified by White on move 14 with the prophylactic e4 push, but White fails to comprehend his weaknesses (including the hanging Nc3) and ends up with his pieces offside and ineffective by move 18.  Black's subsequent careful, relentless crushing of White is instructive and shows how this kind of positional dominance can nullify any hope of an opponent saving themselves with tactics, as none exist.

This was the last tournament game I played before starting this blog and getting serious about chess improvement.  Although for the next annotated game I plan to look at a specific game from earlier in my chess career, after that we will get to see more contemporary games and what lessons they may hold for my ongoing development as a player.

[Event "?"] [Site "?"] [Date "????.??.??"] [Round "?"] [White "ChessAdmin"] [Black "Class A"] [Result "0-1"] [ECO "A24"] [Annotator "ChessAdmin/Fritz/Houdini"] [PlyCount "90"] {A24: English Opening vs King's Indian: Lines without ...Nc6} 1. c4 g6 2. Nf3 Bg7 3. Nc3 Nf6 4. g3 O-O 5. Bg2 d6 6. O-O c6 {a key decision in the KID lines. The c6 pawn blunts White's bishop on the long diagonal, but also can become a target of White's b4-b5 push.} 7. d3 e5 8. Bg5 {the dark-square bishop can be difficult to develop in these lines. White decides to quickly trade it off.} ( 8. Rb1 {is typically played first, for example} a5 9. a3 Re8 10. Bg5) 8... h6 9. Bxf6 Bxf6 {Black has the pair of bishops, notes Fritz.} 10. Rb1 (10. b4 $5 { I rightly saw would lose the exchange, but Houdini thinks White has compensation for it.} e4 11. Nxe4 Bxa1 12. Qxa1 {White is the equivalent of a pawn down materially, but the beautiful a1-h8 diagonal and his development lead are worth it.}) 10... Be6 {joining the fight for d5.} 11. b4 d5 {at the time I felt this was a little premature, but Black does well to assert his strength in the center.} 12. Nd2 Qd7 (12... e4 $5 {would further pressure White and is the logical follow-up to the central push by Black. The move works because of the hanging Nc3. For example} 13. cxd5 cxd5 14. Na4 exd3) 13. cxd5 cxd5 14. a4 $2 {for the second time in the tournament, I fail to consider the best move on the board because it involves advancing the e-pawn.} (14. e4 { would stop Black's central threats and reassert White's control over d5. The blocking of the Bg2 is less important.} d4 15. Nd5 $14) (14. b5 $11 {is what my opponent thought would be best here, preventing the knight from coming out. It's certainly better than the text move.}) 14... Nc6 (14... Rc8 {looks a little unnatural, but would immediately attack White's weak points and give Black positional dominance, as both Fritz and Houdini show.} 15. Rc1 Nc6 16. b5 Nb4 $17) 15. Qc1 $2 {this is just asking for trouble. Again e4 would be best here, although not as good as on the previous move.} Bg7 (15... Rac8 {is the obvious way to exploit the Queen's vulnerability on the c-file.} 16. e4 Nd4 $17 ) 16. Nb3 b6 $17 (16... Nxb4 $2 {pawn snatching would backfire here on Black.} 17. Nc5 Qe7 18. Rxb4 Qxc5 19. Nxd5 Rac8 20. Qxc5 Rxc5 21. Ne7+ Kh7 22. Rxb7 $14 ) 17. Qa3 Rfc8 {Black's dominant positional advantage is apparent, with White's pieces contorting themselves to try and cover his weak points, while Black is admirably centralized and flexible, with multiple threats and the initiative.} 18. Rfc1 $6 (18. b5 $5 Ne7 19. Rbc1 $17) 18... Bf8 $19 {Black now puts the screws to White in a simple but effective manner.} 19. Na2 $2 { this will bury White's knight and effectively take the piece out of the game.} (19. Nd2 $19 {it would be better simply to abandon the pawn, although White has poor prospects in any case.}) 19... a5 20. Qb2 axb4 21. Rc2 Bd6 22. Rbc1 Na5 (22... Rxa4 {would be simpler.}) 23. Nxa5 bxa5 24. d4 {White desperately tries for some sort of counterplay.} e4 (24... Qxa4 {again taking the pawn would be the simple and most effective way to proceed. White would have no hope of stopping the two connected passed pawns.}) 25. Rxc8+ Rxc8 26. Rxc8+ Qxc8 {the material exchanges still don't provide White any relief, largely due to the out-of-play Na2. The engines evaluate black as being over the equivalent of a piece up.} 27. Qd2 Kg7 28. f3 f5 29. fxe4 fxe4 30. e3 Bd7 31. Qd1 Qc4 32. Nc1 Qc6 {Black finally gets around to picking up the a-pawn.} 33. Bf1 Qxa4 {from here the game is completely hopeless for White.} 34. Qe2 Qc6 35. Qd2 a4 36. Na2 Qb7 37. Qb2 b3 38. Nc1 Qb4 39. Kf2 Qa3 40. Qb1 Qb4 41. Be2 Qc3 42. h4 Ba3 43. Na2 Qb2 44. Qxb2 Bxb2 45. Nb4 a3 0-1

The importance of CCT: example #6 - Dortmund 2013 round 1

As part of the continuing series on the importance of CCT (Checks, Captures and Threats) to your thinking process, here is the round 1 game from Dortmund between Vladimir Kramnik and Wang Hao.  Visually it's easy to understand why Black assumed that his last move 25...Ra2 was safe, as it cannot be safely attacked by White's rook and White's queen is on the same file with a White pawn in between them.  With that assumption, Black evidently did not bother with falsifying his move and calculating his opponent's CCT.  However, following the queen check on b8, which forces 26...Kh7, White has the queen fork on b1 and it's all over.

Mastery Concept: Clearing a square for a piece

This Mastery Concept post, as others in the series will do, highlights an idea that was new (to me) as part of my chess studies since this blog was created.  Exposure to these types of "mastery" ideas - ones that are more sophisticated than simply "I go here, my opponent goes there" -  I believe are essential to the improvement process.  Humans are not particularly good brute-force calculators, so starting your thinking process with calculation is unlikely to provide you with the tools needed to find the strongest candidate moves.  At the master level, the process is in fact usually reversed: an idea is first recognized on the board, then calculation follows to determine if the associated candidate moves work.  This is the primary benefit of pattern recognition.

One repeated observation of amateur play is that amateurs fall into the trap of assuming that each move must be significant in itself - make a threat, defend a weakness, etc.  Master play is significantly deeper in recognizing that a move may be used in order to prepare a much greater threat, set up a combination, and so on.

One of the more subtler concepts that I've run across, and it is perhaps more valuable to recognize because of its subtlety, is clearing squares for pieces.  The piece that is moving is therefore not the significant actor in the chess drama, it is the piece that will replace it on the square.  This may be an obvious concept to many, but for those of us who too easily overlook ideas or valuable candidate moves, the insight can help our thinking and give us an extra edge.  It can also be a combinative idea or a positional one, so it has broad applications.

Chess being a practical game, here are two very clear illustrations of how this mastery concept can work in practice.  Chesstempo also has an illustration of a piece clearance sacrifice as part of its reference on tactical motifs.

1.  Ernesto Real de Azua - Vinay Bhat (2000)
This game was highlighted on GM Bhat's There and Back Again blog and was included in Daniel Naroditsky's Mastering Positional Chess (New in Chess, 2010).  The key move 19...b5! dramatically paves the way for ..Nb6 and gets Black out of a serious jam.  The link above has more of GM Bhat's analysis and commentary.

[Event "Wch U16"] [Site "Oropesa del Mar"] [Date "2000.10.14"] [Round "4"] [White "Real de Azua, Ernesto"] [Black "Bhat, Vinay S"] [Result "1/2-1/2"] [ECO "C00"] [WhiteElo "2173"] [BlackElo "2396"] [PlyCount "80"] [EventDate "2000.10.11"] [EventType "swiss"] [EventRounds "11"] [EventCountry "ESP"] [Source "ChessBase"] [SourceDate "2000.11.22"] 1. e4 e6 2. d3 d5 3. Nd2 Nf6 4. Ngf3 b6 5. g3 dxe4 6. dxe4 Bb7 7. Qe2 Ba6 8. c4 Nc6 9. e5 Nd7 10. Bg2 Nc5 11. O-O Qd7 12. a3 Bb7 13. b4 Qd3 14. Qd1 Nd7 15. Bb2 O-O-O 16. Re1 h6 17. Re3 Qh7 18. Qa4 Be7 19. Nb3 b5 20. cxb5 Nb6 21. bxc6 Nxa4 22. cxb7%2B Kb8 23. Bd4 Rd5 24. Nfd2 Rhd8 25. Be4 Qg8 26. Bxd5 Rxd5 27. Na5 Qe8 28. Ndb3 c5 29. bxc5 Bd8 30. Nc4 Bc7 31. Nd6 Bxd6 32. exd6 Qb5 33. Bxg7 Nxc5 34. Nxc5 Qxc5 35. Bf8 Qc2 36. Bxh6 Rd1%2B 37. Rxd1 Qxd1%2B 38. Kg2 Qd5%2B 39. Kg1 Kxb7 40. Bf8 Qd1%2B 1/2-1/2" />

2.  Alexander Chernin - Anthony Miles (1985)
This is a classic modern game featuring a pawn sacrifice in what Hans Kmoch in Pawn Power in Chess termed a "sweeper-sealer-twist" - the pawn is moved (sacrificed in this case) in order to replace the pawn with a piece on the square.  Here 12. e5! is the move that the rest of the game revolves around, as the e4 square is pivotal for White.

[Event "Tunisien"] [Site "Tunisien"] [Date "1985.??.??"] [EventDate "?"] [Round "8"] [Result "1-0"] [White "Alexander Chernin"] [Black "Anthony Miles"] [ECO "E10"] [WhiteElo "?"] [BlackElo "?"] [PlyCount "63"] 1.d4 Nf6 2.Nf3 e6 3.c4 c5 4.d5 b5 5.Bg5 h6 6.Bxf6 Qxf6 7.Qc2 exd5 8.cxd5 d6 9.e4 a6 10.a4 b4 11.Nbd2 Bg4 12.e5 dxe5 13.Ne4 Qf4 14.Nfd2 Bf5 15.Bd3 Bxe4 16.Nxe4 Nd7 17.g3 Qg4 18.h3 Qh5 19.d6 Qg6 20.Rd1 b3 21.Qe2 f5 22.g4 c4 23.Bb1 fxg4 24.Qxc4 Qf7 25.Qc6 Rd8 26.hxg4 g6 27.Nc5 Rg8 28.Be4 Bg7 29.Nxa6 Qf4 30.Nc7%2B Kf7 31.Qc4%2B Kf6 32.Nd5%2B 1-0" />