Annotated Game #193: How strategic errors can snowball

This final-round tournament game illustrates a curious phenomenon, namely how small strategic errors you make tend to "snowball" - gathering weight as time goes on, until your game is crushed by them.  I believe this is because strategic errors reflect an incorrect mindset on the part of the player, and/or a fatally flawed understanding of the position.  Relatively small errors early on are clues that you are thinking wrongly about the position.  Later this often leads to incorrect plans and, consistent with that, more significant errors.

Below, we can see that a rather pleasant position for White around move 7 starts losing momentum after move 10.  I have the (flawed) idea of setting up a Hedgehog-type position, but succeed instead in playing too passively and not seizing enough space to properly develop and maneuver my pieces.  My opponent does a good job of playing natural moves that improve his position on the queenside, until I (too early) become desperate for a solution, which rapidly sends my game downhill.  Moral of the story: seek to understand the position's requirements, especially for your pieces' activity, rather than trying to impose an arbitrary strategy onto the board.

[Event "?"] [Site "?"] [Date "????.??.??"] [Round "?"] [White "ChessAdmin"] [Black "Class C"] [Result "0-1"] [ECO "A16"] [Annotator "ChessAdmin/Komodo 11.2"] [PlyCount "62"] {[%mdl 8192] A16: English Opening: 1...Nf6 with ...d5 A16: English Opening: 1.. .Nf6 with ...d5} 1. c4 e6 2. Nc3 Nf6 3. Nf3 Nc6 {with this move, Black is essentially playing an English Four Knights variation with his e-pawn on e6 rather than e5. This is solid but unambitious.} 4. g3 d5 5. cxd5 {I saw no benefit to allowing d5-d4, and the exchange gives White a target on the long diagonal.} Nxd5 6. Bg2 Nxc3 $146 7. bxc3 $14 (7. dxc3 {going directly into the endgame is level, but unnecessary for White, who has a nice positional plus based on the Bg2.}) 7... Bd6 8. O-O O-O 9. Rb1 {immediately positioning the rook on the half-open file and pressuring b7.} Rb8 10. d3 {although this isn't a bad move, I would say that I strategically start to lose momentum here. I shouldn't be afraid to play d4, with more central control over c5 and e5. Controlling the c4 and e4 squares doesn't do as much for me. I was thinking (a little too vaguely) about setting up a Hedgehog structure for White, but what results isn't that.} (10. d4 e5 {if Black challenges in the center, then exchanging off would be to White's advantage, so I could just improve my pieces further. For example} 11. Qd3 h6 {otherwise Ng5 is a strong follow-up move} 12. Rd1 $14) 10... b6 11. Be3 $11 {now I have problems with cramped space and developing my pieces to useful squares.} (11. Bb2 $5 {would be a little more active.}) 11... Bb7 12. Qc2 h6 {Secures g5} 13. Rb2 {by now it's clear I lack a viable strategy. Going for doubled rooks on the b-file is next to useless.} Qd7 14. Rfb1 Ne7 15. a4 {my plan, such as it is, is to try and crack open the queenside with my a-pawn. However, Black can avoid any problems by simply ignoring the pawn, so it's a rather bad plan.} c5 {this actually gives the a-pawn advance a bit more bite. Now Black cannot play b6-b5 in response, for example.} (15... e5 16. a5 f5 $11) 16. a5 Nd5 17. Bd2 $6 { continuing to make even more passive moves. I was worried about ...Nxe3, but failed to calculate that Black could not ignore the pawn capture on b6 first.} (17. axb6 axb6 {now the doubled rooks have a purpose in life on the b-file} 18. Bd2 Bc7 $11) 17... bxa5 $11 {the correct decision. The a-pawns of course are weak, but Black is temporarily a pawn ahead and having a passed a-pawn is compensation.} 18. e4 {the engine points out that it's better to go after the a-file immediately, but at least I'm thinking a bit more actively now. However, not calculating consequences and alternatives results in a weaker game.} (18. Ra1 Bc7 19. Qc1 {avoiding a potential future attack from a knight landing on b4.} Rfe8 20. c4 $11) 18... Nb6 19. Rb5 a4 $15 {with Black's extra a-pawn now protected, he has a small edge, although the battle still revolves around his weak queenside pawn structure.} 20. Be3 {this delays the rook move and allows Black to start making threats.} (20. Ra5 $5) 20... Bc6 21. Bxc5 $4 {this was just desperation, also based on flawed calculation.} (21. Ra5 $15 {this is the best bet to save the position, notes Komodo via the Fritz interface.}) 21... Bxb5 $19 22. Bxd6 Qxd6 23. Rxb5 {now I'm just a full exchange down with no compensation, plus Black's a-pawns have fewer defenders in front of them.} Rfc8 24. Rh5 {now I really start getting desperate and hope to get something going on the kingside with my rook.} Qa3 {my opponent correctly ignores it.} 25. c4 Qb3 26. Qd2 a3 27. Bf1 {note how useless this once-promising bishop turned out to be.} Qb2 (27... Nxc4 {is a cute tactic:} 28. dxc4 Qxf3 $19) 28. Qf4 a2 29. Ne5 {one last gasp before resigning.} Rb7 30. Rxh6 gxh6 31. Qxh6 Qxe5 0-1

Annotated Game #192: The problem of mental perspective

When you retrospectively analyze your games, it's amazing how certain fundamental issues with your play can seem to spring forth and fairly shout their existence, as you see them repeated over a series of games.  At the time, you are often unaware of these holes or problems in your game, or at least the fact that they form a consistent pattern.  Perhaps it is largely a problem of mental perspective; to take an example from the physical realm, in a similar way you can have an obvious blemish on your face, but can't see it yourself until you hold up a mirror and look at it from an outside perspective.

Identifying these consistent flaws in my game has led to some significant improvements over time, ranging from major changes (coming up with a standard thought process) to relatively minor but still measurable advances (Annotated Game #63: Third time's the charm).  In the most recent analysis series, what has jumped out at me is the theme of unnecessarily complicated moves, which this game shares along with Annotated Game #191 and Annotated Game #189.

In the case of this game, which was a nail-biting tactics fest in a weird Caro-Kann sideline, I could have consolidated my advantage on move 16 or again on move 19, but find a more complicated way to win; at least I can't complain about the result.  Looking back on it now, I was definitely under more psychological pressure than was warranted and a calmer approach would have been more effective.  Something to remember for future such games that feature early direct pressure on the king.

[Event "?"] [Site "?"] [Date "????.??.??"] [Round "?"] [White "Class B"] [Black "ChessAdmin"] [Result "0-1"] [ECO "B11"] [Annotator "ChessAdmin/Komodo 11.2"] [PlyCount "44"] {[%mdl 8256] B11: Caro-Kann: Two Knights Variation} 1. e4 c6 2. Nc3 d5 3. Qf3 { a relatively rare variation, developing the queen early instead of the knight, which would lead to the Two Knights variation of the Caro-Kann.} d4 {here I gave some thought to capturing on e4, but decided that my opponent would likely be familiar with that, and I didn't like the fact that he would have two pieces developed to my none. Instead, I seize some more space and inconvenience the Nc3. The drawback is that it is another pawn move and doesn't do anything directly to aid my development.} 4. Bc4 {something I hadn't considered, but is the standard move in the database. White is playing in classic attacking mode, hitting the weak f7 square early.} e6 5. Nce2 b5 { another pawn move, but it is with tempo, as it hits the bishop.} 6. Bb3 a5 { continuing the straightforward theme of trying to push White's pieces around with my pawns.} (6... Bb7 7. d3 c5 8. Qg3 Ne7 9. Nf3 Ng6 10. h4 Nd7 11. h5 Ne7 12. a3 Nb6 13. Ba2 Nc6 14. Bf4 Rc8 15. Rd1 Qd7 16. Ng5 Be7 17. Nxe6 fxe6 18. Qxg7 Rf8 19. Qxh7 Bf6 20. Qg6+ Ke7 21. h6 {Berescu,A (2386)-Streikus,S (2267) Patras 1999 1-0 (40)}) 7. a3 $146 {giving the bishop an out on a2.} Bb7 { I felt it was time to start developing my pieces. This also helps address potential tactical threats by White on the long diagonal.} 8. d3 c5 {although it's another pawn move, it's valuable to reinforce d4 and also open up the Bb7's scope.} 9. Nh3 {the only current square for the knight.} Be7 {with the idea of fighting for the g5 square.} (9... a4 {would have been good to throw in at this point, more to stop White from himself playing a3-a4 and threatening to open up the a4-e8 diagonal for his bishop.}) 10. Nef4 {my opponent is now fully focused on the kingside, hoping to break through with a direct attack.} (10. a4 bxa4 11. Bxa4+ Bc6 $14) 10... Ra6 {a creatively awkward way to protect the e6 pawn (White's target) along the 6th rank.} (10... Qb6 {is a much more natural version of the same idea, with the queen being well-placed on b6.}) 11. Qg4 {this looks threatening, but the g-pawn is poisoned if the queen takes it.} (11. Nh5 $5 {is the way to attack it.} Kf8 12. N3f4 Nf6 $14) 11... Nf6 $15 {my opening problems are now solved, as White has nothing better than to retreat the queen. However, my opponent goes 'all in'.} 12. Qxg7 $2 {Black cannot castle king side} Rg8 $19 13. Qh6 Bf8 {trapping the queen, so White has to give up material to free her.} 14. Nxe6 Rxe6 15. Qd2 Nxe4 $1 {Komodo/Fritz awarded this an exclamation point. I give back some material to open the center and provide full scope to my rooks.} 16. dxe4 Rxe4+ (16... Bh6 {as an "in-between move" would have been more powerful here, shutting down the h6-c1 diagonal. I end up playing the move later, but to less effect.}) 17. Kd1 c4 {with the idea of cutting the bishop off from the f7 target.} 18. Ba2 Rxg2 19. Re1 {despite winning by a wide margin, I was still (overly) fearful about White counterplay and felt I was hanging on by a thread. Calm play would have made things easier from here, although I still find moves that are good enough.} Bh6 {an interesting sacrificial idea that wins, but is unnecessarily complicated.} (19... Qe7 {is simple and very good.}) 20. Rxe4+ ( 20. Qxh6 {accepting the sacrifice is the main line:} Rxe1+ 21. Kxe1 Qe7+ 22. Kf1 Rxh2 $19) 20... Bxe4 21. Qxh6 {now this leads to more immediate consequences for White.} Bf3+ 22. Kd2 Qe7 {simple, quiet, powerful and best.} ( 22... Qe7 23. Bxc4 bxc4 24. Qe6 Qxe6 25. b3 Qe2#) 0-1

Training quote of the day #13: Peter Heine Nielsen

At any rate, he [Carlsen] was focused and wanted to play his best chess without thinking about winning or losing. If you start to think about results you start to get nervous. When Magnus plays freely and does not care too much you see his best side.

-- GM Peter Heine Nielsen, "Inside Team Carlsen" interview 

Annotated Game #191: Having to be brilliant to win a won game

This tournament game is notable for my failure to win a won game in a more normal and less dramatic fashion.  After spotting a tactical sequence that puts me a piece up, I make all the right moves, but my opponent refuses to give in.  Sometimes this is annoying (when you're the one winning), but is also not a bad strategy at the club level, if there's no immediate knockout, as in this game.

Later on, I break the rule of not allowing my opponent counterplay (or even just the appearance of it), in part by making an unnecessarily complicated move (as in Annotated Game #189), and as a consequence he almost forces a drawing tactic.  However, I earn a "!!" from Fritz when I find (in some desperation) the only winning move, a rook sacrifice.  It's also worth highlighting the final decision to exchange queen for rook and force a won K+P endgame.  This should be a rather basic choice for a player, but it's also a sign that you have confidence in your calculating ability and knowledge of endgames.

Furthermore, as a general rule, I think it's good for a player to trust themselves when they are sure they have found a forced win, and not triple-check things.  This is a characteristic you often see in master games and can be misunderstood when engines give significantly higher valuations to other (also winning) continuations.  Even if it takes longer, the chosen route may well be easier, which is probably why the player was able to calculate it first over other, more complicated alternatives.

[Event "?"] [Site "?"] [Date "????.??.??"] [Round "?"] [White "Class B"] [Black "ChessAdmin"] [Result "0-1"] [ECO "A12"] [Annotator "ChessAdmin/Komodo 11.2"] [PlyCount "110"] [EventType "simul"] {[%mdl 8256] A12: English Opening: 1...c6 with b3 by White} 1. c4 c6 2. b3 Nf6 3. Bb2 {it's unusual to develop the bishop and the queenside so quickly, but my opponent appeared to be experienced in using this "system" type line, given his quick play in the opening.} d5 4. Nf3 Bf5 {the most common reply, getting Black into a Slav-type structure, which is what I had been aiming for.} 5. g3 { my opponent chooses a double fianchetto system, which was not a surprise, given his early fianchetto of the dark-square bishop.} e6 6. Bg2 Nbd7 7. O-O Be7 {here the database is evenly split between this and ...Bd6, although the .. .h6 idea (see next move) is also popular (and scores comparatively well).} 8. d3 O-O (8... h6 {this prophylactic move, controlling g5 and providing a retreat square for the Bf5, scores much better in the database. For example} 9. Nbd2 O-O 10. Qc2 Bh7 11. Ne5 Nxe5 12. Bxe5 Bd6 13. Qb2 a5 14. Nf3 Bxe5 15. Nxe5 Qd6 16. a3 d4 17. f4 Nd7 18. Nxd7 Qxd7 19. b4 Rfe8 20. b5 c5 21. b6 Re7 22. Rab1 e5 23. Qb5 Qc8 24. fxe5 Rxe5 25. Bd5 Bf5 26. Qb2 Re7 27. Qd2 Be6 28. e4 dxe3 29. Qxe3 Qd7 30. Rfe1 Rc8 31. Bg2 Ree8 32. Qf3 Rc6 33. Qf2 Rd6 34. Qxc5 Rxd3 35. Qc7 Qxc7 36. bxc7 Rc8 37. Rxb7 Rd7 38. c5 Rdxc7 39. Rxc7 Rxc7 40. c6 Kf8 41. Re5 a4 42. Ra5 {1/2-1/2 (42) Svidler,P (2744)-Caruana,F (2767) Sochi 2012}) 9. Nbd2 (9. Nh4 $5 {would now harass the Bf5, but this is not a serious consideration for Black. For example} Bg4 10. h3 Bh5 $11) 9... Re8 10. a3 a5 { played to restrain the b3-b4 advance.} 11. Bc3 {this move to me indicated that my opponent wanted to try to force the b-pawn advance. However, this takes time and also moves the bishop to an undefended square, which soon becomes important.} c5 $146 {a logical response, reinforcing my own control of b4 and now also influencing d4, which my opponent immediately challenges.} 12. d4 { it's important to remember that every pawn advance leaves weaknesses behind it, in this case the e4 square.} (12. Nh4 Bg4 13. h3 Bh5 14. g4 Bg6 15. Nxg6 hxg6 $11) 12... Ne4 {this immediately exploits the lack of pawn control of e4, but more patience perhaps was in order.} (12... Qb6 $5 {would put additional pressure on both flank and center and connect the rooks.}) 13. Nxe4 Bxe4 14. cxd5 Bxd5 {the position is equal here, but I feel Black has a much easier task and has energy built up, waiting to be released after an exchange of the c-pawn.} 15. dxc5 $6 {this just plays to Black's strengths and immediately helps activate my pieces.} (15. Rc1) 15... Nxc5 $15 16. b4 {White seemed to be fixated on getting this advance in, and seemed to be happy to have achieved it, despite the significant problems he now has in the center.} Ne4 {now the other knight occupies e4 and threatens the hanging Bc3, but White can no longer get rid of it.} 17. Qd3 $2 {defending the Bc3, but missing the follow-on skewer tactic on the long diagonal.} (17. Bd4 $15 {is just about the only chance, notes Komodo via the Fritz interface.}) 17... Nxc3 $19 18. Qxc3 Bf6 {my opponent thought for some time here, eventually coming up with the move with perhaps the most practical chances of avoiding pain, but I am able to find the correct tactical sequence.} 19. Ne5 {temporarily blocking the tactic, but now Black has} Bxg2 20. Kxg2 Qd5+ {with a double attack on e5.} 21. Qf3 Bxe5 { I was happy to get the queens off here, given the material balance.} 22. Qxd5 exd5 23. Rad1 {my opponent chooses to play on, even down a full piece with no compensation. This is common at the club level, and almost pays off for him later on.} axb4 {tidying up on the queenside first.} 24. axb4 Bc3 (24... d4 { is objectively superior. There is no particular reason to give up the strong advanced central d-pawn, even if it is isolated, as it is easily defended. However, I felt it would be easier to make progress by in effect exchanging it for the e-pawn.}) 25. Rxd5 Rxe2 26. Kf3 Rb2 27. b5 g6 {giving my king "luft" (room to escape a potential threat of back rank mate).} 28. Rfd1 Raa2 {so far my game is effectively playing itself, with obvious moves targeting White's weaknesses.} 29. Rf1 Ba5 {covering the d8 square and preparing to move to the a7-g1 diagonal to further pressure f2.} 30. h4 Bb6 31. h5 f5 $6 {here I get too fancy and make an unnecessarily complicated move in response, significantly loosening my king position.} (31... Bxf2 {is simplest, as White has no real threats.} 32. Rfd1 Ra3+ 33. R5d3 Rxd3+ 34. Rxd3 Bc5 $19) 32. g4 ( 32. Kf4 $5) 32... Rxf2+ 33. Rxf2 Rxf2+ 34. Kg3 f4+ 35. Kh4 {up to this point I've played accurately and increased my advantage. With further accurate play there would be no danger, but I've left my king somewhat exposed and my opponent can now generate counterplay after my inaccuracies.} Rc2 (35... Rh2+ 36. Kg5 f3 37. Rd3 f2 38. Rf3 gxh5 39. gxh5 Rh1 40. Kf5 f1=Q 41. Rxf1 Rxf1+ 42. Ke4 Rh1 43. Kf3 Re1 44. h6 Kf7 45. Kg2 Kg6 46. Kg3 Kg5 47. Kf3 Re3+ 48. Kg2 Kg4 49. Kh2 Re1 50. Kg2 Ra1 51. Kh2 Kf3 52. Kh3 Rh1#) 36. hxg6 hxg6 (36... h6 37. g5 hxg5+ 38. Kxg5 f3 39. Kg4 f2 40. Rd1 Re2 41. Rf1 Re1 42. Rxf2 Bxf2 43. Kf3 Bb6 44. Kf4 Kg7 45. Kg5 Rf1 46. Kh5 Rg1 47. Kh4 Kxg6 48. Kh3 Kf6 49. Kh4 Kf5 50. Kh3 Kf4 51. Kh2 Kf3 52. Kh3 Rh1#) 37. Rd6 {I had missed this as part of the sequence, which still leaves White losing, but now looking more dangerous.} Bf2+ {I thought for a long time here and could not come up with anything better.} (37... Bc5 $1 {ignoring the g-pawn is best.} 38. Rd3 (38. Rxg6+ $2 Kf7 39. Kg5 Be7+ {and White is finished.}) 38... Kf7 $19) 38. Kg5 Be3 39. Kxg6 { again, I'm still winning, but the pressure is starting to get to me. It's never fun when your opponent has a mate-in-one threat (Rd8).} Rd2 40. Re6 Rd8 41. Rf6 Rf8 {at this point I just wanted to try and exchange rooks to simplify down to a won piece-up endgame.} 42. Re6 f3 {here a bishop move is much simpler, but I had calculated that White couldn't take it without losing. I'm correct in the end, but had to find a desperation tactic to make it work.} 43. Rxe3 f2 44. Re7 {at first I despaired after seeing this move. The threat is a perpetual check on g7-h7. However, I soon saw the saving grace:} Rf6+ $3 { a powerful sacrifice which decides the game, comments Komodo via the Fritz interface.} 45. Kxf6 f1=Q+ {and the pawn queens with tempo, preserving my won game.} 46. Kg6 Qxb5 {the simplest way to proceed, taking more material off the board and covering the e8 square.} 47. Rg7+ {I had to calculate the following sequence before playing the queen move, but it wasn't that hard. White soon runs out of checks.} Kf8 48. Rf7+ Ke8 49. Rf5 Qd3 {a good illustration of why the queen dominates a rook in the endgame, she can work all of the angles and do things like impose pins.} 50. Kg5 b5 51. Re5+ Kf7 52. Rf5+ Kg7 53. Kf4 b4 { again, the simple but effective approach. My opponent cannot stop the b-pawn.} 54. Rg5+ Kf6 55. Rf5+ (55. Rh5 {does not save the day} b3 56. Rh6+ Kg7 57. Re6 b2 58. Re7+ Kg6 59. Rb7 Qd4+ 60. Kg3 Qe3+ 61. Kh2 Qf2+ 62. Kh1 Qf3+ 63. Kg1 Qxb7 64. Kf2 b1=Q 65. g5 Kxg5 66. Kg3 Qe1+ 67. Kh2 Qbh1#) 55... Qxf5+ $1 { and with both of us having calculated out the resulting won pawn endgame, my opponent resigned.} (55... Qxf5+ 56. gxf5 b3 57. Ke4 b2 58. Kd3 b1=Q+ 59. Kc4 Qb6 60. Kc3 Kxf5 61. Kd3 Qc5 62. Ke2 Ke4 63. Kd2 Qc7 64. Ke2 Qc2+ 65. Kf1 Kf3 66. Ke1 Qe2#) 0-1