26 October 2014

Annotated Game #138: When complex situations demand complex solutions

This eighth-round tournament game was notable for its middlegame complexity. Both myself and my opponent found ourselves somewhat adrift, with a number of difficult and unclear decisions to make, although it led to an exciting battle all the way to the end. The complications begin after I prematurely relieve the central pawn tension on move 15, then allow Black to win the exchange. At the time, I was rather disgusted with myself, but decided (correctly) to fight on and search for compensation. In return, I got a pawn back and some active play. Both king positions are vulnerable - mine more so - but after both sides engage in ill-advised pawn-grabbing and miss subtle attacking possibilities, I manage to force a draw.

This was a very demanding game for both of us and it illustrates well how Class players too often go for moves that are more obvious, or that simplify the game to our own detriment. Instead, we should not be afraid of complexity, but rather strive to break down the position to the best of our abilities and make clear evaluations of each element. This would have helped me on the move 15 decision, for example, which showed poor judgment along with a failure to look far enough ahead at my opponent's possibilities.

[Event "?"] [Site "?"] [Date "????.??.??"] [Round "?"] [White "ChessAdmin"] [Black "Class B"] [Result "1/2-1/2"] [ECO "A26"] [Annotator "ChessAdmin/Houdini/Komodo 8"] [PlyCount "67"] [EventDate "2012.07.??"] [EventRounds "9"] {A26: English Opening vs King's Indian with ...Nc6 and d3} 1. c4 e5 2. Nc3 Nc6 3. g3 g6 4. Bg2 Bg7 5. Nf3 d6 6. d3 h6 {this is played earlier on in the sequence than is normal in this line. Black has in mind an early thrust with his kingside pawns.} 7. O-O f5 8. Rb1 Nf6 (8... a5 9. a3 Be6 10. Nd2 Nf6 11. b4 axb4 12. axb4 O-O 13. Nd5 Nd7 14. Bb2 Kh7 15. Bc3 Bf7 16. Ra1 Rxa1 17. Qxa1 Nd4 18. Bxd4 exd4 19. b5 Re8 20. Re1 Nc5 21. Nb4 g5 22. Qa2 Bg6 23. Nb3 {Gagare,S (2117)-Barua,D (2483) Mumbai 2009 1/2-1/2 (48)}) 9. b4 g5 {Black accelerates his kingside plan of gaining space with his pawns and eventually assaulting White's king position. White however can prevent Black from fully seizing the initiative with some immediate threats on the queenside and in the center.} 10. b5 $14 {Komodo 8 shows a slight plus here (around half a pawn). White enjoys a small lead in development and should be able to usefully exploit the space advantage on the queenside, as well as control of d5.} Ne7 11. Qc2 $146 { White has a lot of choices here, with none of them being obviously better. Developing the queen to b3 appears more effective, though, as it fights for d5 and the a2-g8 diagonal offers greater possibilities. The text move eyes the e2 pawn and the e4 square for defensive purposes, but after a future Nd2 this queen placement is not as useful.} f4 {an obvious follow-up, also opening up the diagonal for the light-square bishop.} 12. Ba3 {the only effective square to develop the bishop, given Black's pawn structure.} O-O 13. c5 {note how with the queen on b3 this would have more bite to it, as it would come with discovered check.} (13. Nd2 {as a preparatory move looks effective here, as the Nf3 is doing little where it is and the long diagonal is now opened up for the bishop.}) 13... fxg3 (13... Be6 $11) 14. fxg3 $11 {an uninspired, if safe choice.} (14. Qb3+ $5 {leads to a more imbalanced position, for example} Kh8 15. hxg3 Nf5 16. cxd6 cxd6 17. e4 {this blocks the Bg2, but drives away Black's well-placed knight and establishes more central control in return, giving White a small plus.} Ne7 18. Qb4 $14) 14... Ng4 15. cxd6 {resolving the tension may simplify the situation, but it is in White's best interests to make Black have to focus on potential unresolved issues in the center, rather than on his kingside play. The disappearance of the c-pawn also allows Black to exploit the a7-g1 diagonal, which I missed during the game.} cxd6 16. Qb3+ Kh8 17. Qb4 Qb6+ $15 18. Kh1 Nf2+ 19. Rxf2 {this is forced, but Black despite gaining the exchange does not in fact gain a huge advantage as a result.} Qxf2 20. Qxd6 Nf5 21. Qc7 {White's activity in Black's backfield partially compensates for the material. Black also does not have a real attack on the kingside, although he tries his best.} Ne3 22. Rg1 Bf5 23. Bxf8 {I claw back the material, although Black still has a positional advantage, with more threats available to him.} Rxf8 24. Qxb7 {I thought for a long time here and chose poorly. Black however fails to find the continuation that would give him a significant advantage.} (24. Qd6 $5 $15) 24... g4 {the obvious move, but not best.} (24... e4 $5 25. dxe4 Bg4 $17 {is found by the engines. Now White's best is} 26. e5 Nxg2 27. Rxg2 Bxf3 28. exf3 Qe1+ 29. Rg1 Qxc3 $17) 25. Nh4 $11 {I was pleased to have found this move, which holds everything together nicely for White, although it remains a complicated defense and I do not play it well. } (25. Qxa7 $2 gxf3 26. Bxf3 e4 27. dxe4 Bh3 $19 {and now an exchange sacrifice on f3 will be decisive.}) 25... Bh7 26. Ne4 $2 (26. b6 {is one of those "little moves" that engines find so easily.} axb6 27. Qxb6 e4 28. Nxe4 Qxe2 29. Qb1 $11) (26. Qxa7 $6 {is inferior:} e4 27. Nxe4 Qxe2 {and unlike in the previous variation, White's queen remains locked out of the defense on a7.} ) (26. Nd1 {is the correct knight move.} Qxe2 $2 27. Nxe3 Qxe3 28. a4 Bxd3 29. a5 $11 {and Black will have to give up his kingside attack in order to stop White's queenside pawns.}) 26... Qxe2 $17 {I am now in some trouble and start making more desperate moves.} 27. Nc5 $2 {luckily Black goes pawn hunting, ignoring the point of my active play.} (27. Qxa7 Nxg2 28. Nxg2 Qxd3 29. Nc5 $19 ) 27... Qxa2 (27... Nd1 {instead would seal the game for Black, as White would be forced to give up material and/or make his king vulnerable to mate threats.} 28. h3 gxh3 $19) 28. Be4 {this looks good, but has one major flaw.} (28. Qe7 $11 {White would be fine here, due to the follow-up threat of Ne6.}) 28... Rf2 {Black again fails to find the best continuation.} (28... Qf2 $5 {threatening . ..Nf1 and mate on h2.} 29. Bg2 Nxg2 30. Ne6 Qf7 31. Qxf7 Rxf7 32. Kxg2 Bxd3 $17 ) 29. Qb8+ $11 Bf8 30. Qxe5+ Kg8 31. Bxh7+ $4 {in another complex situation, my calculation again fails me. Here simplification does not help the situation. } (31. Qe6+ {would take the game safely into drawish territory.} Qxe6 32. Nxe6 Bxe4+ 33. dxe4 $11) 31... Kxh7 $19 32. Qe4+ {this looks like it should be fine, with Black's king so exposed, but interposing the knight on f5 would give Black the necessary extra tempo to activate his mate threat on h2.} Kg8 { Black has a mate threat} (32... Nf5 $1 33. Qb7+ Kg8 {and now White has nothing better than} 34. Ng2 Bxc5 $19) 33. Qg6+ $4 {another huge miscalculation, but my opponent is having similar troubles calculating ahead in a complex position. } (33. Qe6+ $142 {saving the game} Qxe6 34. Nxe6 $11) 33... Bg7 $2 (33... Kh8 { eventually wins for Black.} 34. Qf6+ Rxf6 35. Ne4 Rf7 36. b6 Qe2 37. bxa7 Rf1 38. h3 gxh3 39. Nf6 Qf2 40. Ng6+ Kg7 41. Nh5+ Kxg6 42. Nf4+ Kf7 43. Rxf1 Nxf1 44. a8=Q Qh2#) 34. Qe8+ $11 {I finally reach a position where I can legitimately force a draw.} 1/2-1/2

17 October 2014

Annotated Game #137: A (mostly) clean win in the Caro-Kann

This seventh-round win may not be completely clean, but it sure looks better than the last few games and shows how effective the Caro-Kann can be as a counterattacking opening.  My opponent avoids theory early on, probably not having prepared anything against the Caro-Kann, and enters a harmless variation that lets Black equalize quickly.

White nevertheless plays quite aggressively as the middlegame phase is entered, signaling with 11. fxg3 that he wants to try for a kingside attack.  However, with both bishops already exchanged off and a solid structure for Black in place, this plan is over-optimistic.  Ignoring an interesting sacrificial theme on f6, which would have forced a draw with a perpetual check, White instead overextends his kingside pawns and essentially traps his own rook on the h-file.  I am then able to switch to operations on the c-file and break into White's back ranks, finishing off his king as it tries to run up the board.  Although as you can see in the notes it would have been better for me to execute some of my ideas a move earlier than I actually played them, I'm generally pleased with my performance.  There are a number of other tweaks that I found in analysis that should also help with future play of similar positions, including avoiding wasted moves and looking out for sacs on f6.

[Event "?"] [Site "?"] [Date "????.??.??"] [Round "?"] [White "Class C"] [Black "ChessAdmin"] [Result "0-1"] [ECO "B11"] [Annotator "ChessAdmin/Houdini/Komodo 8"] [PlyCount "68"] [EventDate "2012.07.??"] [EventType "swiss"] [EventRounds "9"] {B11: Caro-Kann: Two Knights Variation} 1. e4 c6 2. Nc3 d5 3. exd5 cxd5 4. Bb5+ Bd7 5. Bxd7+ Qxd7 6. d4 Nc6 7. Nf3 e6 8. Bf4 Bd6 9. Bg3 $146 Nf6 {an interesting alternative would be to have the knight head for f5 via e7, given the presence of the Bg3 and the chance to add pressure on d4.} 10. O-O Bxg3 { an example of how Class players prematurely release tension. Exchanging now does not offer any advantage to Black.} (10... Rc8 $5) 11. fxg3 {a weakening move that indicates my opponent is looking to play aggressively by opening the f-file for his rook.} a6 {Controls b5, but this is essentially a wasted move, as White is not threatening anything on the kingside. I could (and should) have castled immediately, but wanted to see more of my opponent's intentions first.} 12. Qd2 O-O {I correctly assess that Black is solid and should not worry too much about White's threatening gestures.} 13. Rae1 Rac8 14. Ne5 { an impatient move. White was evidently hoping for an exchange, which would be to his favor with a pawn established on e5 and no good retreat square for the Nf6. However, I do not oblige and simply retreat the queen to a better square.} Qd8 15. h3 {White returns my earlier favor of a wasted move (...a6) with one of his own.} (15. Nxc6 bxc6 16. Na4 Ne4 $15) 15... b5 (15... Nxe5 {now is good for Black, since the Nf6 has d7 for a retreat square.} 16. dxe5 Nd7 17. Rf2 Qb6 18. Na4 Qc7 $17 {Komodo 8 here considers Black to have by far the better position, up a pawn equivalent. White's pieces are scattered and uncoordinated, while Black will have strong pressure on the queenside and against the e5 pawn. }) 16. a3 {Consolidates b4, notes Houdini via the Fritz interface.} Qc7 { a passive move and it also overlooks a tactical shot on f6, now only protected by the g-pawn, which would give White a forced draw.} (16... Nxe5 17. Rxe5 Rc4 $17) 17. g4 (17. Rxf6 gxf6 18. Qh6 fxe5 19. Qg5+ Kh8 20. Qf6+ $11) 17... Nxe5 $15 {now I play the right move.} 18. Rxe5 Nd7 19. Rh5 {White continues to play agressively and over-optimistically.} Rfe8 (19... Nb6 $5 {threatening to move to c4 and also freeing up the queen on the seventh rank.} 20. g5 Qe7 $15 { lending support to the f-file.}) 20. Qf2 {a rather obvious threat. I parry it easily, but could have obtained an advantage instead.} Nf6 {this puts the knight back into the tactical line of fire, which White could have exploited with a similar sacrificial theme as in the previous variation, the removal of the g-pawn.} (20... f6 21. g5 f5 $17) 21. Rh4 (21. Rg5 Kh8 22. Rxg7 Kxg7 23. Qxf6+ Kg8 24. Qg5+ Kh8 {the best option Black has is to continue the repetition of moves.}) 21... Rf8 {a rather ineffectual move.} (21... h6 { this little prophylactic move would have been excellent, controlling g5 and shoring up Black's defense.} 22. Rxh6 Nxg4 23. hxg4 gxh6 24. Qf6 $15) 22. g5 $11 Ne8 23. Qf3 {another obvious threat, this time to head to h5.} g6 {if White had a dark-square bishop, Black could be in trouble. In this case, however, White's attack has nowhere left to go.} 24. Rh6 {White's optimistic desire to attack now goes too far. The rook visually looks more threatening on h6, perhaps the reason for my opponent's choice, but it in fact does nothing for him there and becomes a liability shortly.} (24. Rf4 Nd6 $15) 24... f5 ( 24... Nd6 {immediately is better.} 25. Qf6 Rfe8 $17) 25. g4 $2 Nd6 $19 26. Qe3 Ne4 27. Nxe4 $2 {an oversight. But White was lost anyway, states Houdini via the Frtiz interface.} (27. gxf5 exf5 28. Nxd5 $19 Qd8 29. Nf6+ Nxf6 30. gxf6 Qxf6 31. c3 Rce8 32. Qf4 {White has to maintain control over the c1-h6 diagonal to prevent a queen fork on g5, which allows Black to penetrate.} Re2 $19) 27... fxe4 {opening the f-file now makes all the difference for Black.} ( 27... dxe4 $6 28. c3 fxg4 29. Rf6 $16) 28. Rf6 (28. c3 Rxf1+ 29. Kxf1 Rf8+ 30. Kg2 Rf3 $19) 28... Qxc2 {White is now effectively helpless as Black penetrates the back ranks with the heavy pieces.} 29. Qf4 Rxf6 30. Qxf6 Qb1+ $19 31. Kf2 Rc2+ 32. Kg3 Qg1+ 33. Kf4 Qh2+ 34. Ke3 Qg3+ {missing the more elegant mate in one in favor of the mate in two.} (34... Qd2#) 0-1

12 October 2014

Annotated Game #136: What exactly happened?

In this sixth-round tournament game, if one looks at it in the early middlegame then White appears to be sitting pretty, having accomplished all he could hope for out of the opening by move 18.  Indeed, as of move 26 I still had all the cards.  At this point, however, I "lose the thread" of the game and start a strategic downhill slide, becoming distracted from my queenside-based pressure while not properly defending the kingside.  The ineffectiveness of my strategic flailing is highlighted around move 29, as pieces begin to simply shuffle back and forth.  Despite this slide and Black's subsequent takeover of the initiative, my position was objectively at least equal until the board sight blunder on move 40, where my opponent seals the win with material gain.

So was it a simple tactical error that lost the game?  In reality, my mental state was poor after failing to grasp what was needed in the position and handing the initiative and its accompanying pressure over to the opponent.  The pressure of defending successfully (even if not optimally) eventually exhausted me and contributed substantially to the actual game-losing error.  On the other hand, if I had stuck religiously to my thought process, using CCT would have prevented the loss.  In the end, the result came from a combination of factors - tactical, strategic and psychological - as is the case with most chess games.

Even though I understand how the game evolved after analysis, it still makes me shake my head and wonder what exactly happened, especially after having a position that any English Opening player would love to see.  At least the analytic process should help me play stronger in future such situations.
[Event "?"] [Site "?"] [Date "????.??.??"] [Round "?"] [White "ChessAdmin"] [Black "Class B"] [Result "0-1"] [ECO "A24"] [Annotator "ChessAdmin/Houdini/Komodo 8"] [PlyCount "114"] [EventDate "2012.07.??"] [EventType "swiss"] [EventRounds "9"] {A24: English Opening vs King's Indian: Lines without ...Nc6} 1. c4 Nf6 2. Nc3 g6 3. g3 Bg7 4. Bg2 O-O 5. Nf3 d6 6. O-O e5 7. d3 c6 8. Rb1 Na6 {Black evidently wants to keep the d7 square free so he can develop his light-square bishop, but a6 is not a good square for the knight.} 9. b4 Bd7 10. b5 Nc5 11. Nd2 Rb8 12. a4 (12. bxc6 $5 Bxc6 13. Ba3 {might put more pressure on Black.}) 12... Qc7 13. Ba3 Ne6 14. e3 {Controls d4+f4} (14. Nde4 Nxe4 15. Nxe4 c5 16. e3 {would be an improved version of the idea.}) 14... Rfd8 15. Qb3 b6 $6 {an unnecessary weakening move.} (15... Nc5 $5 16. Bxc5 dxc5 17. Nde4 $11) 16. bxc6 Bxc6 17. Nb5 Bxb5 18. axb5 $16 {White has achieved his positional goals on the queenside, with a powerful Bg2 helping control the light squares and the Ba3 exerting pressure on d6. Finally, the backward a-pawn can become a target.} Nc5 {Black correctly blocks the bishop, which I decide is best traded for the centrally placed knight.} 19. Bxc5 dxc5 {Opposite coloured bishops appeared, notes Houdini via the Fritz interface. Right now White's is much better.} 20. Ne4 Ne8 (20... Nxe4 $5 21. Bxe4 f5 22. Bd5+ Kh8) 21. Rfd1 f5 {gaining space and kicking the knight.} 22. Nc3 {with an eye on the d5 square.} Nf6 23. Nd5 { if Black exchanges on d5, then White's bishop becomes even more dominant from its central position.} Qf7 24. Nxf6+ Qxf6 {forced, due to the skewer otherwise after Bd5.} 25. Bd5+ Kh8 26. Ra1 {White remains comfortably in control and correctly increases the pressure against the backward a-pawn.} Rd7 27. Ra3 { this is not in fact a bad move, but it marks the beginning of my strategic confusion. I did not understand that my queen should be redeployed along the second rank, as it rather useless on b3. This should be accompanied by increased pressure along the a-file with doubled rooks. In the game I understood the benefits of doubling rooks, but got distracted by Black's pawn advances. As a final point, moving Ra6 would have been just as good as the text move for doubling purposes and would have also blockaded the a-pawn and exerted some indirect pressure along the sixth rank.} f4 {Black wins space, notes Houdini.} 28. Rf1 {White is still fine objectively, but this move starts the strategic slide.} Rf8 {with the obvious threat of ...fxe3, as the Rf1 is now under-protected.} 29. Raa1 $11 {I've now simply wasted several tempi on the queenside and have effectively lost the initiative, although the position remains equal.} (29. Qd1 {would reposition the queen to a much more effective square and accomplish the same thing as the text move to protect the Rf1.}) 29... f3 {this certainly looks threatening, establishing a mate possibility on g2, but White can deal with this resaonbly in several ways.} (29... Qg5 $5) 30. h3 (30. Be4 Qe6 31. Qd1 Rdf7 32. Kh1 Qh3 33. Rg1 {and White has neutralized Black's threats, with good prospect of resuming his queenside pressure.}) 30... g5 31. Kh2 Qh6 {here I only saw the h-file threat and did not note that Black now could open up on the long diagonal with the Bg7.} 32. Rh1 $6 (32. Be4 $14 { is a clever way of blocking the advance of the e-pawn, while maintaining protection of the d3 pawn and controlling f5 and g6.}) 32... e4 $11 33. Rad1 exd3 {this should let White off the hook.} (33... g4 $5) 34. Qxd3 $14 Be5 35. Kg1 {this should still be good enough for a draw, but bolder, more active defense would have been better.} (35. Qe4) 35... Qd6 {although Black is not making objectively strong moves, he retains the initiative and succeeded in pressuring me psychologically as well.} (35... Qg7) 36. Qf1 $6 (36. Qc2 $5 { would have retained some chance for advantage. For example} Qc7 37. g4 Qd8 38. Kf1) 36... Bxg3 $11 37. Bxf3 {I was proud of finding this defensive move, which I assessed would lead to a draw.} Bxf2+ 38. Kxf2 Qf6 39. Kg2 (39. Rxd7 { is a more creative solution.} Qxf3+ 40. Ke1 Qxe3+ 41. Qe2 $11) 39... Rdf7 40. Bd5 $4 {the losing move. I simply miss the long-range check on b2.} (40. Rd5 $11 {was a good chance to save the game, notes Houdini.}) 40... Qb2+ $19 41. Kg1 Rxf1+ 42. Rxf1 Rxf1+ 43. Kxf1 Qd2 {the power of the queen in the endgame is now demonstrated and my opponent plays well in converting the point. I didn't want to give up prematurely, but the struggle is essentially hopeless.} 44. e4 Qd1+ 45. Kg2 Qe2+ 46. Kg1 Kg7 47. Rh2 Qe1+ 48. Kg2 Kg6 49. Rh1 Qe2+ 50. Kg3 (50. Kg1 h5 51. Rh2 Qe1+ 52. Kg2 g4 53. hxg4 hxg4 $19) 50... Qe3+ 51. Kg2 Kh5 52. Rf1 Kh4 53. Rf3 Qe2+ 54. Rf2 Qe1 55. Rf3 h5 56. Ra3 $2 (56. Rf2 $19 { what else? notes Houdini. Instead, the end comes sooner.}) 56... Qd2+ 57. Kf3 g4+ 0-1

05 October 2014

Annotated Game #135: Stopping the bleeding

The best thing that can be said about this fifth-round tournament game is that I escaped with a draw and stopped the bleeding on the scorechart.  As Black, I commit a couple of minor inaccuracies in the opening, especially by misplacing the queen's knight, which then grow into major structural deficiencies.  I was fortunate that my opponent could not find the winning idea in the end, which would be to use a pawn lever to pry open Black's position.  This was another example and an unfortunate continuation of my weak play in the tournament, with worse to come before it gets better.

[Event "?"] [Site "?"] [Date "????.??.??"] [Round "?"] [White "Class B"] [Black "ChessAdmin"] [Result "1/2-1/2"] [ECO "D12"] [Annotator "ChessAdmin/Houdini/Komodo 8"] [PlyCount "72"] [EventDate "2012.07.??"] [EventType "swiss"] [EventRounds "9"] {D12: Slav Defence: 3 Nf3 Nf6 4 e3 Bf5} 1. d4 d5 2. c4 c6 3. Nf3 Nf6 4. e3 Bf5 5. Nc3 e6 6. Qb3 Qb6 7. Nh4 Bg6 {if Black wants to postpone the queen exchange, normally ...Be4 is played to provoke f3 first.} (7... Be4 8. Qxb6 axb6 9. f3 Bc2 10. cxd5 cxd5 11. Bd2 Nc6 12. Bb5 Bb4 13. g4 g5 14. Ng2 h5 15. gxh5 Rxh5 16. h4 Ke7 17. Kf2 Bxc3 18. Bxc3 g4 19. Nf4 Rh6 20. h5 gxf3 21. Rh4 Be4 22. Rg1 Rxa2 23. Bxc6 bxc6 24. Bb4+ Ke8 25. Ba3 b5 26. Kg3 b4 27. Bxb4 Rxb2 28. Bc5 Rh8 29. Ra1 Rg8+ 30. Ng6 fxg6 {0-1 (30) Rau,H (2464)-Jaracz,P (2558) Germany 2012}) 8. Be2 Qxb3 9. axb3 Na6 $146 {the idea is to jump to b4 if possible, otherwise develop via c7. However, this is not the most challenging continuation and could become awkward for Black.} (9... Bc2 {is much stronger.} 10. c5 Bxb3 11. Nf3 Bc2 $11 {Komodo considers this equal. White has an advantage in development and space that offsets the loss of the doubled pawn, but I'd still prefer to play Black.}) 10. Nxg6 $14 hxg6 {White has the pair of bishops, notes Houdini via the Fritz interface.} 11. O-O Bd6 12. g3 {without any Black pieces able to exploit the resulting light-square weakness, this is the best way to block the threat on the h-file.} Nc7 {without any obvious targets reachable from b4, I decided to move the knight off the rim to c7. This may not have been best, however, and this knight struggles to find a useful place.} (12... Nb4 13. Bd2 a5 $14) 13. f3 {Covers e4+g4} a6 {this was a somewhat prophylactic idea, geared toward preventing a White piece occupying b5 after a pawn exchange on d5, if I chose to recapture with the c-pawn. It also frees up the Ra8 to move away from defending the pawn.} 14. e4 {a logical follow-up to White's previous move.} Bb4 {A comfortable square for the black bishop} (14... e5 {is preferred by Komodo 8. Black fights for the center this way.}) 15. Bf4 { targeting the awkwardly placed Nc7.} Kd7 $6 (15... O-O-O $5) 16. Bd3 Ng8 $6 ( 16... Rad8 17. Bg5 Be7 18. e5 $14) 17. Na4 $16 {White has taken over the initiative and Black's pieces are poorly coordinated, with no counterplay.} Re8 18. Nc3 (18. cxd5 {would start a sequence opening up the position to White's advantage, using his two bishops and more active rooks.} exd5 19. Nb6+ Kd8 20. exd5 Nxd5 21. Nxd5 cxd5 $16) 18... Ne7 19. Kg2 Bxc3 (19... f6 $5) 20. bxc3 $16 Rh7 {trying to generate some counterplay on the h-file.} 21. h3 Reh8 22. Rh1 Ne8 23. cxd5 exd5 (23... cxd5 {would maintain a stronger central pawn presence. }) 24. g4 Nc7 25. Bd2 Re8 $6 (25... Ne6 {is the logical follow-up.} 26. exd5 cxd5 27. b4 $16) 26. Rae1 Rhh8 27. f4 Rhf8 {I still have no real idea how to improve my position at this point.} 28. Rhf1 dxe4 (28... f5 {is a better way to challenge in the center, putting Black's pawns and pieces to work.}) 29. Bxe4 f5 30. Bc2 Nb5 31. Bd3 Nc7 {this knight is wasting a truly impressive number of tempi without doing anything to justify its movements.} 32. Re2 Ncd5 33. Kg3 Nc7 $6 (33... fxg4 34. hxg4 g5 35. f5 $16) 34. Rfe1 (34. Re5 {is the best way to make progress.} Nc8 35. gxf5 gxf5 36. Rxf5) 34... Ncd5 $2 {the engine doesn't like this, as it keeps the Re8 masked and allows White to dominate the file with his rooks.} (34... Nc8 $5) 35. Re5 Nc7 36. R5e2 (36. c4 {is what the engines point out as the way to continue pressing White's advantage, as it takes away the d5 square and allows him to ratchet up additional piece pressure with Bb4 or Ba5.}) 36... Ncd5 {my opponent did not see an obvious way to make progress, so accepted the draw.} 1/2-1/2