29 March 2016

Improvement Program List - March 2016

In keeping with the eclectic training program format I am following, here is the list of chess resources that I am currently working on.  Once they are all completed, I'll generate a new list.  As I finish them, I'll be posting my "completed" thoughts for each.  The previous list can be found here.  As always, I'll also be pursuing my regular and constant training activities such as analyzing my own games, doing tactics exercises and going over annotated master games.

Mastering Opening Strategy by GM Johan Hellsten (openings)

Chess Strategy for Club Players (2nd edition) by Herman Grooten (middlegame)

Chess Endgames 1: Basic Knowledge for Beginners by GM Karsten Mueller (fritztrainer - endgames)

20 March 2016

DVD completed: Essential Endgame Knowledge with IM Dr. Danny Kopec


I recently (and finally!) completed this DVD, which although non-interactive (having been produced before interactivity became common) was certainly a useful video lecture.  At 1 hour 50 minutes long, it provides a foundation of basic endgame knowledge, focusing on universal concepts and principles by using a number of classic, real-world and composed examples.

The contents are broken into three main sections:

I. Elementary Endings - includes the standard mates with major pieces (Queen, Rook, 2 Bishops, Bishop and Knight) and no pawns, along with positions with a single pawn on the board.  All of these (with the exception of the last one) are demonstrated in a video of IM Kopec using a (very nice) wooden board, a presentation method which actually works pretty well.
  • K+Q vs K
  • K+R vs K (the first two are both are demonstrated using a "shrink the box" technique, not the only one possible but easy to remember)
  • K+2 bishops (demonstrated using a "shrink the triangle" technique consistent with the previous concept) (Workout at Chess.com)
  • K+P vs K (includes the idea of the opposition, rule of two ranks, principle of lead with the king, principle of maximizing distance (in files) between kings, stepping into the "pawn square", rook pawn exception)
  • K+B+N vs K.  This is the tough one that sometimes professional players can't get, since it is by far the most complex and the most rare.  The presentation for the technique is given using a 2-D computer board and IM Kopec focuses on presenting concepts such as the "good formation" of the bishop and knight (on the same color square) that the viewer can use.
II. King and Pawn Endings - concepts include:
  • Better king position
  • Better pawns - structure (more space), mobility (more tempi available)
  • Zugzwang - sometimes it's enough to do nothing and force your opponent to move
  • Strong and weak pawn structures
  • Triangulation
III. Rook and Pawn Endings

First are the three primary factors, in order of importance:
  • Relative positions of the rooks and their activity - rook activity is primary, even over material; rooks should be behind passed pawns (both yours and your opponent's)
  • Better king - more advanced, more active
  • Pawn structure; the superiority of "relatively outside" passed pawns
Other thematic content:
My comments:
  • It would have been nice to have seen some of the examples played out more, including the last two classic games listed above.
  • There are a significant number of "verbal typos" made throughout the lecture (wrong square announced, etc.) although most of them are corrected.  It's a little annoying sometimes, but not a reason to avoid watching the lecture.
  • The mix of 3-D wood board and 2-D computer board (including all of the rook and pawn endgame material) actually works pretty well and gives some variety in the first part of the DVD, as if you were really working with a coach over a wood board.
Of course practice makes perfect and working out endgame positions with a computer opponent can be very helpful.  The Chess.com Computer Workout section is one such online resource.

17 March 2016

Annotated Game #152: To the pain


The quote from The Princess Bride is apt to describe this last-round tournament game, which for me was one long torture session.  I misplay the opening, which took a rather weird (or at least unfamiliar) turn in an English Four Knights around moves 8-9.  I ended up with a painfully inferior position where my opponent had all of the chances.  I then spent a long time simply surviving, then clawing my way back into having real chances, but fell just short of a positive result, in large part due to the mental exhaustion of the effort required in getting to that point.

These types of games are very tough to play and also difficult emotionally to analyze, since you get to relive some of that pain along the way.  However, on the positive side, doing that can help burn into your mind how not to play like that the next time you face a similar situation.

ChessAdmin - Class A

Result: 0-1
Site: ?
Date: ?
A28: English Opening: Four Knights Variation
[...] 1.c4 e5 2.¤c3 ¤f6 3.¤f3 ¤c6 4.e3 d5 5.cxd5 ¤xd5 6.¥b5 ¤xc3 7.bxc3 ¥d6 8.d4 e4 this is the first time I had faced this move, which is unusual but not necessarily bad. From Black's perspective, it gives up the center but gains space on the kingside for further operations. 9.¤d2 £g5 this move discombobulated me a bit, since it was unexpected, although it's a natural follow-up to Black's previous one. Normally the queens don't come out this early in an English. 10.¥xc6+?!
10.¥f1 an example of where concrete considerations override standard opening principles. Here White moves a piece twice, back to its starting square, but it's clearly the best move. The king is secure enough on e1 and White can get play on the kingside and in the center. Eventually I hit on this characteristic of the position in the game, but only in desperation and after digging myself a large hole. 10...£g6
10...bxc6 11.g3³ weakening the light-square complex and making the king much less secure. 11...£g6
11...¥g4!? is more challenging, immediately exploiting the holes in White's position.
12.c4?! a slow move that also opens the a5-e1 diagonal, further weakening the king. Black with his two bishops is much better placed to exploit a more open position. (12.£a4 hitting the weak c6 square 12...¥d7 13.¥a3) 12...c5 13.¥b2 here the bishop on the long diagonal isn't very effective. Ba3 would still be better, also keeping the b-file open. (13.O-O!?³) 13...O-Oµ White has an awful position strategically speaking, as it's full of holes and none of the pieces are really doing much, while Black's are well positioned. The engine gives Black a full pawn equivalent advantage. 14.¦b1 ¥g4 15.£c2 ¦ae8 (15...cxd4 16.¥xd4 c5 17.¥c3 ¥h3µ) 16.d5 locking the center here looks useful, since the pawn advance gains White space and opens the long diagonal, but in concrete terms it actually does nothing for my game and in fact reduces my dynamic possibilities. (16.dxc5 ¥xc5 17.¥c3µ) 16...f5 normally an f-pawn advance at this stage would be a weakening move for Black, even if it has a good attacking purpose, but none of my pieces can do anything in the center or elsewhere to counter it. 17.¤b3? there's not a lot that I can do here, but this is still bad. The idea was to activate the knight and get it to c6 via a5, but this is a slow plan and takes the knight away from a defensive role on the kingside. (17.h3 ¥h5) 17...£h6 an effective move, looking to penetrate the kingside. The immediate f-pawn advance would be even more effective:
17...f4 18.gxf4 ¥h3 at the cost of a pawn, Black has pried open the kingside and White's destruction is imminent. Let's see how that would play out: 19.¢d1 £g2 20.¦e1 ¥g4+ 21.¢c1 ¦xf4!22.£c3
22.exf4?22...¥xf4+ 23.¤d2 e3 24.fxe3 ¦xe3 25.¦d1 ¥xd1 26.£xd1 ¦f3
22...¥e5 23.£c2 ¦xf2−⁠+
18.£c3 sadly, my only counterplay at this point is threatening a rather obvious mate on the long diagonal. 18...¦e7 (18...¥f3 19.¦f1 ¦f7−⁠+) 19.£a5? by this point I'm playing rather randomly and desperately.
19.h4!? would at least try to address Black's threats on the kingside.
19...¥f3 keeping a steady advantage.
19...¥xg3!? is an effective tactical blow. 20.£xc5 ¥xf2+ 21.¢d2 ¦ef7−⁠+ and with ...f4 coming, White's position will collapse.
20.¢d2 at least by this point I understand the seriousness of my position and make the correct choice to sacrifice the exchange for defensive reasons. Although Black still has a won game, this marks an initial psychological turning point in my climb back into the game. 20...¥xh1 21.¦xh1 £h3 22.¦e1 it would have been wiser to continue my king's flight. (22.¢c1 ¦b8 23.£d2−⁠+) 22...£xh2 23.¦e2 I've now "turtled up" my position so there is no imminent breakthrough by Black, although the material deficit means I should still lose. 23...h5 24.¤xc5? (24.¢c2 continuing the king flight is what the engine recommends.) 24...f4 (24...h4!?25.¤b3 h3 26.c5 £g1−⁠+) 25.gxf4 h4 here my opponent misses the ...Qg1 idea, which would seal the win. (25...£g1 26.¤b3 £b1−⁠+) 26.¤e6 although Black is still winning by far, I finally am starting to have some potential threats appear on the board. 26...¦b8 27.¢c2 protecting the Bb2 and opening the diagonal for the queen to return. 27...c5 here my opponent starts to go astray. His winning advantage is on the kingside, but he shifts play back and forth now between that and the queenside, to the detriment of both.
27...h3!? and further material loss by White is inevitable, as the pawn can't be stopped otherwise.
28.£d2?? this should lead immediately to a loss. Furthermore, White can actually obtain a winning position, but it would have required me to sacrifice the Ne6, which I didn't even consider.
28.dxc6! is possible and White now causes serious headaches for Black, using the open lines around his king. 28...¦xe6 29.£d5 £h3 30.f5 this is the tactical point of the sequence. 30...¦be8 31.¦d2 it's important not to pull the trigger too soon on the Re6, which is going nowhere. 31...¥c7 32.¦d1 ¦8e7 33.fxe6 £xe6 34.¦h1+⁠− and shockingly White is now winning.
28...£g1 (28...h3 again is more to the point.) 29.¤g5 with the idea of preventing ...h3, but Black could still in fact play it. Black also has a win using tactics on the b-file, but ignores it. (29.¦e1 £g2 30.£e2 h3−⁠+) 29...¦eb7 with Black having delayed this, I can still (desperately) defend. (29...¦xb2+ 30.¢xb2 ¦b7+ 31.¢a3 £b1 32.£c3 h3−⁠+) 30.¥c1 £h1? this allowed me a chance to get closer to equality, but I didn't see it at the time. (30...¦b4) 31.¦e1?
31.£e1 had to be tried to avoid defeat, notes Komodo via the Fritz interface. 31...£xe1 32.¦xe1 ¦b1 33.¤xe4 ¥c7 34.¦d1 h3µ
31...£g2−⁠+32.¦e2? (32.£e2 ¥e7 33.f3 £xe2+ 34.¦xe2 ¥xg5 35.fxe4−⁠+) 32...£g4? this time, however, I seize the chance to claw my way back into the game.
32...¦b1 finishes off the opponent, says the engine with no sympathy whatsoever. 33.¥a3 ¦a1−⁠+
33.¤xe4³33...£d7 and now I can even equalize. 34.¤c3 it was an exhausting struggle to get to this point. Material is now roughly balanced and my defenses after the king walk are holding. The h-pawn is no longer a threat, either, as it does not have enough support to get past my second rank. 34...¦f8 letting up pressure on the b-file, which is a mistake. (34...¦b4 35.£d3 £b7 36.£e4) 35.f3² the correct response. The second rank is opened for the Q+R battery and I can start mobilizing my central pawns. 35...¦f5 (35...£f5+ 36.£d3 £h5 37.¦h2²) 36.£e1 (36.¦h2± first is better, keeping the h-pawn firmly in check.) 36...¦h5 37.e4 h3 38.£g3 ¥c7 (38...¦b4 39.¢d3±) 39.¦h2 ¦b6?! this does nothing for Black. (39...¥a5!? was necessary here. 40.£g4 £e8 41.e5²) 40.e5±40...¦bh6 I don't understand what my opponent was thinking with this maneuver. I believe he was rather exhausted himself by this point. (40...£f5+ 41.¤e4 ¦g6 42.£f2±) 41.e6 this is too impatient.
41.£g4 is much better, supporting the pawn advance first. 41...£e8 42.d6 ¥xd6 43.exd6 ¦xd6 44.¦xh3 ¦xh3 45.£xh3±
41...£e8²42.£f2 shifting attention to the c-pawn. 42...£g6+? a check for a check's sake, seemingly. The next move centralizes my knight to good effect, making it significantly more powerful despite the pin. (42...¦g6!?²) 43.¤e4+⁠− now once the king moves, the knight is very well placed to support the d6 push, as well as controlling f6 and attacking c5 again. 43...£f5 44.£xc5 ¥xf4 45.¥xf4? with this move White loses his initiative
45.e7 passed pawns must be pushed! 45...¢f7 46.¥xf4 £xf4 47.¦e2+⁠−
45...£xf4 now the best I can do is a draw, which was a significant letdown. 46.£f2 necessary to protect f3. 46...¦g6 47.d6 ¦xe6 and now, after a long, tiring battle, I fail to falsify my next move and check my opponent's tactics. 48.¦h1?? (48.d7 was possible 48...£b8 49.£d2) 48...¦xe4 this capture removes the guard of the Qf2 and so the pawn recapture isn't possible. Game over.
Powered by Aquarium

12 March 2016

Annotated Game #151: The Critical Position

This fourth-round tournament game illustrates well the importance of understanding the critical position in a game.  Most games have one position (occasionally more, if it's a long game) that contains a major decision point and require significant thought, both in terms of calculating and evaluating it properly.  Here the critical position occurs on move 17, something which I recognized during the game and is also evident during analysis.  I had deliberately unbalanced the position with my opening choice (9...Bf5!?), which lead to having an open g-file and a strong light-square complex, at the expense of the dark squares and my kingside pawn structure.  Over the next several moves, I correctly exploited the ideas for Black inherent in the position and developed a good initiative.  Unfortunately I failed to then find the best (really only) idea for continuing, 17...Nxg5, which immediately turned over the initiative to my opponent.

It's interesting to observe how often when one fails to play the critical position correctly, it dooms the rest of your game.  Partly that is due to objective factors, but there is also a significant psychological component.  For example, I also failed to find better options - admittedly, much harder to calculate - on moves 18 and 26.  I think that is a combination of the position actually being significantly more difficult to play, along with the earlier psychological blow coming from the sudden shift in momentum.  Aside from the concrete lessons that analysis of this game teaches me, remembering to keep looking for ways to reverse course, even after suffering a turnaround in the game, is a more general lesson to keep in mind.

Class C - ChessAdmin

Result: 1-0
Site: ?
Date: ?
B13: Caro-Kann: Exchange Variation and Panov-Botvinnik Attack
[...] 1.e4 c6 2.d4 d5 3.exd5 cxd5 4.¥f4 (4.¥d3 is the Exchange Variation main line.) 4...¤c6
4...£b6 is the only original try in this position to take advantage of the early dark-square bishop move, but it doesn't work. 5.¤c3 ¤f6 (5...£xb2?6.¤xd5+⁠−) 6.¤b5 ¤a6²
5.c3 ¤f6 this move is fine, but represents a bit of lazy thinking on my part. I should have pondered more the difference between this position and the normal Exchange Variation with an earlier Bd3.
5...¥f5 played now is the way to take advantage of the lack of a White Bd3. 6.¥d3 ¥xd3 7.£xd3 e6 with the idea of playing ...Bd6 and exchanging the other pair of bishops.
6.¥d3 now we're back to the standard Exchange Variation line. 6...g6 7.¤f3 ¥g7 8.¤bd2 O-O 9.O-O ¥f5 a deliberately unbalancing move. Safer would be ...Nh5. 10.¥xf5 gxf5 11.£c2 e6 12.¥g5 £c7 taking advantage of the bishop leaving the h2-b8 diagonal. 13.£d3 ¤e4 the natural move, hitting the Bg5 and powerfully centralizing the knight. 14.£e3 f6 correctly mobilizing the extra f-pawn. So far I am doing well in understanding the unique characteristics of the position and using them to my advantage. 15.¥f4?! (15.¥h6!?) 15...e5µ is again the strong, natural reaction. The extra f-pawn is a benefit rather than a drawback, supporting e5. 16.dxe5 fxe5 17.¥g5 Black has an active position, notes Komodo via the Fritz interface. Unfortunately my calculation and evaluation skills now fail me in this unfamiliar and complex position. 17...f4?! this is the "easy" way to rationalize the position, but unfortunately it gives White an advantage.
17...¤xg5!? I considered for a long time, along with the text move, but I did not understand how to properly follow it up, incorrectly evaluating that it would cause me to lose the initiative. 18.£xg5 and now Black in fact has a pleasant choice of how to proceed, for example 18...e4 (18...¦f6!?) (18...¦ad8) 19.¤h4 h6 20.£e3 f4 21.£h3 £f7µ
17...£f7 is also good, something I did not really consider. 18.¤xe4 fxe4 19.¤h4 £h5³
18.£d3 at the time, I did not appreciate how easily White could fix his problems with this move. The d5 pawn is hanging, which is key. 18...¤f6? this is obviously bad, but at the time I did not see an alternative.
18...¤xd2!? and Black is still in the game, notes the engine. 19.£xd5+ ¢h8 20.¤xd2 h6 21.¥h4 ¦ad8 I would have also needed to find this non-obvious move. 22.¥xd8 ¦xd8 23.£b3 ¦xd2²
19.¥xf6±19...¦xf6 not the best move, but I am already, with some desperation, thinking about how I can try to counterattack rather than simply lose without a fight. (19...¥xf6 20.£xd5+ £f7 21.£xf7+ ¢xf7±) 20.£xd5+ ¢h8 21.¤c4 (21.¤e4 ¦d8 22.£b3 ¦g6+⁠−) 21...¦g6±22.¦fe1 ¦g8 my only hope at this point is to get something going on the g-file. 23.¤h4
23.g3 is the simpler way to defuse g-file threats. 23...fxg3 24.hxg3 ¦f8 25.¤h4 ¦gf6 26.¦e2±
23...¦g4 (23...¦h6!?24.¤f5 ¦f6 25.¤xg7 £xg7 26.g3±) 24.¤d6 with the threat of Nf7+ 24...¦f8 25.¤hf5 £d7 pinning the Nd6 and therefore threatening ...Rxf5. 26.¤xg7? this allows an unusual saving tactic for Black...which however is difficult to see.
26.¦ad1 would protect the Qd5 and unpin the knight. 26...¦xf5 27.¤xf5 £xf5±
26...£xg7? the second-best move.
26...f3! the idea is to press the attack while temporarily ignoring the need to recapture a piece. White cannot simply ignore the threat. 27.¤e6 (27.g3 ¦g6µ) 27...¦xg2+ 28.¢h1 ¦f6µ and material equality will be restored, with Black having a more threatening position.
27.¦ad1?! this again allows a creative tactic...which again is not found. (27.g3± is necessary here.) 27...f3? the right general idea, but wrong sequence.
27...e4!28.¦xe4 (28.g3 now does not work: 28...e3 29.fxe3 fxg3 30.h4 g2µ) 28...¦xg2+ 29.¢h1 f3
28.g3+⁠−28...¦h4 29.¦e4 ¦xe4 (29...¦h6 30.h4+⁠−) 30.£xe4 despite being only a pawn up at the moment, White has a thoroughly winning game and I no longer have any counterplay, so the rest is easy for him. 30...¦f4 31.£e1 h5
31...£d7 would be the best try, but White can now easily defend against a mate threat on g2 or h2. 32.b4 a6 33.a4+⁠−
32.¢h1 h4 33.£g1 ¦g4 34.£f1 hxg3? this loses quickly, but frankly sometimes that is better than inevitably losing slowly. (34...£g6+⁠−) 35.fxg3 sealing the win.
35.£h3+ seems even better, comments the engine. 35...£h7 36.¤f7+ ¢g8 37.£xg4+ ¢xf7 38.fxg3+⁠−
35...e4 36.£h3+ ¢g8 37.¤f5 £g5 38.¤h6+ ¢g7 39.£xg4
Powered by Aquarium

09 March 2016

Annotated Game #150: Drawing a Win

This third-round game features a seesaw battle in the Symmetrical English, which in fact turned out to be rather unbalanced.  My opponent was evidently unfamiliar with the opening, although I misunderstood some key ideas as well (see for example move 13).  I get my plan back on track and by around move 22 am doing quite well.  Black of course does not give up and complicates things in the next several moves, while I only find lesser moves in response.

The main error on my part comes on move 26, when I make a false assumption about the solidity of my position and make an "obvious move" (see Annotated Game #149) which is tactically refuted by my opponent.  Nevertheless, I fight back and do the same to my opponent on move 31, coming up with a winning "in-between move" rather than the obvious recapture.  However, in the end I fail to find the winning idea with time running out on my clock, so I force a draw.  Disappointing, but at least the lesson learned in analysis of how I could have won the position will be with me for the future.

ChessAdmin - Class B

Result: 1/2-1/2
Site: ?
Date: ?
A38: Symmetrical English vs ...g6: 4 Bg2 Bg7 5 Nf3 Nf6
[...] 1.c4 c5 2.¤f3 ¤c6 3.¤c3 ¤f6 4.g3 g6 5.¥g2 ¥g7 6.O-O O-O 7.d3 d6 I suspected my opponent had not faced the English much and therefore decided to copy my moves for lack of any other ideas. 8.¥d2 h6 breaking the symmetry finally. Since the bishop had just moved to d2, I assume that the pawn advance was aimed at preventing a future Ng5, although this does not seem particuarly urgent, so developing a piece would make more sense. 9.¦b1 e5 this seizes space in the center, but leaves a big hole on d5. 10.a3 with the idea of forcing through the pawn advance b4. 10...a5 the obvious reaction 11.¤e1 opening up the long diagonal and looking to transfer the knight to support b4. 11...¦e8 12.¤c2 ¤d4
12...a4 13.b4 axb3 14.¦xb3 ¤d4 would be an improved version of the idea.
13.¤xd4 this was unnecessary and distracted from the whole point of the previous plan to build up the b4 advance. (13.b4 axb4 14.axb4²) 13...exd4 14.¤d5 (14.¤b5!?14...¥g4 15.¦e1 £d7) 14...¤xd5 15.¥xd5 the bishop looks nicely placed, but has no real follow-up threats and can eventually be attacked after preparing ...Be6. 15...£e7 16.¦e1 ¥h3?! bishops on h3 like this always look menacing, but without an ability for Black to further attack the g2 square (for example, with his queen), the bishop in fact is out of play here. (16...¥e6 17.¥g2) 17.b4² I finally return to the main idea of the position for White. 17...axb4 18.axb4 ¦a7 19.bxc5±19...dxc5 20.¦b6 threatening to win material after Rxg6, due to the pin on the f-pawn (the Bd5 actually contributing now to concrete threats). (20.e4!?) 20...¢h7 21.£b1 now it's my turn to line up the heavy pieces against a pawn, to greater effect. 21...¥c8 (21...¥e5 22.¦xb7 ¦xb7 23.£xb7 £xb7 24.¥xb7 ¦b8 25.¥g2±) 22.¥f4± now this bishop finally gets in the game, with a strong threat to fork on d6. 22...g5? missing the point of the last move. (22...¦d8!?±) 23.¥d6 £d7 24.¥xc5 White now dominates and has a winning position, although the game itself is still far from actually won. 24...£g4? here I fail to think more creatively to take advantage of Black's error. (24...¦a5 25.¥d6+⁠−) 25.f3 this is fine for maintaining the advantage, but not decisive.
25.¥xf7 secures the win, notes Komodo 8 via the Fritz interface. 25...¦g8
25...¦d8 26.¦xh6 ¥xh6 27.£b6 forking the Rd8, Ra7 and the g6 square with the queen.
26.¦xh6 this is the key idea I failed to find, which features both a discovered attack on the Ra7 and opening lines against Black's kind; see also the previous variation. 26...¥xh6 27.¥xg8 ¢xg8 28.¥xa7+⁠−
25...£f5 26.¥e4?? an unfortunate move that relinquishes the win, states Komodo 8. Also an "obvious move" that I should have checked more throughly. (26.£b2 is a simple way to continue. 26...¢g8 27.¦d6+⁠−) (26.¦xh6 also still works.) 26...¦xe4± Black is able to exchange the rook for two bishops, which is quite favorable and something I failed to account for, with the hanging Bc5. The false assumption was that the Be4 was immune to capture, being protected by pawns, but going the one step further in calculation would have pointed out this fallacy. 27.fxe4? the curse of the "obvious move" strikes again!
27.¦b5 would still leave White slightly better. 27...¦xe2 28.¦xe2²
27...£xc5µ28.¦f1 still not taking into account my opponent's best replies. (28.£b5 £c7 29.¦b1µ) 28...¥e6?!
28...¥h3!? now would be quite effective, since it would threaten White's king on the back rank. 29.¦d1 (29.¦xf7??29...£xb6 30.£xb6 ¦a1 31.£b1 ¦xb1 32.¢f2 ¦f1#) 29...£c7µ
29.¦xb7 ¦xb7 30.£xb7 the position is now equal, as White's extra pawns and active queen position compensate for Black's two bishops. 30...¥xc4?? this should be the losing move. (30...£a5 was possible) 31.¦f5+⁠− the simplest way to win. Here I give myself credit for not just automatically playing the obvious dxc4 (which is also good according to the computer, but creates a much more complex and unclear situation). Instead I see that the queen can be chased away from protecting the c4 square first.
31.dxc4 d3 32.¢g2 dxe2 33.¦xf7 e1=¤ 34.¢f1 £xc4 35.¢xe1 is the complex computer line, which also gives Black a lot of queen checks to follow, although it's rated better for White (eventually).
31...£a3 32.dxc4 £c1 33.¢g2
33.¦f1 is simpler and better, as the rook is not as useful on the f5 square. 33...£e3 34.¦f2 £c1 35.¢g2 £xc4 36.£xf7 £xf7 37.¦xf7+⁠−
33...£xc4 34.¦xf7 £xe2 35.¦f2 £e3 in comparison with the above variation, now it's much harder (at least for me) to find a line where White can make progress. With time low on the clock, I instead end up repeating moves and drawing. 36.¦f7
36.£d5 is the best move, getting behind Black's passed d-pawn while maintaining support of the e4 pawn and enabling the queen to move laterally to the f-file. The idea is to check on f5 - which Black cannot block - then threaten to redeploy the rook to the 8th rank to target the Black king. 36...g4 technically best, in order to divert the queen to the g-file after Qf5+.
36...d3 37.£f5 ¢g8 38.¦a2 ¥f8 39.£g6 ¢h8 40.£e8 ¢g7 41.¦f2 and wins.
37.£f5 ¢g8 38.¦a2 £f3 39.£xf3 gxf3 40.¢xf3+⁠− and the endgame win for White is trivial.
36...£e2 37.¦f2 Twofold repetition 37...£e3 38.¦f7 £e2
Powered by Aquarium