20 August 2016

Commentary: 2016 U.S. Championship, Round 11 (Krush - Paikidze)

This last commentary game from the 2016 US Championship is the decisive round 11 encounter between GM Irina Krush and IM Nazi Paikidze in the Women's section.  (Original ChessBase commentary can be found here.)  Krush by this point in the tournament, with 6 points, was out of the running, while Paikidze had to win as Black in order to catch up to WGM Tatev Abrahamyan.

As with many games, knowing the context is important to understanding the choices made by the players.  Paikidze as Black could not afford to be passive, while Krush as White had no need to strive for a win.  This dynamic I think helped shape the game from the start in terms of the opening choice (a King's Indian Attack).  White with 17. f4 provokes a complex middlegame with a number of tactical ideas lurking in the variations - mostly to Black's benefit.  Black in response sacrifices a pawn and has the initiative for almost the entire game, although Krush at one point had fought back to near-equality.  Paikidze's play illustrates some important tactical and positional concepts for improving players and the factors involved are well worth studying.

Krush, Irina (2465) - Paikidze, Nazi (2346)

Result: 0-1
Site: Saint Louis USA
Date: 2016.04.25
[...] 1.¤f3 ¤f6 a noncommital response to White's first move, while ruling out an immediate e4 as follow-up. 2.g3 d5 3.¥g2 c6 4.O-O ¥g4 this is slightly more challenging than the other standard move developing the bishop to f5. 5.d3 continuing with the standard plan of the King's Indian Attack setup. White will eventually play e4. 5...¤bd7 the knight needs to be developed in any case, and this provides the option of supporting an ...e5 push. 6.h3 ¥h5 7.£e1 getting off the h5-d1 diagonal and behind the e-pawn. 7...e5 played the vast majority of the time, forming a pawn duo in the center. (7...e6 is certainly possible, but is unambitious and drawish.) 8.e4 dxe4 not necessarily obligatory, but almost always played. The following game shows how problems can develop for Black by delaying it.
8...¥d6 9.exd5 ¥xf3 10.¥xf3 ¤xd5 11.¤c3 ¤xc3 12.bxc3 O-O 13.¦b1 £c7 14.¥d2 f5 15.£e2 ¦ae8 16.¥g2 ¢h8 17.£h5 ¥c5 18.¥g5 ¥b6 19.¦b4 ¦e6 20.¦h4 h6 21.d4 f4 22.gxf4 exd4 23.£g4 ¦g6 24.¥e4 ¦xf4 25.£xf4 £xf4 26.¦xf4 ¦xg5+ 27.¢h1 ¤f6 28.cxd4 ¥xd4 29.¦d1 ¥b6 30.¥f5 ¢g8 31.c4 ¥c7 32.¦f3 b5 33.cxb5 cxb5 34.¦b1 a6 35.¦c1 ¥e5 36.¦c6 ¢f7 37.¦xa6 ¥d4 38.¦d6 ¥c5 39.¦c6 ¥e7 40.¦c7 g6 41.¦e3 1-0 (41) Movsesian,S (2695)-Zontakh,A (2546) Loo 2013
9.dxe4 ¥c5 developing the bishop to its most effective diagonal. 10.a4 a5 preventing a b4 advance. 11.¤a3 actually the most common move played here, but scoring only 46 percent for White in the database. The point is to transfer the knight to c4. 11...O-O 12.¤c4 £c7 protecting e5 and connecting the rooks. 13.¥d2 b6 the obvious move, to ensure the a5 pawn is protected and Black's pieces are not tied down to it. 14.¤h4 intending to go to f5, but the knight ends up stuck here for a long time before exchanging itself for the bishop on g6. 14...¦fe8 developing the rook, which was doing nothing on f8. 15.¢h1 getting off the a7-g1 diagonal and preparing to push the f-pawn. 15...¥g6 anticipating the push g4 and pressuring e4, essentially inviting the following exchange. 16.¤xg6 hxg6 although White now has the two bishops, the individual minor piece trade is a fine idea for Black. Her light-square bishop was not doing anything very important and the White knight on the kingside otherwise could effectively support a pawn advance and/or could go to f5. 17.f4 a natural move, but perhaps White could have taken some more time to prepare it. Black is able to launch a counterstroke on the queenside. 17...b5 the tactics work in Black's favor if White accepts the pawn sacrifice. 18.¥xa5
18.¤xa5 is inferior, as the Na5 is out on a limb and its protectors can become overloaded, while Black has multiple other threats. 18...exf4 19.axb5 f3 20.¦xf3
20.¥xf3 cxb5³ and now Black can threaten the g3 and c2 pawns after ...Bd6.
20...cxb5 21.¦b3 ¥d6³22.¦e3 (22.¦xb5?22...¤c5µ)
18...£c8 19.axb5 cxb5 20.¤d2 the engine assesses the position as equal, as White's pieces are not as well coordinated as her opponent's and Black can start making threats along the e-file. 20...exf4 21.gxf4 ¤d5 a key move in the sequence, as the knight takes advantage of the pinned e-pawn to use d5 as an outpost and threaten to go to e3. 22.¦f3 defending the e3 square, albeit awkwardly.
22.b4!? is a recurring idea in this position that the engines identify. White at least temporarily gives back the pawn in order to better activate her pieces and deflect Black's threats. For example 22...¤xb4 23.£b1 ¤c6 24.£xb5 ¤xa5 25.¦xa5 ¦xa5 26.£xa5 ¤f6
22...f5?! this invites the advance of the e-pawn, which essentially solves White's problems with it.
22...¤7f6 would increase the pressure and not allow for the advance, as if 23.e5?!23...¤h5µ and now the f-pawn is under fire.
23.e5 g5 evidently this was Paikidze's idea, to pressure the e-pawn by undermining its support. White is faced with some complex choices. 24.fxg5? this was unnecessary and justifies Black's play.
24.£d1 would get the queen out of the pin first and improve on the idea. 24...gxf4? (24...¤7b6 25.fxg5 ¦xe5 26.b4 ¥xb4 27.¥xb4 ¤xb4 28.¦xa8 ¤xa8 29.¦b3±) (24...¤xf4?25.¦xf4 gxf4 26.¥d5+ ¢f8 27.£h5±) 25.¤c4 bxc4 26.£xd5++⁠−
24.b4!? has similar ideas as in the variation above. The hanging Nd5 and the open long diagonal give White some tactical possibilities and Black has to be careful.
24...¤xe5 Black now takes over the initiative. The two centralized knights in combination with the Re8 and Bc5 can make a variety of threats in this wide-open position. 25.¦f2? preserving the rook in this way just leads to more trouble for White. The engines suggest a positional exchange sacrifice. (25.£g3!?25...¤xf3 26.¥xf3µ) (25.¦f1 ¤e3µ) 25...¤e3−⁠+ it's clear by this point that for the investment of a pawn, Black's pieces are now dominating the game. This is a more positional road to victory.
25...¤d3! is even stronger, with a double attack on the queen and rook. White loses material in all lines, for example 26.¥xd5+ ¢h7 27.cxd3 ¦xe1+ 28.¦xe1 ¥xf2−⁠+
26.¤b3 protecting the c2 pawn by opening the second rank, but it would be safer to get the queen out of danger with Qb1. 26...¤xg2 Black again passes up the ...Nd3 tactic. 27.¦xg2µ27...f4 following the precept that passed pawns must be pushed, although this reduces the pressure of Black's pieces. (27...¤f3!?) 28.£c3 ¤c4
28...f3 is favored by the engines and is the logical continuation of the previous move's idea.
29.£f3 White has been doing a good job of containing Black's threats as best she can and the engines show only a slight advantage for Black at this point. 29...£f5³30.¤xc5 White logically wants to eliminate Black's strong bishop, but now the Ba5 is hanging. This is a case of where "doing something" in a position is actually inferior to waiting.
30.g6!? is the engine recommendation, a waiting move that also restricts Black's king. 30...¦ac8 31.£g4 £xg4 32.¦xg4 ¥d6³
30...£xc5 31.b4 £f5µ looking at how the position has transformed, White's bishop is now largely locked away, although may get back into the action via c7. Meanwhile Black's control of the e-file and the well-placed Nc4 are key advantages; the Ra8 can also easily get into the action. 32.¦f2 ¦e4 33.¦g1 ¦ae8 34.¥c7? an apparently logical idea, to increase pressure on the f-pawn and have the bishop do something useful, but now Black's domination of the e-file and her rooks will decide the game.
34.¢h2 would protect the h-pawn, which is vulnerable to pressure along the 3rd rank.
(34.£g4) 34...¦e3!−⁠+35.£xf4 ¦xh3+ a good example of the principle of looking for tactical exchanges, in this case the f-pawn for the h-pawn, since White could not take and also protect at the same time. Obviously the loss of the h-pawn hurts White much more than the f-pawn does Black, due to White's vulnerable king. 36.¢g2 ¤e3+ 37.£xe3 forced. 37...£g4+ here the value of the tactic of gaining a tempo is illustrated, with White's queen moving out of danger. 38.£g3 ¦xg3+ 39.¥xg3 ¦e3 40.¢h2 £h5+ a nice little tactic to pick up the g-pawn and give Black a passed pawn on the kingside. 41.¢g2 £xg5 White does not have compensation for being down material (R+B vs Q). It is instructive to see how Black's Q+R combination holds the initiative and how White is essentially helpless to do anything from this point forward. 42.¢h2 ¦e6 43.¦gg2 £h5+ 44.¢g1 £d1+ 45.¦f1 £d4+ 46.¦ff2 ¦e1+ 47.¢h2 £d1 48.¥f4 £h5+ 49.¢g3 ¦h1 50.¦h2 ¦g1+ 51.¦hg2 ¦h1 52.¦h2 £g6+ 53.¢h3 £e6+ 54.¢g3 ¦e1 55.¦hg2 £g6+ 56.¢h2 £e4 57.¥g5 £xb4 an illustration of the power of the queen to reposition herself with tempo and then pick up additional material in an endgame. 58.¥f4 (58.¥f6?58...£d6+ 59.¢h3 ¦h1+ 60.¢g4 gxf6−⁠+) 58...£e7 59.¢g3 ¦e6 60.¢h3 £d7 61.¢h2 ¦e4 62.¢g3 £f5 63.¦f3 g5 Black is now able to bring another piece into the attack. 64.¥xg5 one last attempt at setting a trap. 64...¦g4+ the tactical intermediate move that finishes things off. (64...£xg5+??65.¢h3)
Powered by Aquarium

13 August 2016

Commentary: 2016 U.S. Championship, Round 8 (Yu - Melekhina)

I selected this next commentary game both based on its excitement factor and it being a Symmetrical English (not two qualities you often see together).  A standard imbalanced position follows after Black's (FM Alisa Melekhina) 5...e5, which can lead to a slow maneuvering game.  In this case, however, White (Jennifer Yu, also the winner of the previous commentary game) chose to pursue a non-standard and perhaps somewhat risky kingside strategy starting on move 12, rather than focusing on the usual queenside and central play revolving around d5.

Melekhina reacted well and picked up the gauntlet by castling on opposite sides, but her apparently safe-looking move 14 became the root of later problems by opening the f-file.  Yu then took advantage of her opponent moving her bishops away from protecting key squares not once, but twice, then found some creative tactical resources to win.  An excellent and informative struggle between two fine players (although this was not Melekhina's tournament).

(You can also see the original US Championship round 8 reportage at ChessBase here, although it seems that the commentator was working from an incorrect scoresheet when referring to this game.)

Yu, Jennifer R (2157) - Melekhina, Alisa (2205)

Result: 1-0
Site: Saint Louis USA
Date: 2016.04.22
[...] 1.c4 c5 2.¤c3 g6 3.g3 ¥g7 4.¥g2 ¤c6 5.¤f3 e5 breaking the symmetry and establishing a central pawn presence. 6.O-O ¤ge7 the idea here is not to block the f-pawn's advance later. 7.¤e1 White's idea is to redeploy the knight via c2. This is slow, but the opening is largely about maneuver rather than attack. 7...d6 8.¤c2 from here the knight can support the b4 advance or move to e3 to increase domination of the d5 square. 8...h5!? this move scores well in the database - although see the next annotation - but is not often played. There is only one game listed in 2015 with it, for example. (8...¥e6 is the more conventional choice, along with castling.) 9.d3 both Komodo and the database indicate that the reaction h4 should be avoided. In the small sample (19) of games available, it has been played roughly half of the time and scores badly at 25 percent. That said, White appears to be OK in the line, although it allows some additional attacking ideas for Black, as in the following game:
9.h4 g5 10.hxg5 h4 11.¤e3 hxg3 12.fxg3 ¥e6 13.¤cd5 £d7 14.¤f6+ ¥xf6 15.¦xf6 ¦g8 16.¤d5 O-O-O 17.d3 ¤f5 18.¦xf5 ¥xf5 19.¤f6 £c7 20.£f1 ¥e6 21.¤xg8 ¦xg8 22.¥d2 ¤d4 23.£f2 ¥g4 24.¥d5 ¤xe2+ 25.¢g2 ¤d4 26.£xf7 £xf7 27.¥xf7 ¥f3+ 28.¢f2 ¦h8 29.¦e1 ¦h2+ 30.¢e3 ¥h1 31.¦xh1 ¦e2# 0-1 (31) Markos,J (2327)-Navara,D (2433) Pardubice 2000
9...h4 the most logical follow-up. If Black is going to advance the h-pawn, she should go all in. 10.¤e3 most played here, although the engine evaluates that first proceeding with standard play centered around the b-file is fine.
10.a3 a5 11.¦b1 a4 12.¥g5 f6 13.¥d2 h3 14.¥h1 O-O 15.¤e3 is one possibility.
10...¥e6 this gets the bishop out, but Black did not need to develop it this early, as it is doing fine on its original square for now. It does nothing to impede White's next move. 11.¤ed5 f6 this seems a little premature and commital.
11...h3!? is the engine's choice, which would avoid White's later gxh4.
12.£e1 Yu here is signaling a shift in commitment to a kingside strategy, placing her queen on the e1-h4 diagonal and preparing her next move. (12.¦b1!? would continue with queenside and central play.) 12...£d7 13.f4 O-O-O Melekhina notes the strategic shift and castles on the opposite wing, making White's threat of expansion on the kingside less urgent. 14.fxe5 fxe5?! this looks like a logical and "clean" move visually, but immediately gives White some advantage to play with, including the open f-file and the initiative. From here on out, the game gets wilder.
14...¥xd5!? with the idea of exchanging material and reducing potential White threats. 15.¤xd5 ¤xe5 16.gxh4 ¤xd5 17.¥xd5 f5
15.gxh4 ugly-looking but effective. Black has some compensation for the pawn, due to the weak doubled h-pawns, but White does an admirable job of covering the weaknesses and playing actively. 15...¥h3? it turns out that Black needs to worry more about her white-square weaknesses, especially on f7, with the absence of this bishop. White has a number of ways to take advantage of this.
15...¥xd5 is still a good idea, but leaves White in better shape compared with the above variation: 16.¤xd5 ¤xd5 17.¥xd5 ¤d4 18.¥g5²
16.¥g5
16.¦f7! and Black has some unsolvable problems related to the 7th rank and king position, for example 16...¥xg2 17.¢xg2 ¥h6 18.¥xh6 ¦xh6 19.¤xe7+ ¤xe7 20.¤d5 ¦e8 21.£a5+⁠−
16...¥xg2 17.¢xg2 ¦df8 18.h3 White evidently was concerned about ...Qg4+ here, although the engine shows that is not necessary.
18.¤xe7+ ¤xe7 19.¦xf8+ ¦xf8 20.¥xe7 £xe7 21.¤d5± and Black no longer has any real compensation for the pawn.
18...¤f5 a good consolidating move by Black. With this and the previous rook move, she has shut down threats along the f-file. White can also no longer trade down material, as in the previous variation. 19.¤e4 centralizing the knight and recognizing that the e4 square is superior to b5 for it now. 19...¥h6?! continuing with the theme of moving bishops away from controlling key squares, in this case f6. This time White takes advantage of it better. (19...¤cd4) 20.¤ef6±20...£e6 21.e4 White has regained the initiative and revived the utility of the f-file. 21...¤xh4+ a piece sacrifice based on an interesting tactical idea for Black. (21...¤fd4± is the safer choice.) 22.¥xh4 ¥g5? unfortunately for Black, this bishop move doesn't work. Yu spots the refutation, which is not obvious. Two white pieces are hanging (the Bh4 and the Nf6) and the Bh4 can't move without allowing ... Qxh3. However, White finds a creative solution by giving back the piece.
22...¥f4 is the only good continuation here, with the threat of ...g5 and ... Qxh3. 23.¦xf4 exf4 24.¤xf4 £e5 25.¤xg6 ¦hg8 26.¤xg8 ¦xg8 27.¢h1 ¦xg6 28.£f2± still works out fine for White, however.
23.¤c7!23...¢xc7 24.¤d5+ The Nf6 escapes with tempo, thanks to the sacrifice of its brother. 24...¢d7 now Black has problems with hanging pieces instead and loses at minimum the exchange. 25.¦xf8 ¥xh4 (25...¦xf8 26.¥xg5+⁠−) (25...¦xh4 26.£g3 ¦h5 27.¦af1+⁠−) 26.£f1
26.¦xh8 might be simpler, with two rooks vs. queen in a position where the rooks will dominate. 26...¥xe1 27.¦xe1+⁠−
26...¦h5
26...¦xf8 27.£xf8 ¥e7 28.£g7+⁠− and Black cannot stop the rook transfer to f1 and then f6 or f7.
27.£f7+ £xf7 28.¦xf7+ ¤e7 29.¦af1 an illustration of the importance of the open f-file in the game, along with the weak 7th rank. Black's fate is now sealed. 29...¦h8 30.¦g7 ¦e8 31.¦h7 ¥g5 32.h4 ¥f4 33.¤f6+
Powered by Aquarium

07 August 2016

Analyzing your chess games - it's now trendy!



The chess world is full of trends - popular openings (as in the above graphic, credit to Randy Olson / ChessBase), tournament formats (rapid chess is now a thing), and so on.  The chess improvement community, as a subset of this, has also been subject to trends - for example, a large part of the original blogosphere was focused on solving series of tactical problems via the Michael de la Maza method.  Although I wished them well in that regard, the "seven circles of hell" method of redoing tactical problems always seemed a little hyped and perhaps even suspect.  In the end, it almost always led to burnout.

This blog was founded in part to keep me honest and committed to chess training, with a big part of that being committed to the idea and practice of analyzing your own games.  Although it's a common practice for serious chessplayers, it's not something that has been prominently discussed or even necessarily present in the general consciousness of the chess improvement community.  Until more recently, that is.

In the past couple of months I've noticed an uptick in references, examples and useful highlights of the benefits of analyzing your games (and having them analyzed by others), more so than in any time since the 2012 post that's linked above.  Closely related to that is the idea that looking at others' thoughtful analyses in annotated games can provide unique insight into chess concepts, how an effective thinking process works, and can boost your own understanding significantly.

Here are some of those recent examples, most of which should also be good long-term resources for chess training:

1.  Jon Speelman's Agony Column

This new ChessBase column isn't about GM Speelman's games - although he's certainly racked up enough of them during his international career.  Here's what it is about, in his own words:
A game of chess is a battle both against your opponent and yourself. You have to make decisions almost every move (apart from very obvious ones including recaptures). The problem is to express yourself while avoiding blunders, though absolutely everybody from Magnus Carlsen down makes these occasionally,and to follow your desires while maintaining sufficient balance to remain within what is reasonable in the position.
This sounds both high level and abstract and obviously top players will be able, when on song, to produce games far beyond the level of club players. However, it still applies to club players, who must make decisions based on their understanding. They too must find their way through a maze of possibilities - or if you prefer, peer into the fog.
In this column I'd like to dispel some of that fog through analysing readers' games and/or answering specific questions. When working with students I normally start by asking to look at a game they are proud of and one they are not. And ideally I'd like readers to send in one of both - Ecstasy and Agony - though if you'd prefer just the former to be in print that's totally understandable. I can either analyse the whole game or focus on a particular position.
With the proliferation of strong chess engines, it's become not only easy to analyse your games with them but hard to resist their use. They provide a merciless commentary on the tactics, which we all often miss, but have only a limited connection to what is actually happening when two human beings do battle across the chess board: and often skew the viewpoint of spectators when watching games online.
I therefore propose to analyse readers' games as much as possible without an engine on. What I'm interested in, is identifying the critical decisions and "flow" of the game and neither concept is endemic to the silicon assistance we now employ. The engines are very addictive though so I imagine that I will check with one of our silicon "best enemies" for a second opinion and to error check.
I always think it's a miracle when I or anybody else plays a really good game or even avoids significant tactical mistakes. So I'm certainly not intending to be critical of anything that readers are kind and brave enough to send.
He's done a great job so far and in Agony Column #12 you can see a hybrid of original analysis and Speelman's commentary, submitted by a fellow blogger at the Hebden Bridge Chess Club.  I particularly like how Speelman focuses on practical lessons and insights for improving your game, not blindly following whatever the computer engine says is the best move (one of the pitfalls of computer analysis).


2.  Move by Move Chess Improvement (NM Julian Lin)

This recent post ("How to really improve at chess, no gimmicks, no lying, just the cold hard truth") caught my eye.  A relevant excerpt:
The most powerful thing I can leave you with is to do the following: analyze one of your games, but focus on doing the following: aim to find out as many of your mistakes as possible and note them. You might even want to suggest a better move than the one you played. Then do one more thing: try to come up with an understanding of why you made the mistake. Is this something that comes up often in your games? Then extend this further: how can you prevent this mistake in the future? How can you change your thinking to evolve as a player and not make this mistake again?

3.  Seven Questions to Ask Yourself After Each Game

This article at the Chess Improver site, by fellow chess improvement blogger Bryan Castro, helps provide some structure to your self-evaluation process.  One thing that I particularly like is that it emphasizes identifying positive examples from your play, along with mistakes; reinforcing the positives is something that I think can easily be overlooked in the process.  We need to recognize our mistakes but not punish ourselves too severely, which is just counterproductive; look towards the future and understand how to avoid falling into the same traps, that's a better approach.  On the flip side, we need to identify our best practices and learn how to repeat our successes in the future, without becoming overconfident.


4.  IM Silman's Chess.com articles with reader games

One of the more entertaining forums for showing improving players' games and analysis are the periodic articles by IM Jeremy Silman on Chess.com.  The recent "If the Board Says ATTACK, Then ATTACK!" is an excellent example of how to gain insights by looking at a game both conceptually and concretely, while having some fun.  Silman is sometimes panned for being too harsh in his comments, but I find it refreshing to see the attitude that chess is a game that should not be taken too seriously (by non-professionals at least) and that a sense of humor over our play is a good thing in the long run.


5.  dana blogs chess (NM Dana Mackenzie)

It's hard to find a better advertisement for analyzing your own games than the above-linked blog, especially the recent series on his most memorable games.  Game #5 particularly stood out to me, since the "never give up" theme I think is terribly important for practical play - many times I've despaired but still put effort into finding the best move, and have been rewarded for it.  Dana's analysis and his solicitation of commentary is something to be emulated.