23 April 2017

Annotated Game #172: Light and Dark

This next tournament game is probably the best one of mine that illustrates the idea of weak-square complexes and how one can exploit them, which for Class players is a sometimes mysterious concept.  Here we have a clear light vs. dark situation on the board, with my opponent's pawns placed on dark squares, which have the effect of restricting his dark-square bishop, but more importantly leaving open the light squares to be dominated by my pieces.  This situation was evident by move 17 and demonstrated by the effectiveness of my centralized knight (these weaknesses aren't just for bishops to exploit).

Although even well into the middlegame my opponent either had a small advantage or at least equality, despite his weaknesses, I felt comfortable playing the position and was able to identify good ideas for making progress.  I also correctly identified many of the important positional ideas (including strong and weak squares) and found dynamic moves like the (temporary) pawn sacrifice idea on move 19.  The turning point of the game was the sequence that began on move 23, which involved my finding some unexpected intermediate moves and placed my opponent under significant pressure for the first time; this led him to err with 26...Rf7?!  Although not a losing move in itself, I was subsequently able to maintain the initiative for the rest of the game and win an interesting, dynamic minor piece endgame.  Even considering some weaker play in the opening and early middlegame, I feel this game serves to highlight some of the signposts of progress that I have been making in strengthening my game.

ChessAdmin - Class C

Result: 1-0

[...] 1.c4 c5 2.¤f3 ¤f6 3.¤c3 ¤c6 Symmetrical Four Knights variation. 4.g3 e6 indicating my opponent is going to take a relatively cautious approach to the opening, at least early on. 5.¥g2 ¥e7 6.O-O O-O 7.d3 I thought for a while here on the best approach to take. The main alternative is d4 (and by far most often played), while b3 is also a possibility, with a double fianchetto position. 7...d5 8.cxd5 played to reduce Black's central pawn presence and make pressure from the Bg2 down the long diagonal more meaningful. 8...exd5 9.¥g5 in the English it's sometimes hard to know what to do with the dark-square bishop. I didn't see a future for it on the queenside and on f4 it could be harrassed by ...Nh5, so I picked g5. When playing this move, White has to be prepared to exchange it for the Nf6, so evaluating the effects of that piece trade is important. 9...d4 10.¥xf6 I had foreseen Black's last and considered that the resulting position was good for me, with the centralized knight vs. a locked-in Bf6. 10...¥xf6 11.¤e4 ¥e7³ objectively speaking, Black is a little better here. He has the two bishops and a small space advantage. That said, the position is relatively easy for me to play, with some clear ideas for making progress. 12.¦c1 done in the expectation of provoking Black's next move, which opens up the long diagonal for the Bg2. 12...b6 13.¤ed2 while the knight looked good in e4, it had no squares other than d2 open and could therefore be threatened by ...f5 (which Black plays shortly). Redeploying it gives it an equally good square and improves my piece coordination, is what I thought. 13...¦b8 my opponent appears concerned about the rook on the open long diagonal, so moves it. 14.a4 with the idea of preserving the c4 square for the knight, by restricting the ...b5 advance.
14.¤c4!? can in fact be played immediately to good effect. For example 14...b5?!15.¤fe5 taking advantage of the hanging Nc6 and the R+Q fork on the square. 15...¤xe5 16.¤xe5 ¥b7 17.¤c6 ¥xc6 18.¥xc6
14.a3!? would take away b4 from the Nc6, which is helpful in several variations.
14...f5 I felt that this move now was too loosening for Black's kingside. The knight is prevented from returning to e4, and the pawn then continues to f4 to try to weaken White's kingside pawn shield, but that does not appear to be sufficient reason for Black to weaken the a2-f8 diagonal and the light squares in general. (14...¥g4!?) 15.¤c4 f4 16.¤fe5 now my other knight gets into the action and releases the Bg2's power. 16...¤xe5?!
16...¤b4 is a significantly better choice, giving the knight an excellent outpost on b4.
17.¤xe5 the Ne5 now eyes the weak c6 square. I had thought that if Black exchanged the light-squared bishop for the knight (for example after trying ...Bb7, which I was thinking of following by playing the Nc6 fork) then that would leave me with a significantly positive imbalance between the remaining minor pieces (my light-square vs. Black's dark-square bishop). 17...£d6 the best option for Black, removing the queen from the fork and centralizing it.
17...¥b7? would in fact have been a significant blunder, but for other reasons: 18.£b3+!18...¢h8 19.¤f7+ ¦xf7 20.£xf7+⁠−
18.¤c6 now an exchange is not forced, but the Nc6 still causes Black difficulties. 18...¦b7 19.b4 I felt that this was necessary to energize my position and use my pieces most effectively, particularly the Rc1 (which is not otherwise playing). Komodo agrees. 19...¦c7 Black passes up the (temporary) pawn sacrifice.
19...cxb4?!20.¦c4 I had spotted this idea, targeting Black's weak pawns on the 4th rank. 20...fxg3 21.hxg3 ¥f6 22.¤xb4² and White has a slight advantage due to better piece activity.
20.bxc5 bxc5 now the position is still equal, but I have nice pressure against the c-pawn and comfortable play on the queenside. 21.¤a5 fxg3 it's often difficult to decide which pawn recapture to make in this situation. I decided that the open f-file would benefit me more than Black, who does not have his rooks connected on the back rank, and that the resulting pawn formation would be a bit more solid, not offering Black any prospects of an attack down the h-file. The drawback of the text move, as I immediately realized, was that I lose control of the e3 square, so I had to watch that carefully. (The engine gives an assessment of equality to both pawn recaptures, incidentally). 22.fxg3 ¥g5 here my opponent evidently did not consider the intermediate moves I could play in response to the threat against the Rc1, which end up giving me the initiative. I felt this justified my decision to open the f-file. 23.£b3+ ¥e6 24.¦xf8+ forcing the recapture with the king, as the Be6 otherwise would be left undefended (deflection tactic against the Qd6). 24...¢xf8 25.£b8+ placing the queen on the back rank and pinning the Rc7. 25...£d8 my opponent thought for some time here and found the best reply. (25...¥c8?26.¦f1+ ¥f6 27.¥b7±) 26.¦f1+ ¦f7?! this allows me to win the a-pawn. (26...¢g8) 27.¦xf7+ ¥xf7 28.£xa7 Black has some compensation in the form of the two bishops heading into the endgame, so I only have a small advantage. When calculating the pawn capture, I also needed to be very careful about evaluating Black's next move, which is very trappy. 28...¥e3+ I had thought a good deal about this position prior to initiating the previous sequence, so was prepared. 29.¢h1 better than f1, although it gives the king no squares. Either way is fine for White, however, according to the engine. 29...£e7 30.£xe7+ at the time I was happy to enter the endgame with the advantage of a passed a-pawn, although I figured that combating the two bishops could make it a hard slog. At least with the queens off, I did not have to worry about mating threats. The engine evaluates keeping the queens on as significantly better for White, since the queen can more effectively shepherd the a-pawn forward. However, queen endgames are also very complex, so I think I made a decent practical decision in trading material. 30...¢xe7 31.¤c6+ ¢d6 32.a5 I had calculated this out prior to the knight move, as Black does not have sufficient time to capture the knight before the pawn queens. The idea is to block the Bf7 from getting over to defend, while the Be3 is also out of the action on the queenside. Now Black has to find an "only move" at this point to defend. 32...¥e8? a reasonable try, but not sufficient.
32...c4 is the only move that preserves equality for Black and is not necessarily easy to find (for humans). 33.dxc4
33.a6 c3 34.¤b4 ¢c5 35.a7 ¢xb4 36.a8=£ c2 as White cannot keep the pawn from queening. For example 37.£b8+ (37.£f8+ ¢c3 38.£xf7 c1=£+) 37...¢c3 38.£c7+ ¢d2
33...¥xc4 34.¥f3
33.a6 ¢c7 34.a7 by this point I knew that I would have to give up the a-pawn for Black's d-pawn, but was not sure when would be best. After some thought, I figured that it would be better to have Black's king a little further away. The engine disagrees.
34.¤xd4 makes the knight a much more threatening piece and introduces some tactical ideas. 34...¢b6 35.¤f5 ¥d4 I had seen this far and didn't consider it any better than the game continuation, but after 36.¤xd4 cxd4 37.¥b7+⁠− White has the easy winning strategy of activating his king and clearing away Black's d-pawn.
34...¢b7 35.¤xd4+ transforming the advantage of the passed a-pawn, by taking advantage of the discovered check. An example of a tactical trade, in this case the a-pawn for the d-pawn. 35...¢xa7 36.¤f5 ¥d4 I felt at the time that this was a losing move, giving away the benefits of the two bishops and clarifying my advantage. The engine is less harsh in its evaluation, not seeing the evaluation as any worse, although from a practical standpoint it made my mental task easier. 37.¤xd4 cxd4 38.¥e4 I thought for a while about this or Bd5, they are both good centralization moves. Since it provokes Black's next (unforced) error, I'm glad I went with it. 38...g6?+⁠− Now Black has made his kingside pawns vulnerable to penetration by my king and/or bishop. This was the actual losing move. (38...h6±) 39.¢g2 ¢b6 40.¢f3 the plan is very obvious for White here, to threaten the d4-pawn and tie Black's king to its defense, then go after the kingside pawns. 40...¢c5 41.¢f4 h6 42.h4 guarding g5 against a supported Black pawn advance 42...¥f7 43.¢e5 g5 44.hxg5 hxg5 45.g4 played as a prophylactic move, to keep Black's bishop from h5. 45...¥b3 with the idea of moving to d1. 46.¥f3 the safest route to victory. Now that Black can only move the king or bishop, eventually he will be put in zugzwang; my bishop protects both e2 and g4 and the king has full freedom. 46...¥d1 47.¢e4 I did this rather than move directly to e5 to have Black essentially lose a tempo with the bishop, although it's not truly necessary. 47...¥b3 48.¢f5 ¢b4 49.¢xg5 ¢c3 50.¢f4
Powered by Aquarium

20 April 2017

How do you know you are becoming a stronger chess player?

Chess strength is a funny thing.  It's hard to define precisely, so we rely on rating systems (primarily Elo-based) as a proxy statistic for it.  Yet clearly there must be something substantive behind the explanation for why players have particular levels of strength, so we can talk about things like "master" and "expert" versus "Class B" and "Class C" players.  (I am currently a Class B player, per the USCF scale above.)

There are some helpful, if not necessarily definitive, attempts at providing "roadmaps" or the like to chess knowledge at each level.  Here is one posted at Chesstempo.  You can also infer what knowledge is considered standard to have at beginner, intermediate and expert levels from resources such as the Chess.com study plans.  Another approach is defining specific characteristics and skills that set the higher levels apart from others, as done by GM Andy Soltis in What it Takes to Become a Chess Master.  Soltis' book I think gets at some crucial concepts, including how masters are able to much better understand (and apply) things like compensation for sacrificed material in the absence of concrete winning tactics.  This includes "positional" sacrifices of the exchange or a pawn or two, where there is no combination on the board, yet the master understands that in the long term, their chances are better (or at least as good) as before the sacrifice.

All of the above approaches to explaining playing strength have their uses, but one of the conundrums of chess strength is that it often does not reflect the extent of a player's knowledge, at least on a one-to-one basis.  Some things are directly correlated with your ability to win; knowing basic mates and mating patterns are fundamental to success.  Others are helpful, but not 100% required (for example how to play the Philidor and Lucena ending positions).  Finally, some pieces of chess knowledge have a very low (yet non-zero) percentage chance of ever being directly helpful (such as knowing the K+B+N v K mate).  Soltis' work outlining particular crucial areas of skill I think comes closer to a cognitive approach, rather than simply giving a list of "must knows", but of course this approach is necessarily subjective, rather than rigorously scientific.

Similar to Soltis' approach, but on a more practical level for the improving Class player, I'd like to document some phenomena that appear as concrete indicators of improving chess strength.  It's not an exhaustive list, but I think it is useful to share.  I make no claim to having originated any of these ideas, but in reading widely and through analyzing my own games I have been struck by how important some of these are when they appear, and I have personally experienced all of them at one time or another.  Naturally there are many more improvements yet to come...

I've arranged the below phenomena in what seems to me to be a logical progression from ones based on more concrete/tactical/conscious considerations, to those that are more cognitive/strategic/unconscious in nature.  Again, these are meant to be taken as positive signposts you may see on the road to mastery, whenever you observe them in your games.  I think it's important for improving players to explicitly recognize the positives in their game, along with the many errors and negatives, otherwise chess can start becoming toxic and demotivational.  We (including professional players) will always fall short of perfection, but that's to be expected, and therefore not overly lamented.
  • I hesitate to start with this one because of its obviousness, but blunders (making less of them) really is fundamental.  If you keep having the same frequency of game-ending blunders (i.e. losing a piece or overlooking mating threats) over time, then by definition you will make little progress.  Having at least a basic blundercheck thinking process is necessary; over time, it hopefully will become more and more unconscious (but still necessary).  One practical observation I've read before, and tend to agree with based on personal observation, is that by Class B level the majority of games are no longer decided by blunders, and by Class A only a very few are.  In other words, you more often have to beat your opponent, rather than just waiting for them to beat themselves.
  • When analyzing your games, you have an increasing number of "!" annotations.  This is related to the above indicator on blunders (the "?" moves), but it is not simply the elimination of errors, but rather a demonstrable ability to find the key (sometimes only) move that gives you a breakthrough in the position.  You can always give yourself a "!" if you're feeling generous, but it's worth more when coming from an unbiased annotator, which in many cases will mean your computer engine.  Although there are some pitfalls of computer analysis, engines can almost always correctly identify the standout strong moves ("!") as well as the blunders.
  • Something a little less obvious, but a definite sign of improvement in strength and sophistication, is spotting and deliberately using intermediate moves when conceiving and calculating move sequences.  One thing I have seen repeatedly in my game analyses is that I (or my opponent) often may have a good idea, but it is not executed to the full extent of its power, whether it may be a strategic pawn break or a particular tactic that becomes much more effective with the insertion of an intermediate move.  I think that learning how to keep searching for more effective moves after finding a good idea is the key, including taking a particular idea and then looking creatively at the different possibilities of how and when it could be most effectively played.  Annotated Game #171 has some good examples of effective intermediate moves.
  • As part of your evaluation of the position, you start naturally "thinking in squares" rather than just about the pieces and threats to them.  Weak squares in your opponent's camp become magnets for tactical and strategic ideas; one common example is if f2/f7 become underprotected next to a castled king.  Excellent squares should also suggest themselves (see below) for your pieces - for example leading to the repositioning of a knight, perhaps even using a tactic, to get it to a dominating outpost on the opponent's side of the board.  Defending your own weak squares and realizing when they might be created (after a pawn push, when a key piece moves away) is also very important, the more so the higher level you get.  (See this example commentary game.)  Going back to the thinking process, Botvinnik suggested thinking on your opponent's time about positional considerations rather than calculating variations - squares are a big part of this process.
  • Allied to the automatic recognition of "thinking in squares" is the development of a more automatic/unconscious visualization skill, in other words the ability to picture future board positions in your mind.  I have found that this grows naturally in the context of a consistently applied training/study plan, as long as you sometimes move the pieces in your head rather than always on the board.  One good example of doing this is when variations are given in an annotated game and you visualize them, rather than playing them out physically.  This can also be practiced in reading chess books without a board, starting with ones that have plentiful diagrams (such as Logical Chess: Move by Move).  At a basic level, you should not need to have the board coordinates (A-H, 1-8) printed on the board for you to immediately identify a square.  The next stage is being able to mentally picture (away from a board) the color of a particular square (d4 is black, b7 is white, etc.) quickly; if you can't do this, then your internal board sight is not functioning on an unconscious level and the need to process this consciously will slow you down.  Finally, being able to play a game "blindfold" (without sight of the board, whether or not you actually are wearing a blindfold) is more than a parlor trick, it is a sign that you have strong visualization skills that will strengthen your calculation and evaluation of variations.  Blindfold chess skill is again not a one-to-one correspondence with overall chess skill - not everyone can be Timur Gareev and he is not at the super-GM level (yet) - but it is a strong indicator of your general strength level.
  • While the above are largely conscious mental efforts related to the process of calculation - with visualization being, I would argue, also a partially unconscious function - one unconscious-origin phenomenon is when a move "suggests itself" without you doing any calculation at all.  At certain level this phenomenon is related to what we call "natural moves" - which to a lower strength player do not necessarily appear natural at all.  Geometrically these can typically be defined as when an individual piece can be moved to a square where it will have maximum influence across the board, particularly into the opponent's camp.  More sophisticated versions of this idea occur when we have a mental library of successful positional patterns built up, so we make an instant comparison to what we have seen work before.  To quote Magnus Carlsen: "Of course, analysis can sometimes give more accurate results than intuition but usually it’s just a lot of work. I normally do what my intuition tells me to do. Most of the time spent thinking is just to double-check." 
  • Finally is an observation that I once read by a GM (the source regrettably does not occur to me right now, maybe Yermolinsky?) who noted that when you gain strength you do not in fact feel any stronger yourself; rather, your opponents start seeming weaker (in the same Elo range as you).  This is naturally a largely unconscious impression, but I believe it's a valid one and probably one of the most powerful indicators that you have in fact shifted significantly past a previous milestone in chess strength.

17 April 2017

Annotated Game #171: A matter of technique

This next tournament game, which started my win streak, recalls a number of previous themes from my analyzed games.  The problem of going ahead and playing an "obvious move" without checking tactics is present - for both White and Black, in this case.  Strategic themes are identified on the kingside (White's Bh6 threat), the center (the e5 square and various tactics, as well as an eventual pawn break), and the queenside (missed opportunities to exploit White's weaknesses there).  Tactical themes include hanging pieces (White's Bf4), deflection, and intermediate moves (exploited by me to very good effect).

Overall I chose as the primary theme the sometimes cliche' comment in a game annotation, that the win from a certain point is "a matter of technique".  As long as one's technique is adequate, that is.  One of the pitfalls of computer analysis is to always look for the best move, even when you have already identified a winning move, which from a practical perspective should (by definition) always be enough.  This is an especially important consideration after a long struggle when you may be tired and your board sight and calculating ability (unlike a computer engine's) is not at 100%.  In this game I won by going into an advantageous endgame which I felt confident of winning, but should have earlier seen a deflection tactic (25...h5!) which would have wrapped things up much sooner.  After that, it really is around move 36 that I knew I would win (rather than just thinking it...there is a difference), but still almost faltered due to a board sight issue on the long diagonal.  The final blow doesn't come until I transition into a won king and pawn endgame, which takes two tries but finally succeeds.

This game was a struggle and I missed some good ideas, but being able to make some strong, confident decisions along the way to victory helped keep up the positive momentum in subsequent games.

Class C - ChessAdmin

Result: 0-1

[...] 1.e4 c6 2.d4 d5 3.exd5 cxd5 4.¤f3 this is not a recognized variation in the Caro-Kann, although of course it's not a terrible move in itself. It doesn't challenge Black in any way, unlike the Panov (c4) or the Exchange (Bd3) variations. 4...¤c6 5.¥f4 ¥g4 Black is able to deploy the bishop to a very effective square before playing ...e6. I would say that Black has already equalized. 6.¥e2 ¤f6 7.h3 ¥h5 8.c3 e6 9.¤bd2 ¥d6 White's Bf4 is an excellently placed piece, so deserves to be challenged. 10.¤e5 ¥xe2 11.£xe2 O-O much better to castle before trying to play further in the center. 12.¤df3 White has established a strong-looking and supported outpost on e5. The flip side is that I get a similar one on e4. 12...¤e4 also an excellent outpost for my knight on the e-file and one that cannot be challenged without inflicting positional damage on White. Importantly, it also helps guard the g5 square. 13.¤xc6 I am happy with this exchange, which gets rid of a strong centralized knight in exchange for my Nc6, which is doing relatively little at the moment. 13...bxc6 14.¤e5?! again a knight occupies e5, but with fewer pieces on the board White has less in the way of potential threats associated with it. White also gives himself a tactical problem, in that the Bf4 is hanging if the Ne5 moves. This is not immediately critical, but soon becomes a consideration. (14.¥xd6 ¤xd6 15.O-O) 14...£c7 now ...f6 is a threat. 15.£g4 protecting the Bf4 (once). I now have to watch out for the Bh6 idea, but it doesn't work tactically under the current circumstances. (15.¥h2!?) 15...c5µ a classic Caro-Kann pawn lever.
15...f6 16.¤d3 saves the bishop, but doesn't prevent the pawn advance. 16...e5 17.dxe5 fxe5 18.¥e3 c5µ this variation is a better version of the ...c5 lever idea, with a large rolling pawn center for Black and White's king looking unsafe.
15...¦ab8 is preferred by Komodo, taking more direct advantage of the Ne4 placement and White's queenside weaknesses.
16.f3? creating a major dark square weakness for White. This is an "obvious move" to kick the knight, but...
16.¥h6?16...¥xe5 17.dxe5 £xe5µ is what I saw during the game.
(16.O-O ¦ab8µ) 16...cxd4! I spotted this tactic/intermediate move, the point being that the Ne5 is (temporarily) no longer sufficiently protected. 17.cxd4 (17.fxe4 ¥xe5 18.¥xe5 £xe5 19.cxd4 £xd4−⁠+) 17...¥b4+ it was clear here that activating a piece along the d1-a5 diagonal would be best, but it was unclear to me which one. In the end I chose the "safer" alternative. Visually it also looked good, with my queen maybe able to penetrate on c2.
17...£a5+ I also considered for a while, and Komodo considers it much superior. Black is able to break through in the different variations. 18.¢f1
18.¢e2 ¥xe5 19.fxe4 f5 20.exf5 ¥xf4 21.£xf4 ¦xf5 and White's open king will be hunted down.
18...£b5+ 19.¢g1 £xb2 20.¢h2 ¤f2−⁠+
18.¢f1 here I could not come up with much of a plan, other than exchanging pieces. Simply retreating the knight would have been fine. 18...¤d2+?! (18...¤f6 19.£h4 ¦ac8µ) 19.¢g1 I was instead expecting a piece trade on d2, which would have left me with an unopposed, outstanding dark-square bishop. The text move attempts to hide the White king on the h-file, but takes additional time and leaves the king still exposed on a dark square.
19.¥xd2? would in fact be much worse for White. 19...¥xd2−⁠+ and now White's queenside is very weak, notably the b-pawn, and not defendable.
19...¦fe8?! with the idea of supporting an eventual f6 and e5 advance. Unfortunately the rook placement on e8 is an error, as pointed out by the engine.
19...¦fc8 instead allows the knight to be extricated via c4, while protecting against White's Bh6 idea. 20.¥h6 f5 21.£g3 ¤c4
19...¤c4 is the safe alternative, not leaving the knight out on a ledge.
20.a3? another "obvious move" that fails, this time putting White back into losing territory for the rest of the game.
20.¤d3 is the way to accomplish White's idea, attacking the Bb4 while gaining a tempo by forcing Black's queen to move. 20...£c4 21.¤xb4 £xd4+ 22.¢h2 £xb4 23.b3± and now the knight has no escape.
20...¤b3 another intermediate move tactic, this time that saves the knight. I should have looked at this idea sooner, however, as White's move was actually a surprise. 21.¦d1 ¥d6 now I can breathe easier. White has no good responses to the multiple threats down the c-file and on the b8-h2 diagonal. 22.¢h2
22.¤d3!?22...¦ac8 23.¥e5 ¥xe5 24.dxe5 £b6+ 25.¢h2 ¦c2µ
22...f6 23.¤d3 ¤xd4 winning the d4 pawn is the the first concrete benefit I have to show for my positional advantages. I spotted this idea when first considering the plan of kicking the White knight with ..f6. 24.¥xd6 £xd6+ 25.¤f4? unnecessarily pinning a piece. (25.f4) 25...¤e2 not a very imaginative move, but there are multiple paths to victory and it utilizes the pin to good effect. I can win now by trading down into an endgame.
25...h5! wins the Nf4 with a deflection tactic. I should have looked harder for other ways to exploit the pin.
26.g3 ¤xf4 27.gxf4 evidently hoping to try for some counterplay along the half-open g-file. 27...¦ac8−⁠+ finally I get the rook developed. 28.¦c1 ¦c7 played to keep an eye on g7 while preparing to double on the c-file. 29.¦he1 ¦ce7 here I decided that the c-file would be of no real use to White, while opening up and dominating the e-file would be to my benefit.
29...¦xc1 would be simpler and follow the rule of exchanging down rooks when in a winning position. White indeed can do nothing with the c-file, so logically this is superior than keeping the double rooks on.
30.¢h1 e5 31.fxe5 ¦xe5 32.¦xe5 at this point there's nothing better for White. Ceding the e-file to the doubled rooks would be worse. 32...£xe5 the piece trades have benefited my strategy of clearing the way for the passed d-pawn and dominating the e-file. The queen is very well placed on e5 for central mobility. At this point there is little doubt that I have a won game, but still have to win it. 33.b4 £e3 this gives White too much latitude. (33...f5) 34.¦g1
34.¦c7!? would be annoying. 34...g5 35.¦c8 ¢g7 36.¦xe8 £xe8µ
34...g6 here it was nice to play a correct "obvious move". 35.¦f1 f5 now I get the idea of blocking the queen's penetration on the h3-c8 diagonal. 36.£g2 £xa3 at this point it really is "a matter of technique" for the win, although White still has a fair amount of fight to put up. 37.f4 ¢g7 38.¦b1 (38.£xd5 £xh3+) 38...¦e3 I originally thought this would be decisive, but White finds a good resource that I missed. 39.£b2+ ¦c3 not necessarily my first choice, but I touched the rook to capture the h3 pawn before I understood that the queen move was a check. My board sight failed me on the long diagonal here, in part probably due to tiredness. Luckily this changes nothing about the outcome. 40.£xa3 ¦xa3 41.¢g2 now my rook is more awkwardly placed and White has managed to protect the h-pawn, but the d-pawn is still dominant. 41...¦a6 played to regroup the rook behind the d-pawn. 42.b5 ¦d6 43.¦a1 ¦d7 this is a safe place for the rook, from which I felt I could continue playing without worrying much about calculation. 44.¢f3 d4 passed pawns must be pushed! 45.¢e2 ¢f7 actually sending the king to the wrong side. The Rd7 can cover everything while Black's king romps on the kingside. (45...¢h6) 46.¦a6 ¢e8 47.b6 likely played out of a common Class player desire for simplication, which often is the wrong instinct. 47...axb6 48.¦xb6 d3+ 49.¢d2 ¢f7 here I calculated that the resulting K+P endgame would be winning, so did not have a problem jettisoning the d-pawn. 50.¦b3 h6 51.¦xd3 at this point Komodo shows a mate in 26 for Black. (51.¦b5−⁠+) 51...¦xd3+ 52.¢xd3 g5 53.fxg5 hxg5 54.¢e3 ¢g6 55.¢f3 ¢h5 again I head my king the less effective way. (55...¢f6) 56.¢g3 f4+ 57.¢f3 ¢h4 58.¢g2 here I had some problems calculating the win due to fatigue, so repeated the position. 58...¢h5 59.¢f3 ¢h4 60.¢g2 f3+ the key move which I found, but had to make sure did not lead to stalemate. 61.¢h2 g4 62.hxg4 ¢xg4 63.¢h1 ¢h3 64.¢g1 ¢g3 having gained the opposition the final time, the win is now evident. 65.¢h1 ¢f2 66.¢h2 ¢e1 67.¢g1 f2+
Powered by Aquarium

09 April 2017

Annotated Game #170: It's the little moves that matter

The next tournament I played in featured some rather big swings in momentum.  After my usual "warm-up" game to get the rust out after an extended period of not playing (the game below), I won the next three games and then lost the last two.  This result really pointed out the importance of energy management and psychological factors in improving my tournament performances.

The main theme for this game is the importance of "little moves" - ones that don't look like they do much (if anything) on the board, but can have big strategic or tactical impact.  The first one was a strategic one on move 9, while the other two were crucial tactics, including what should have been a game-winning combination.  Another theme is that of the swindle, as after the combination fail I have to fight hard and cunningly to get to a (largely undeserved) draw.

ChessAdmin - Class C

Result: 1/2-1/2

[...] 1.c4 c6 2.¤f3 e6 an unusual and rather inflexible way of heading for a Semi-Slav type setup. 3.g3 d5 4.¥g2 ¤f6 5.O-O ¤bd7 6.b3 ¥e7 I've seen more games with the ...Bd6 but both are equal choices for the bishop. 7.¥b2 the bishop serves to both restrain ...d4 and to pressure the beautiful long a1-h8 diagonal. 7...O-O 8.d3 here I thought for a while and eventually opted for the database favorite. This retains flexibility for White in the center - keeping the Bb2 unblocked - while giving the Nb1 the option of developing to d2. 8...dxc4 this is rarely played, as it gives up Black's best, well-supported central pawn (the whole reason for the Semi-Slav pawn formation in the first place). 9.dxc4 I opted for the central pawn recapture to maintain the pawn structure symmetry, thinking that Black might somehow be able to do something with an open b-file later on. This isn't a bad move, but most players (and Komodo) prefer recapturing towards the center with the b-pawn. Among other things that leaves the d3 pawn controlling e4. A "little move" that matters. 9...£b6 done to clear the d8 square for a rook, but this is less effective a maneuver without the rooks connected on the back rank. The locked-in Bc8 is doing nothing for Black except get in the way. 10.£c2 ¦d8 11.¤c3 developing my last minor piece and connecting the rooks, which will allow me to challenge on the d-file. 11...¤f8?! I was a little puzzled by this knight maneuver, which seems a bit of a time-waster. 12.¦ad1² I now have a pleasant game and am ahead in development. 12...¤g6 13.¦xd8+ £xd8 14.¦d1 although I have the d-file, which is nice, the piece exchange did help ease Black's cramped game a bit. 14...£c7 although the idea of seizing the d-file was easy to find, now I thought for a while about a suitable plan. The Nc3 isn't doing a whole lot and has limited prospects, so I thought exchanging it for the Nf6 and opening the long diagonal again would be a good idea. 15.¤e4 the (rather obvious) threat with this move is exchanging twice on f6 and doubling Black's f-pawns. Komodo prefers a more sophisticated version of the idea of occupying e4 with a knight, first playing Nd2. 15...¤d7?! another surprising retreat, as I had expected an exchange on e4 (indeed best per the engine). I understood the importance of the d-file, but failed to find the best idea, which would be to use the un-exchanged Ne4 to move to an even better outpost on d6. 16.£d2 the idea was to increase pressure on the d-file, which could further hinder the development of the Bc8, and also to support a possible knight hop to d6. (16.c5! would prepare Nd6 and throttle Black.) 16...¤ge5 17.¤xe5! here I had another long think and came up with a nice-looking combination that doesn't quite work as played. This move however is indeed winning. 17...¤xe5 18.¥xe5 £xe5 19.£d8+!+⁠− a nice-looking back-rank tactic. 19...¥f8 however, now I wasn't able to find the correct forcing path, due to a lack of creative thinking. As played in the game, the try to win a piece ends up leaving me down two pawns. 20.¤d6?! this actually can work out OK for White, but there is an immediate win. I was too fixated on trying to win the Bc8 and did not see the potential for also threatening f7.
20.£e8! a "little move" with the threat of Rd8 to follow, doubling up on the 8th rank and winning.
20...£xe2 21.¤xc8? this is the right idea (or one of them) in the position, but wrong sequencing. I understood by this point that I would be down multiple pawns afterwards, but didn't see anything better.
21.¦f1 another quiet, little move that is the best. 21...£e5 (21...£xa2?22.£e8) 22.¤xc8 £b8 23.¤e7+ ¢h8 24.£d7±
21...¦xc8 this is the problem, the rook sacrifices itself with a deflection tactic and the Rd1 is now hanging. I was not patient and analytic enough to see that simply removing the Rd1 from being en prise (as given in the above variation) would be enough to make the tactic succeed. 22.£xc8 £xd1+µ23.¥f1 £a1 24.£xb7 (24.a4 again, patience would have been a virtue here.) 24...£xa2 25.£xc6?! this gives Black an outside passed pawn on the a-file. 25...£xb3 now it's looking pretty ugly for me, although not hopeless, given the complexity of queen endings. I try to avoid trading queens at all costs, keep up threats to the a-pawn, and freeze the Black bishop on the back rank, while maintaining my c-pawn and advancing it. 26.£a8 £b6 27.¥d3 with the threat of Bxh7, getting a pawn back by a deflection tactic against the Bf8. 27...£b3?28.c5? now I thought here for a while and chose to go for another risky, poorly-calculated "combination", thinking that I could queen the c-pawn if given a head start.
28.¥xh7+! the original idea is still the best. 28...¢xh7 29.£xf8 I mis-evaluted this position, which the engine gives as completely equal. Black's king is too exposed and the f-pawn is weak.
28...£xd3−⁠+29.c6 £d1+ 30.¢g2 £d5+ 31.f3 g6 my original idea hinged on being able to queen on c8 after sacrificing the queen for the Bf8, but now I see this will not be possible. 32.£c8 the only hope, I felt, of still trying to queen the pawn. Now we are in swindle territory. 32...¢g7 33.c7 ¥d6 34.£d8 temporarily staving off disaster by pinning the bishop. 34...£d2+ 35.¢h3 £h6+ 36.¢g2 ¥xc7 Black has nothing better. We are now back to the two-pawn material deficit situation. 37.£xc7 while Black is still comfortably winning, there's a big difference both psychologically and materially with the bishop off the board. 37...£d2+ 38.¢h3 £d4 39.f4 with the idea of freezing the e-pawn advance. 39...h5 40.£a5 with the idea of keeping an eye on the a-pawn and on e5. 40...¢g8 41.£c7? now White's idea should be to keep the queen on the a-file to prevent Black's queen from doing the same and advancing the pawn. 41...¢g7 42.£a5 £d1 43.£e5+ ¢h7 44.£c7 making sure to keep the queen as active as possible, threatening the f-pawn as well as the a-pawn. 44...£g4+ 45.¢g2 £e2+ 46.¢h3 at this point, with less time on her clock, my opponent decides to stop trying to convert the queen ending and offered a draw, which I accepted.
Powered by Aquarium