Training quote of the day #21: Emmanuel Lasker

The most intelligent inspection of any number of fine paintings will not make the observer a painter, nor will listening to a number of operas make the hearer a musician, but good judges of music and painting may be so formed.  Chess differs from these.  The intelligent perusal of fine games cannot fail to make the reader a better player and a better judge of the play of others.

-- Emanuel Lasker, World Champion 1894-1921

(Cited in The Art of Planning in Chess: Move by Move by GM Neil McDonald) 

Commentary: Tata Steel 2019, Round 1 (Van Foreest - Anand)

As mentioned in my previous game post, I'll again start working in commentary on master-level games to my rotation of training analysis. I think it's useful to alternate that with analysis of your own games, at least to some extent.  Different lessons can be learned, in part because the overall quality of play tends to be (much) higher.  I've found that with master games, often it's analyzing why they didn't play a particular line that is illuminating, in addition to critical turning points in games.

The below game is from round 1 of this year's (still currently ongoing) Tata Steel tournament. I selected it because Anand adopts an aggressive setup as Black in the Caro-Kann Exchange Variation that is deliberately designed to cause interesting dynamic and structural imbalances.  Essentially Black wants to exchange bishops on f5 and thereby open the g-file, while castling queenside.  I find Black's typical ideas to be more straightforward and understandable, although not necessarily easy to execute.  The dark-square weakness and White's space on the queenside serve to counterbalance things and engines give White a small plus out of the opening.  However, in the middlegame White runs out of productive ideas, then critically weakens his own dark squares, after which Black finds a threat using his advanced doubled f-pawn that White cannot handle.  Anand's finish is quite strong and worthy of remembering.

[Event "81st Tata Steel Masters 2019"] [Site "Wijk aan Zee"] [Date "2019.01.12"] [Round "1"] [White "Van Foreest, Jorden"] [Black "Anand, Viswanathan"] [Result "0-1"] [ECO "B13"] [WhiteElo "2612"] [BlackElo "2773"] [Annotator "ChessAdmin/Komodo 11.2"] [PlyCount "56"] [EventDate "2019.??.??"] 1. e4 c6 2. d4 d5 3. exd5 cxd5 4. Bd3 {the Exchange Variation has been making a comeback at high levels.} Nf6 (4... Nc6 {is the main line, forcing White to play c3 in response.}) 5. c3 {White plays it anyway, which eventually leads to a transposition back to the main line, at least for Black.} Qc7 {placing the queen on c7 this early used to be a quirky sideline. It's now more mainstream, I'm sure in large part due to its practical success. The main idea is straightforward, to take over the b8-h2 diagonal and prevent White from playing Bf4. As the queen wants to go here in most lines anyway, playing it early and preventing White's bishop from seizing the diagonal makes a lot of sense.} 6. h3 {this is seldom played. The evident idea is to take away the use of the g4 square from Black, for either the knight or bishop.} g6 {a new move in this position according to the database, but not a new idea in the Exchange Variation. Black looks like he is fianchettoing his bishop, but in fact the main idea is to play play ...Bf5 and then open the g-file.} 7. Nf3 Bf5 8. Ne5 { White holds off on the bishop exchange and places his knight on the e5 outpost, ready to take back on d3. This also potentially opens up f3 for his queen.} Nc6 {although this position isn't in the database, for Black it's a standard setup in the Exchange Variation. In this game, White is further behind in development than normal, with only two pieces out to Black's four.} 9. Bf4 { now White has essentially caught up, as Black will have to move the queen again either immediately or after an exchange on e5.} Qb6 (9... Nxe5 {it might seem under general principles that exchanging off White's central knight is a good proposition. However, leaving the a4-e8 diagonal open results in a small plus for White. For example, this continuation leaves Black's king in the center:} 10. Bxf5 Nd3+ 11. Bxd3 Qxf4 12. Bb5+ Nd7 13. Bxd7+ Kxd7 14. O-O $14) ( 9... Bxd3 {engines like this move, but it betrays the original idea of the variation.} 10. Nxd3 Qb6 11. O-O {and White has an easy game, while Black is solid but without real prospects.}) 10. Bxf5 gxf5 {this is what Anand was going for with the variation.} 11. Nd3 {this avoids a possible exchange on e5 and protects b2, but is still a backwards move of the same piece in the opening. Moreover, exchanging on e5 or taking on b2 for Black does not look dangerous.} (11. Nd2 Nxe5 (11... Qxb2 $6 {going pawn hunting will leave Black dangerously behind in development.} 12. Rb1 Qxa2 13. O-O Qa6 14. Ra1 Qb5 15. Qf3 $16 {with Rfb1 a threat.}) 12. dxe5 Nd7 13. O-O Rg8 14. Nb3 $14) 11... e6 { the standard consolidating move for Black in this structure.} 12. Nd2 {getting the final minor piece out.} Rg8 {by this point Black has achieved a dynamic equality. He has some longer-term structural weaknesses (including the h7 pawn and dark-square holes) but in return he has dynamic piece play and the half-open g-file.} 13. O-O O-O-O {the only real place to put the Black king, also developing the queenside rook.} 14. a4 {now that Black has committed with his king, White grabs some space on the queenside. This is not dangerous for Black, though, as a4-a5 is not yet possible.} Ne4 {improving this piece considerably, as the most it was doing on f6 was guarding h5. With Black's king tucked away on the queenside, though, that is not critical.} 15. Rc1 (15. Qh5 Rg7 $11) (15. f3 Ng3 16. Rf2 Be7 $11) 15... Bd6 {a case where exchanging minor pieces is clearly indicated for Black. White's dark-square bishop would otherwise help dominate the dark squares, while the Black counterpart has nowhere else as useful to go.} 16. Bxd6 Nxd6 17. b4 {this type of position is important for Caro-Kann players to understand. White's pawn thrusts on the queenside look scary, but Anand deals with them effectively.} Kb8 {taking the king off the c-file and removing it from potential tactics involving the rook opposite it. Also vacates the c8 square for another piece. This is a good example of a move that "doesn't do anything" immediate but is valuable in the long run, with White not having a concrete threat in the meantime.} 18. Qe2 ( 18. a5 Qc7 19. Nc5 Ne4 $11) 18... Qc7 {proactively retreating the queen. It was doing no good on b6 anyway, now it can move along the 7th rank and also is well-positioned on the b8-h2 diagonal.} 19. Qe3 {White with this move basically admits he has nothing on the queenside and tries to generate some action in the center.} Ne7 {improving his weakest piece. On c6 the knight was shut down by White's pawns.} 20. f3 $6 {perhaps attempting to be prophylactic and shut Black's knight out of e4. However, now White has a more serious dark-square weakness, absent his bishop, and it affects the space in front of his king, with g3 and e3 now becoming more vulnerable.} (20. Kh1 {would step away from the g-file and keep the balance.}) 20... Ng6 {Black's knight immediately gets into play. The threat is now f5-f4, as we shall see.} 21. Ne5 $2 {White must have miscalculated the impact of Black's next move.} (21. Kh1 { as in the previous note is more prudent, but Black still gets an edge.} f4 { the pawn is tactically protected, due to a "removal of the guard" being available if White takes with the knight.} 22. Qe1 (22. Nxf4 $2 Nf5 $1 { chasing away the queen and the Nf4's only protection.}) 22... Nf5 23. Rf2 { and White's pieces are awkwardly tied up.}) 21... f4 22. Qe1 Nf5 23. Nxg6 { this just clears the way for Black on the g-file, but White appears to be losing in all variations.} (23. Rf2 {would be analagous to the above variation, but now White has a lot more problems. The Ne5 is vulnerable to ...f6, among other things.} Ngh4 $1 {and White has no good response to Black's threats, including piling up on the g-file and playing ...Ne3.} 24. Nf1 (24. c4 { doesn't gain White anything after} Rc8 $19) (24. Ng4 h5 25. Nf6 Rg6 26. Qe5 Ne3 {and White loses material.}) 24... f6 25. Ng4 h5 26. Ngh2 Rg6 $19 {it's looking grim for White on the g-file and Black also has the ...e5 pawn lever coming.}) 23... Rxg6 24. Rf2 Rdg8 {simple and effective.} 25. c4 (25. Qe5 { exchanging queens won't help White.} Qxe5 26. dxe5 Nh4 $19 {and the g-pawn falls.}) 25... Ne3 {Black correctly ignores the attempt at counterplay on the c-file.} 26. cxd5 Nxg2 {no need to wait for the capture, as White's queen is now also en prise.} 27. Qe5 Qxe5 28. dxe5 Ne1+ {with the fork ...Nd3 to follow. A strong finish to the game by Anand.} 0-1

Training quote of the day #20: Dommaraju Gukesh

Gukesh is unusual in that he doesn't obsess about his results. He plays every game to win. He tries to be flexible in his opening choices and he studies the classics. He is level-headed enough to know that this aggressive maximal approach is going to lead to losses every so often and he can shrug them off.

...His favourite player is the late Bobby Fischer...In his best games, Gukesh concocts a similar blend of classic positional build-ups coupled with sharp tactics. 

(From the article "Dommaraju Gukesh's journey to becoming India's youngest grandmaster" in the Business Standard)

Training quote of the day #19: Lyev Polugayevsky

From GM Lyev Polugayevsky's Grandmaster Performance:

Some 30 years ago, when I was still a boy, I was given some advice by one of the oldest Soviet chess masters, one of Alexander Alekhine's fellow players back in the 1909 St Petersburg Tournament, Pyotr Romanovsky.  "If you want to play well," he said, "in the first instance study games.  Your own and other peoples'.  Examine them from the viewpoint of the middlegame and the endgame, and only then from the viewpoint of the opening.  This is more important than studying textbooks."

Perhaps such advice is not indisputable, perhaps it will not appeal to everyone, but I accepted and have followed this recommendation all my life.  Of course, on becoming a master, and then a grandmaster, I had to make a detailed acquaintance with opening monographs and with endgame guides, but nevertheless the analysis of games still remains for me the most important thing.

Annotated Game #206: Don't forget your own preparation

In this last-round tournament game, my problems (as Black, in a main line Slav) can be initially traced to forgetting my own preparation.  On move 7 I start believing (incorrectly) that my opponent has played a significant deviation from theory (and therefore could be "punished" for it).  So on move 8 I make a dubious choice.  The line isn't necessarily losing - in the notes you can see that GM Bent Larsen even played it (although he lost that game...) - but it gives my opponent a free positional plus, in a rather imbalanced position (White king in the center, but with a defensive pawn mass and half-open g-file to compensate).  Unfortunately ratings fear and loathing also seemed to play a part in my decision-making process, to no good end.

Despite the favorable position, my opponent shortly gives back the advantage with the loosening 13. f4?! and I aggressively follow up, achieving a tactically winning position by move 19.  At this point a calculation error on my part leads to a "safe" choice, which instead of consolidating the winning advantage takes me into a rook and minor piece endgame, only giving me a pawn for my troubles.  My opponent plays well after this and my own game deteriorates rapidly, in part no doubt due to the psychological letdown, but also due to my failure to recognize the importance of rook activity (the key to success in rook endgames).

Looking back on this tournament, it was generally a disappointment, despite it being the first one where I defeated a Master (and even if only due to a tactical oversight, in Annotated Game #199).  My performance continued to plateau in the middle of the Class B rating range and the quality of my play was generally not satisfactory, being too variable.  Essentially I was both unable to properly convert advantages (as in this game) and made too many poor decisions that overlooked the strength of my opponents' replies.

I'll continue the fundamental practice of looking deeper into my own games for lessons, but I also plan to resume providing Commentary games at the Master level, based on matches of particular interest and relevance.  The ongoing 2019 Tata Steel tournament has provided some excellent recent examples, plus I have a number of other ones I've been saving for personal analysis over the past year or two.

[Event "?"] [Site "?"] [Date "????.??.??"] [Round "?"] [White "Class D"] [Black "ChessAdmin"] [Result "1-0"] [ECO "D16"] [Annotator "ChessAdmin/Komodo 11.2"] [PlyCount "85"] {[%mdl 8256] D16: Slav Defence: 5 a4: Lines with 5...Bg4 and 5...Na6} 1. d4 d5 2. c4 c6 3. Nf3 Nf6 4. Nc3 dxc4 5. a4 Na6 {the relatively rare Lasker Variation in the mainline Slav. Its main benefit is that it avoids a lot of theory.} 6. e4 Bg4 7. Be3 {normally the c4 pawn is immediately recaptured here, but the move order is not in fact important. However, I erroneously thought my opponent was deviating, contributing to the poor decision on move 8.} e6 8. Bxc4 Bb4 $6 {here I forget my opening preparation and play the wrong piece to b4, being excessively optimistic about my prospects, with vague notions of following up with ...Qa5. I have to think that the ratings difference contributed to this faulty mindset. Now White gets an immediate advantage, with the Na6 out of play and better central control.} (8... Nb4 {is standard and the main idea behind Black's 5th move.}) 9. Qd3 Bxf3 {the best reaction, although the doubled pawns aren't really detrimental to White, who can also think about using the half-open g-file.} 10. gxf3 {simply looking at the state of piece development shows White's advantage, with Black's minor pieces scattered and not cooperating well.} Qa5 (10... Nc7 11. Rg1 Nh5 12. Ke2 g6 13. d5 Bd6 14. dxc6 bxc6 15. Qd4 O-O 16. Rgd1 Bxh2 17. Qxd8 Rfxd8 18. Rxd8+ Rxd8 19. Bxa7 Be5 20. a5 Nf4+ 21. Kf1 Rd2 22. Nd1 Na8 23. Be3 Rd7 24. a6 Kf8 25. Nc3 {Polugaevsky,L (2625)-Larsen,B (2620) Riga 1979 1-0 (39)}) 11. Rg1 g6 {I thought for a while here and decided that blunting White's pressure on the g-file first was important.} ({The immediate} 11... O-O {is favored by Komodo.} ) 12. Ke2 {a natural move, connecting the rooks. White's king is also currently well-protected by his pawn mass.} (12. Kf1 $5 O-O $14 13. Ne2 { would alternatively allow White to bring the knight into action.}) 12... O-O 13. f4 $6 {over-ambitious, as it drops White's king safety and allows my queen to get in the game, instead of languishing on a5.} (13. Na2 $5) 13... Qh5+ $15 {immediately taking advantage of the open h5-d1 diagonal.} 14. Kd2 (14. Kf1 Rad8 15. e5 $15) 14... Nc5 {I thought for some time here as well, deciding to take advantage of the tactical possibilities offered by White's K+Q lined up on the d-file.} (14... Qxh2 $6 {is too greedy, opening up Black's kingside to the White rooks.} 15. e5 Nd5 16. Rh1 Qg2 17. Qe2 $14 {with the threat of Rag1 coming.}) (14... Rad8 {is more direct.}) 15. dxc5 {this is in fact the best choice.} (15. Qe2 $2 Ncxe4+ 16. Kc2 Qxe2+ 17. Bxe2 Nd6 $19 {is probably the simplest path to victory, with Black now a pawn up and White's structure shattered heading into the endgame.}) 15... Rad8 16. Bd4 Bxc5 17. Ne2 $2 { a reasonable-looking defensive move, but it should lose.} (17. Bxf6 {would hold out} Rxd3+ 18. Bxd3 Bxf2 19. Rgf1 Qxh2 20. Ne2 $11 {and the engine considers the position equal. Black no longer can make progress against White's king and my opponent has equivalent material. I'd still prefer playing my position, though.}) 17... Bxd4 $19 18. Nxd4 c5 19. Rg5 Rxd4 $2 {I thought that this lead to a "safe" win, but it is in fact a major calculation error. Any Black queen move that threatens f4 wins, as it would result in a fork of the Kd2 and Rg5.} (19... Qxh2 {is the most straightforward.} 20. Rxc5 Qxf4+ 21. Qe3 (21. Kc2 Qxf2+ $19) (21. Ke1 Nxe4 $19) 21... Nxe4+ 22. Ke2 Qxe3+ 23. Kxe3 Nxc5 $19 {is probably the simplest winning continuation.}) (19... Qh4 20. Rxc5 Qxf4+ {etc.}) 20. Qxd4 cxd4 21. Rxh5 Nxh5 $17 {so instead of winning major material, I'm a just pawn up in a R+B vs. R+N endgame.} 22. f5 e5 {reinforcing d4.} (22... Kg7 {is a safer choice, getting off the g8-h2 diagonal.}) 23. Bd5 Nf4 {this visually looks good, centralizing the knight, but is inaccurate. Either activating the rook - rook activity being paramount in endgames - or breaking up White's pawn formation would be better.} (23... Rd8 $5 24. fxg6 hxg6 25. Rc1 Rd7 $17) (23... Rc8 {is also good.}) (23... gxf5 24. Rg1+ Kh8 25. exf5 Nf6 $17) 24. Rc1 (24. fxg6 hxg6 25. Rc1 Nxd5 26. exd5 Rd8 27. Rc7 Rxd5 28. b4 $15 (28. Rxb7 $6 Ra5)) 24... Nxd5 {betraying a lack of imagination. Better would be to undermine the bishop's pawn support first.} (24... gxf5 $5) 25. exd5 $15 {now we have a pure rook endgame and White's rook is better, giving him some compensation for the pawn.} Rd8 26. f6 {I missed this idea and started to feel a lot of pressure. This was unnecessary, however, as I could have immediately acted to free the king with an h-pawn move.} Kf8 $2 {Black is ruining his position, comments Komodo via the Fritz interface.} (26... h5 27. Rc7 {at the time, this looked too scary, but it works out in the end for Black. } Rxd5 28. Rc8+ Kh7 29. Rc7 Rd6 30. Rxf7+ Kh6 {and the f6 pawn is doomed.}) 27. Rc7 $18 {White is now winning.} Ke8 {I understood at this point I was in big trouble.} 28. Re7+ {the correct continuation.} (28. Rxb7 $2 {is what I was hoping for.} Rd7 29. Rb8+ Rd8 30. Rxd8+ Kxd8 31. Kd3 $16) 28... Kf8 29. Rxb7 a6 30. d6 $1 {a great move, using my back-rank problems to freeze the Rd8.} Ke8 { everything loses at this point.} 31. Re7+ Kf8 32. Ra7 {still winning, but not optimal.} (32. d7 $18 {would clinch things for White now, as I'm frozen and White can march his king forward.}) 32... Ke8 33. Rxa6 $6 Kd7 $2 (33... Rb8 { again, the principle of rook activity dominates.} 34. Kc2 Kd7 $14) 34. b4 $18 { now White is back on track.} Ke6 35. b5 Rxd6 $2 {desperation.} (35... Rb8 36. Kd3 $18) 36. b6 Kxf6 37. a5 e4 38. Ra7 Rd5 39. b7 Rb5 40. Ra6+ Kg5 (40... Kg7 41. Rb6 Rxa5 42. b8=Q e3+ 43. fxe3 Ra2+ 44. Kd3 Ra3+ 45. Kxd4 Ra4+ 46. Kd3 Ra3+ 47. Ke2 Ra2+ 48. Kf3 Ra5 49. Qd8 Rf5+ 50. Kg3 Kh6 51. Kg4 Kg7 52. Rb8 h5+ 53. Kg3 h4+ 54. Kg4 Rh5 55. Qd4+ Re5 56. Qxe5+ f6 57. Qe7+ Kh6 58. Rh8#) 41. Rb6 e3+ 42. fxe3 dxe3+ 43. Kxe3 (43. Kxe3 Re5+ 44. Kd4 $18) 1-0

Training quote of the day #18: Mikaela Shiffrin

Mikaela Shiffrin

“You start to think, ‘Yeah, maybe I am unbeatable,’ ” Shiffrin said. “And as soon as you start acting that way, you get beat.”

(From the Washington Post article, "The world's most dominant athlete at the moment is 23 years old, and getting better")

Annotated Game #205: A strategic turning point that is tactical

This next tournament game features an interesting struggle in the English, versus a King's Indian Defense (KID) structure.  As often happens, White tries to press on the queenside while Black focuses on the kingside.  Here, I think both sides neglected the center too much and missed better opportunities, so that is a useful general lesson to take away.

The key strategic turning point occurred around moves 19-21, where Black makes his initial kingside push.  I only considered the idea of pushing g4 too late, when it was forced, and did not consider the full ramifications of Black's strong Nf5.  For me, it is a good example of how tactical threats and strategic considerations can intersect and determine the course of a game.

[Event "?"] [Site "?"] [Date "????.??.??"] [Round "?"] [White "ChessAdmin"] [Black "Class A"] [Result "0-1"] [ECO "A26"] [Annotator "ChessAdmin/Komodo 11.2"] [PlyCount "100"] {[%mdl 8256][%evp 0,100,16,-12,-9,-28,36,28,19,19,11,-1,31,7,50,-10,-2,-17,-4, -20,-14,-11,-11,-9,44,9,10,1,16,21,20,8,8,-4,-9,-18,-7,-7,2,-13,20,7,43,-68, -74,-138,-149,-136,-96,-100,-103,-89,-92,-73,-67,-217,-203,-209,-187,-187,-144, -162,-163,-168,-147,-174,-178,-191,-218,-226,-238,-229,-253,-253,-223,-220, -230,-244,-111,-273,-274,-274,-274,-300,-338,-289,-229,-235,-242,-242,-241, -237,-236,-466,-295,-286,-378,-767,-848,-29997,-29998,-29999,-30000] A26: English Opening vs King's Indian with ...Nc6 and d3} 1. c4 Nf6 2. Nc3 g6 3. g3 Bg7 4. Bg2 O-O 5. Nf3 d6 6. O-O e5 7. d3 {the standard setup for White in the English, facing the KID structure.} Nc6 8. Rb1 Re8 9. Bg5 {a key idea in this variation for White. As soon as the Black rook gets off the f-file, the bishop comes out and can be exchanged for the Nf6. This gets rid of a strong Black attacking piece on the kingside, while not giving Black as much potential momentum on the f-file. Naturally it's not the only way to play, but I find the idea simple to execute.} h6 10. Bxf6 Bxf6 11. b4 Bg4 12. h3 {the first real decision point for White. This move is not the most incisive, as it's not clear that the bishop is particularly well placed on g4. The h3 pawn can also become a target.} (12. b5 $5 {is more straightforward.} Ne7 13. Nd2 $14) 12... Be6 13. Nd2 (13. b5 Ne7 14. Nd2 Qc8 15. Kh2 Bg7 16. Qb3 f5 17. Nd5 c6 18. bxc6 bxc6 19. Nxe7+ Rxe7 20. Qa3 Qd7 21. Qa4 d5 22. Nb3 Bf7 23. Na5 Rc8 24. Rb7 Qxb7 25. Nxb7 Rxb7 26. Rc1 dxc4 27. Bxc6 Rbc7 {Medjkouh,S (2080)-Chumfwa,K (2180) Maputo 2011 1-0 (55)}) 13... Rb8 14. b5 Ne7 15. Qc2 {not necessarily a bad move, but perhaps a little timid. The queen has little prospect on the b1-h diagonal, nor does it do much to support anything else. On the positive side, the rooks are connected.} (15. Nd5 $5 {this is a common theme in the English, the knight leaping into d5 only to be exchanged off with doubled d-pawns. However, the resulting advanced d5 pawn and dynamic features of White's position typically compensate for it.} Nxd5 16. cxd5 Bd7 17. Qc2 {and now the queen has more to do on the half-open c-file.}) 15... Bg7 16. a4 {it would have been better to immediately play Kh2 here, anticipating Black's next.} Qd7 17. Kh2 f5 18. a5 {we now have a classic queenside vs. kingside expansion "race" on.} f4 {this is one of those moves that looks aggressive but in reality is premature. White has several good responses.} (18... c6 19. Rfc1 $11 ) 19. Nd5 {an example of a good idea played at a less-optimal time.} (19. g4 $5 {looks a bit loose, but White is set up well enough on the kingside and in the center to keep Black from breaking through.} Rf8 (19... h5 $2 20. gxh5 gxh5 21. Rg1 $16 {and it's White who is better positioned for action. For example} Kh8 22. Nde4 Nf5 23. Qa4 Nd4 24. Bf3) 20. a6 $14) (19. Qa4 {is another idea, lining up on the a4-e8 diagonal.} fxg3+ 20. fxg3 $14) 19... fxg3+ {this reduces the tension and is a typical Class player inaccuracy.} (19... Nf5 $5) 20. fxg3 $14 Nf5 {we now reach a key strategic decision point, where it is revealed that I don't know what I'm doing in the position. The overlooked tactical threat is the knight fork on the e3 square, which is only temporarily covered by my Nd5. The e2 pawn is also currently undefended.} 21. b6 $6 (21. g4 $5 {kicking the knight is the best solution.} Nd4 22. Qd1 e4 23. e3 $14) 21... axb6 $15 22. axb6 $6 (22. g4 $5 {is now required, but at a disadvantage to the previous variation.}) 22... c6 $17 {kicking the Nd5 and gaining the advantage, although it still wasn't too late to play the g4 idea, which I only find after it's forced.} 23. g4 Nd4 24. Qd1 cxd5 25. cxd5 Bxg4 $1 {a winning, if tricky, sacrificial idea.} 26. hxg4 Qxg4 $6 {this could allow White to escape, but I don't find the correct continuation.} (26... e4 {is the winning continuation, but it requires a further sacrifice.} 27. e3 exd3 28. exd4 Re2 {the key idea, as now Black's queen can penetrate to maximum effect.} 29. Rf4 Qe7 30. Kh3 Qg5 31. Bf3 Qxf4 32. Bxe2 dxe2 33. Qxe2 Qxd4 $19) 27. e4 $2 {this optically looks like a good defensive move, but the d5 pawn is not under threat and the pawn does much less on e4 than on e3.} (27. e3 {threatens the Nd4 and Black therefore does not have enough time to generate winning threats.} Qh4+ 28. Bh3 $11) 27... Ne2 $19 (27... Qh4+ {is even more effective.} 28. Bh3 Rf8 29. Nf3 Nxf3+ 30. Rxf3 {and now Black can choose how he wants to win, with White's pieces overburdened and a hole on the 2nd rank. For example} Rf4 31. Rxf4 exf4 32. Qg4 Qxg4 33. Bxg4 h5 $19) 28. Rf3 Nf4 29. Rg3 Qxd1 {my opponent chooses to head for an endgame with an advantage.} 30. Rxd1 Ra8 31. Nc4 Bf8 32. Bf3 $6 ( 32. Rd2 {defending the 2nd rank is better.}) 32... Ra2+ 33. Rd2 {now things fall apart quickly.} (33. Kh1 Rea8 $19) 33... Rea8 (33... Rxd2+ 34. Nxd2 Ra8 35. Nc4 $19) 34. Kh1 (34. Rxa2 Rxa2+ 35. Kh1 h5 36. Bxh5 Nxh5 37. Rxg6+ Kh7 $19 ) 34... Rxd2 35. Nxd2 h5 {Black's connected passed pawns ensure victory.} 36. Rg1 Ra6 37. Nc4 Kf7 38. Rf1 Be7 (38... Ke7 39. Kg1 $19) 39. Rd1 $2 (39. Bd1 $5 {taking advantage of the pin on the Nf4.}) 39... g5 40. Rf1 g4 41. Bd1 Kg6 { by this point the outcome is inevitable, but I stubbornly play on.} 42. d4 { the only chance for any counterplay, even if it is a false hope.} Kf6 (42... g3 {passed pawns must be pushed!} 43. Rf3 g2+ 44. Kh2 $19) 43. dxe5+ dxe5 44. d6 Bd8 45. d7 {it's great to see my pawn one square away from queening, but unfortunately there's no way of removing the Bd8 that is blocking it.} g3 46. Bf3 (46. Rg1 h4 47. Bb3 $19) 46... h4 47. Ne3 (47. Rd1 {a fruitless try to alter the course of the game} Ra4 48. Ne3 Ra3 49. Ng4+ Ke6 $19) 47... Kg5 48. Nc4 Ra2 {the engine now shows a mate in 10, but I accelerate the process.} 49. Nxe5 (49. Rd1 {does not improve anything} Rh2+ 50. Kg1 Rf2 51. Rd2 Rxf3 52. Rc2 h3 53. Rd2 g2 54. Rd1 Ne2+ 55. Kh2 Rf1 56. Rxf1 gxf1=Q 57. Ne3 Qf2+ 58. Kxh3 Ng1#) 49... Rh2+ 50. Kg1 Nh3# {an aesthetic end, at least.} 0-1

Training quote of the day #17: Will Durant

“We are what we repeatedly do. Excellence, then, is not an act, but a habit.”
— Will Durant (not Aristotle)