Lost in the red mist

By April, I'd successfully survived the peak period at work and was able to reward myself with a fortnight off. (For readers with a background in professional services, it will be enough to say that it was almost all given as TOIL.) I spent much of the first week catching up on a long list of unannotated games, and a rare dip into some other chess study*, before turning my attention to Wantage's final match of the Oxfordshire League season. This was against Didcot, who had already been relegated, and since I was on board 2 in a relatively strong team I was hoping to win despite the black pieces.

Their team appeared quite predictable, and since I was not working I was excellently prepared... for the players on the next two boards down. Yes, they had a new player, 1880-rated Dmitry Kolobkov, something I was unreasonably thrown by. I outsmarted myself in the opening, was more than let back it by my opponent's liberties, but then became the author of my own demise:

[Event "Oxfordshire League Div. 1"] [Site "?"] [Date "2026.04.13"] [Round "14.2"] [White "Kolobkov, Dmitry"] [Black "Young, Daniel J"] [Result "1-0"] [ECO "E74"] [WhiteElo "1880"] [BlackElo "2020"] [Annotator "Young,Daniel J"] [PlyCount "71"] [EventDate "2026.??.??"] [SourceVersionDate "2026.04.14"] [WhiteTeam "Didcot"] [BlackTeam "Wantage"] {This was the final league match of the season - and thankfully a dead rubber, since it brought together the bottom two teams in the division. I was in the middle of a fortnight off work so had taken the opportunity to prepare... only for someone new to appear in their team! Sometimes it just isn't your night.} 1. d4 {(1:20)} Nf6 {(1:20)} 2. c4 g6 {My opponent made his opening moves with alacrity, and I was seized by the thought that I should play something out of the ordinary. But I haven't had many King's Indians recently and wanted to keep my hand in.} 3. Nc3 Bg7 4. e4 d6 5. Be2 O-O 6. Bg5 {The Averbakh Variation aims to keep Black's counterplay in check without being dragged into the sort of mate-or-death races that typify the more popular Classical KID lines. It has to be taken seriously and I had duly studied it... a long time ago.} c5 7. d5 h6 {Black's plans involve ...e6 to disrupt the enemy centre, but it's worth relieving the pin against f6 in advance before White can stop it with 8. Qd2. In Marsh-Young, 4NCL 2025, my opponent now played the modest 8. Be3 and I had no problems. This time the choice was} 8. Bf4 {, a more challenging try which puts d6 in the crosshairs. I knew well that Black should nevertheless play 8... e6!, even though this can mean sacrificing two pawns. I knew it, but as the minutes crept past I realised I simply could not remember the justification. Perhaps the unexpected pairing had thrown me off balance. Anyway, after about 15 minutes I decided to fall back on the prerogative of the stronger player, and made something up.} Na6 $2 {I didn't doubt that this was too slow to be good, but at least we were both on our own now.} (8... e6 $142 {Looking back at my notes to the Marsh game reveals that I'd studied this (and Jones's analysis) in depth. Respect for IP (and the copyright laws) limits me to a brief summary in a blogged game, but the key line I couldn't remember was} 9. dxe6 Bxe6 10. Bxd6 Re8 11. Bxc5 $6 {(Critical but risky - White is spending a lot of time clearing the centre for Black's pieces...)} (11. Nf3 {is more measured and far more popular. After} Qb6 {White will resolve his floating bishop with} 12. Bxb8 {, but now Black has straightforward compensation on the dark squares.} Raxb8 13. Qc2 Nh5 $44 {After} 14. g3 {Black has a satisfactory choice between ...Bg4 and ...Bxc3 to win back his pawn on e4, or the more open-ended 14... Bh3!?.}) (11. e5 $5 Nfd7 12. f4 {bravely tries to hold the centre at the cost of developing, although both 12... Nc6 and 12... g5!? give Black counterplay in the style of the Four Pawns Attack.}) 11... Qa5 {I remembered this far but was disheartened by} 12. b4 {, breaking the pin on c3 with tempo, and convinced myself I must be misremembering the theory. However, it is White who must tread carefully after} Qa6 {and Black regains at least one pawn -} 13. b5 $2 {is too greedy and loses rapidly to} (13. Bd4 $142 Bxc4 14. e5 {is most common, though} Nh5 $15 {forces White to find 15. b5 to stay above water - not} 15. Bxh5 $2 Nc6 $3 {(Jones) with a deadly attack for the sacrificed piece}) 13... Qa5 14. Qc2 {and now the typical King's Indian shot} Nxe4 $19) 9. Qd2 Kh7 10. f3 $6 {(1:16) Now the game becomes a kind of Saemisch, but in this case the Benoni structure is quite convenient for Black since it enables him to fight for all the dark squares that White is weakening by placing pawns on light squares. This also obstructs the development of the g1 knight, and at a moment when Black has no particular pressure on e4. In short, the text isn't necessary and White can instead cut to the chase.} (10. h4 $142 {Simply arguing that Black's queenside shuffling is too slow. In this case I would probably have continued consistently with} Nc7 {, trying to build up to ...b5 and some distant hope of queenside counterplay.} (10... Ng4 {gets the silicon nod, however} 11. f3 {now comes with tempo, and Black can't prevent his knight being driven back since} Ne5 12. Bg3 h5 13. f4 Bh6 14. Nh3 Bxh3 {should be answered with} 15. gxh3 $1 $16 {to maintain control of g4.}) 11. a4 a6 {The simple controlling move 12. a5 looks more than fine for White, but he can also try the more adventurous} 12. h5 g5 13. Bxg5 $5 {, which proves no more than a sham sacrifice given that Black's queenside pieces are so distant.} hxg5 14. h6 Bxh6 (14... Bh8 {keeping lines closed is no better. White can calmly bring up the reserves with} 15. Nf3 $1 {, not fearing} g4 16. Ng5+ Kg8 17. h7+ $18 {and Black has to return the knight, as} Kg7 $2 18. Nf3 {leads to mate on the h-file.}) 15. Qxg5 Ng8 {Black defends everything for now, but with} 16. e5 $1 {(a clearance sacrifice to open the b1-h7 diagonal)} dxe5 17. Nf3 f6 18. Qg3 Rf7 19. Bd3+ Kh8 20. Nh4 $16 {, White is primed to invade on g6 with all the chances. Black's problem is that he is defending without his queenside - a consequence of the sluggish ...Na6-c7 manoeuvre.}) 10... Nc7 {(1:02)} 11. h4 $2 {Now the weakening of the dark squares last move takes concrete form. This time Black gets chances of his own against the h-pawn advance, since the g3 square is his for the taking.} (11. a4 $142 {White could still have kept the upper hand by restraining Black's queenside.} a6 12. Bd3 Nd7 13. Nge2 Ne5 {Here Stockfish suggests a move I would never dream of:} 14. Bxe5 $5 {To give up the dark squared bishop after placing all the central pawns on light squares feels criminal, and yet the computer is banking on the gain of time.} (14. a5 {seems more human, but the argument is that} Nxd3+ 15. Qxd3 Rb8 16. O-O b5 17. axb6 Rxb6 $132 {gives Black chances down the b-file which make up for the weakness of a6.}) 14... Bxe5 {(taking with the pawn would deaden this bishop)} 15. f4 Bg7 16. O-O $14 {It is true that White maintains an imposing pawn centre whilst holding the ...b5 break in check, but despite the engine's optimism I'd be satisfied with this in a practical game, as the dark-squared bishop may yet have its say and the position is imbalanced enough to outplay an opponent.}) 11... Nh5 12. Be3 $6 {White chooses to keep the bishop on an active square, at the cost of surrendering g3.} (12. Bh2 {If White had dropped back here to keep the knight out then Black has} e5 (12... e6 {was my idea, but this is an inferior version since White can target the knight at the edge -} 13. g4 $1 Qxh4+ 14. Kd1 {in the main line Black would be glad to head for f4, but now that square is unprotected and} Nf6 $2 {also fails to the remarkable} 15. Nh3 $3 {, playing to connect the rooks and also highlighting the enemy queen's lack of squares.} Qxh3 16. Kc2 e5 17. Rag1 {Now the situation becomes clear.} Kg8 18. Bg3 Qxh1 19. Rxh1 $16 {Material is approximately level (Black has a rook, knight and pawn for the queen) but White retains good attacking chances on the queenside and has already opened the h-file, whereas Black's play on the other flank is yet to get started.}) 13. Bd3 (13. g4 {As mentioned above, this sacrifice is now a dark square weakness too many in view of} Qxh4+ 14. Kd1 Nf4) 13... b5 $1 {We will see this Benko-style sacrifice again and again in the subsequent variations. Whilst White is unknotting his kingside (the last move was to prepare Nge2), Black must get some counterplay going on the queenside, even at the cost of a pawn. In fact the sacrifice works very well because the weakening moves that White has made on the kingside push him towards long castling, and the open lines will then come in handy.} (13... Qxh4+ $6 14. g3 Qe7 15. Nge2 $44 {will offer White fair play for a pawn after a later g3-g4: he has lost his h-pawn but gained the file in return.}) 14. cxb5 a6 15. bxa6 Bxa6 $15 {Why is Black better despite his pawn deficit? It may be helpful to compare with Young-Gluckman, SCCU Ch. 2023, where Black tried a similar sacrifice but didn't get close to enough for it. In that game, however, he had no pieces in support on the queenside so I could just take it and carry on with a much faster kingside attack. This time Black's heavy pieces have easy access to the queenside, and White's attack has stalled: to make progress he would need to get the h5 knight out of the way, but g2-g4 creates a big hole on f4. White's development is also more suspect than in the earlier game. Let us consider a few more moves:} 16. Nge2 (16. g3 {is an ugly way to defend h4, shutting in the bishop on h2, but not completely stupid. In this case} f5 {takes over the initiative in the centre as well:} 17. exf5 gxf5 18. Nge2 Bxd3 19. Qxd3 Kh8 $15 {, when the rolling centre and White's entombed bishop are worth more than a pawn.}) 16... Bxd3 17. Qxd3 Qxh4+ 18. g3 Qf6 {Now castling queenside invites a terrifying attack after ...Rfb8 and ...Na6(-b4). The machine prefers to remain in the centre, although} 19. Rd1 Rfb8 20. b3 Rb4 {gives Black control on the queenside, and mastery of the dark squares, in a position where material is equal.}) 12... e6 $6 {I didn't want to do it until provoked (and White dutifully obliges) but Black should take the opportunity to land on g3 and exchange on e2, after which he can strike out with ...b5 and obtain similar play to the last variation.} 13. g4 $6 {Pawns don't move backwards!} (13. Bd3 $142 $1 {Now once the black knight lands on g3 it won't have any onward square.} Ng3 14. Rh2 exd5 15. exd5 {White should have no interest in giving Black counterplay with 15. cxd5?! f5. It's true that the text allows Black the e-file instead, but his kingside remains under pressure after} Bf5 16. h5 Bxd3 17. Qxd3 $14 {to prevent the capture on h5.}) 13... Ng3 14. Rh2 exd5 15. exd5 {(1:00) Again, this is the right capture, not allowing Black ...b5 too easily. Now the threat of Bd3 is renewed, and Black has little choice but to make the trade on e2 - although it's a shame to give up the marauding knight for a bishop restricted by its own central pawns, the knight itself doesn't have anywhere to go.} Nxe2 {(57)} 16. Ngxe2 Re8 $6 {A natural developing move of course, however Black's most important strategic idea is ...b5, and he should take the opportunity to play this before White can prevent it.} (16... b5 $142 $1 17. cxb5 Rb8 {(if 17... a6?! then White would be glad to keep lines closed with 18. b6) Stockfish is happy to brave} 18. O-O-O {here rather than remain in the centre, but now} Qe7 19. a4 a6 $44 {creates a very imbalanced position. White has gained a pawn and made inroads on the kingside, but Black has two bishops and more immediate threats to open 'his' side of the board.}) 17. h5 $6 {This is an understandable way to proceed. White allows Black to close the kingside files with ...g5, but in return the f5 square becomes vulnerable. However, even in a sharp position like this it can be worth taking a prophylactic approach: Black's main active idea is ...b5, and playing 17. a4 to stop it keeps all of the kingside ideas in reserve.} g5 18. Qd3+ $6 {Part of the same plan, pointing the queen at the f5 square with tempo. I was surprised that the machine was unimpressed with such a logical move, but it prefers to keep the queen on d2. The reason for this is to pursue a different breakthrough on the kingside.} (18. O-O-O $142 {has the point that after} b5 19. cxb5 Rb8 {, when Black has the usual play for his sacrificed pawn, White can strike out with} 20. f4 $13 {when both sides of the board are burning.}) 18... Kg8 19. Ng3 $2 {And now the advertised plan leads to serious problems. White is playing with fire by leaving his king in the centre for so long, and now I could sense that the time for decisive action was near.} (19. O-O-O $142 {was safer, although there is already a price for White to pay since} b5 20. cxb5 Rb8 $17 {is now faster for Black. The point is that} 21. f4 $2 {no longer threatens g5 enough times, so Black can crash through with} Nxb5 $1 22. Nxb5 Ba6 (22... Rxb5 $144 23. Qxb5 Rxe3 $19 {, opening the b-file towards White's king and removing his important bishop, also seems good}) 23. Nec3 (23. a4 Qa5 24. Nec3 Bxb5 25. axb5 c4 $1 26. Qd2 Rxb5 $1 {breaks through, since} 27. Nxb5 Qa1+ {mates}) 23... Bxb5 24. Nxb5 c4 {, regaining the piece - if} 25. Qxc4 Rc8 {wins.}) 19... Qf6 $6 $17 {As well as attacking f3, this move looks to intensify the pressure against the pinned bishop on e3. These are not easy threats to deal with when your pieces are uncoordinated, but Black can improve this idea with a familiar finesse.} (19... b5 $142 $1 {As always, the opening of queenside files is a welcome development for Black, but in this case the move also helps to undermine White's pawn centre. Now on} 20. cxb5 (20. Nxb5 $6 Nxb5 21. cxb5 Qf6 $19 {wins back the pawn at once.}) 20... Qf6 {hits even harder than in the game because White also has to keep an eye on d5. In the game White's best try is} 21. Qe2 {, which is now defeated by} Qe5 22. Nge4 (22. Kf2 a6 23. bxa6 Bxa6 24. Qd2 Rab8 $19 {threatens the devastating 25... Nxd5}) 22... Bb7 $19 {and White will quickly come to grief once the centre opens.}) 20. Rf2 $6 {(29) Just defending the f-pawn - not 20. Kf2? Bxg4.} Qd4 $1 {(37) So Black attacks something else.} (20... Qf4 $143 $4 {was also tempting but would have led to disaster after} 21. Nce4 {, defending the other knight and relieving the pin on the e-file. Black's queen is now attacked, but} Qe5 22. f4 gxf4 23. Bxf4 $18 {hits an important pawn on d6. If} Qe7 24. O-O-O Rd8 25. g5 {and White's attack lands first.}) 21. Ke2 $8 {This is forced, as 21. Kd2? Rxe3 wins. But now how should Black make progress? All the indicators said that I should have something: White's king is stuck in the centre and caught in a pin, neither of his rooks is helping the defence, and Black has a vice-like grip on the dark squares. But White's knights are primed to jump into e4 and/or f5, sealing the e-file and pressuring d6. I felt that I needed to act quickly before White could consolidate, and was painfully aware that I could do with some of my queenside pieces to help. Unfortunately, this accurate insight turned into panic, which, combined with a desire to punish White's simplistic opening play, resulted in a terrible misjudgement.} Bxg4 $4 {Essentially a sacrifice to gain time to bring the a8 rook into play: I had ideas of ...Rxe3 followed by ...Re8 to win the white queen, and felt I could capitalise on the loose kingside pieces. However, these ideas prove to be insufficient, and that leaves White free to do exactly as I feared and take over with his knights. I can't even say that I overlooked a defensive resource; I felt this sacrifice had to be good, and didn't look deeply enough. A typical case of evaluation by emotion.} (21... Bd7 $2 {I could tell that tepid development would be too slow, for example} 22. Nge4 Qxd3+ 23. Kxd3 {, although even here} b5 {is nice for Black since} 24. Nxd6 bxc4+ $15 25. Nxc4 $2 {would lose material to} Bxc3 26. bxc3 Nxd5 27. Bxc5 Nf4+ 28. Kd2 Rec8) (21... Rxe3+ {If Black is intent on sacrificing something (and I wasn't consciously playing to the gallery; I just played what I thought was best) he can play this, removing a key protector of the dark squares.} 22. Qxe3 Qxc4+ 23. Qd3 Qf4 24. Nge4 Bd4 $17 {Black has excellent compensation with the bishop pair and a pawn for the exchange, with White's king still stuck in the centre, but there is better. With tedious predictability, Black's best move is...}) (21... b5 $142 $1 {This can hardly come as a surprise to the reader who has been following the notes up to this point. This is a less committal solution to the problem of queenside development, which also opens up a new front on the queenside.} 22. cxb5 (22. Nge4 {I think I lost my composure because I knew a knight jump would likely force a queen trade, when my chances of exploiting White's centralised king would seemingly evaporate, but in this case} Qxd3+ 23. Kxd3 bxc4+ $19 {wins a pawn and opens the b-file for Black, since} 24. Kxc4 Ba6+ 25. Kb3 Bxc3 {(not a move to make lightly, but sometimes it is good!)} 26. bxc3 Nxd5 {demolishes the white centre.}) 22... Nxb5 $1 {A similar idea to the 19. O-O-O variation above, exploiting the overloaded white queen.} 23. Nxb5 (23. Qxd4 Nxd4+ 24. Kd2 Rb8 25. b3 {is refuted by} Nxf3+ $1 26. Rxf3 Bxg4 {, removing the defender of the e3 bishop.} 27. Rf2 Bxc3+ 28. Kxc3 Rxe3+) 23... Qxb2+ 24. Kf1 Qxa1+ 25. Kg2 Ba6 $19 {White's king has finally found shelter, but only at the price of the exchange and a pawn.}) 22. fxg4 Qxg4+ (22... Rxe3+ $143 $2 23. Qxe3 Re8 24. Nge4 $18 {refutes my other idea}) 23. Rf3 Bd4 24. Nf5 {Of course White defends by advancing his pieces, rather than meekly retreating them. There's no need to slip into assumptions of attack and defence: White is a piece up with the enemy king in his sights!} Qg2+ (24... Qxh5 25. Nxd4 cxd4 26. Qxd4 f5 {at least forces White to find the only really good move, to whit} 27. Rg1 $18 {and the threat of ...f4 is repulsed.}) 25. Rf2 {(15)} Qg4+ {(11)} 26. Rf3 Qg2+ {By now reality had set in, but of course White has no interest in repeating more than twice.} 27. Rf2 Qg4+ 28. Kd2 Bxe3+ 29. Qxe3 {Flashy, but the most punishing. Black can defend h6 with} Qxh5 {, but now the h-file is opened for White's use.} (29... Rxe3 {of course fails to} 30. Nxh6+ Kg7 31. Nxg4 $18) 30. Ne4 {(12)} Qg6 {(8)} 31. Rh1 h5 {Black has three pawns for the piece, but his kingside is falling apart (those poor queenside pieces...) and White polishes off efficiently.} 32. Rg1 g4 33. Nh6+ Kh8 34. Qc3+ Kh7 35. Rxf7+ {(5)} Qxf7 {(3)} 36. Nxf7 {Seeing now that 36... Rxe4 would be answered with 37. Ng5+, I decided there was nothing to play for even in a blitz finish. Wantage eventually limped to a 3-3 draw, which was a disappointing finish to the league season for the team as well as for me. At least we get to go again next time. For myself, I was clearly out of sorts in this game and don't think I ever quite recovered from the unexpected opponent. I doubted my memory, had a blind spot for the thematic ...b5 break, and lost the game with a totally unnecessary piece sacrifice.} 1-0

I think this game is most notable for its psychological content, although the chess is quite interesting too. Memory blanks do happen (and all the more so with age) and are a risk if you play theoretical openings like the King's Indian, but in hindsight I think it was wrong to improvise on move eight. My reasoning was that sacrificing two pawns for I-couldn't-remember-what was too risky, however the amount I could remember was all correct and it's usually best to play good moves. When I was younger I would go into the theoretical line anyway and trust myself to reconstruct it at the board, which is a reasonable bet in positions you are familiar with.

Nevertheless, I got away with it since White was provoked into ramming his kingside pawns up the board rather than developing his pieces. But I was undone by a baffling refusal to ever play ...b5. How much more thematic a move can you get in a Benoni structure? I've played enough Benko Gambits in my time that I don't think I was put off by having to sacrifice a pawn - I simply never thought about it. Rather, I caught sight of the enemy king in the centre and started hyperventilating. The idea of opening a new front to get at it from the flank was too sophisticated for me in that kind of mood. My ...Qf6-d4 idea was actually fairly good, but the appalling 21... Bxg4?? was not. The attack was doomed to fail with my queenside pieces watching in the corner, and despite White's questionable pawn play his king has plenty of pieces nearby for cover. This was a return of my old 'chessboard policeman' persona: I wanted to punish White for his provocative play and lost my objectivity as a result.

This was obviously a very deflating loss, not least because it came at a moment when I had been working quite intensively on chess. However, I found some considerable solace over the next few weeks with the conclusion of the Wantage club championship. I won both of my last two games in miniatures to finish on an unbeaten 5/6 (in fact my four wins were achieved in just 90 moves between them). When my opponent from the previous entry tripped up in his last game I finally became a club champion, having been the runner-up in four of my last five tries at Wallington and Epsom.

However, as in previous years, my activity dried up with the end of the league season. Surrey's U-2050 team turned in its traditional 7½-8½ defeat in the first round of the national stage, to which I could only contribute a grovelled draw, but otherwise I still couldn't get any slowplay practice in ahead of my usual August event - it's a strange hole in the calendar. This year I'm playing the AM Open at the British Championship in Warwick, from which I am writing this post, and that's where we shall head next.

* Specifically, I started working through Ramesh's book on chess calculation. Whilst the positions are fascinating, the solutions are given in line with the text so it's surprisingly easy to skip straight from questions to answers without realising. That's to say nothing of the difficulty, and I was left with the feeling that you need a training partner to make the most of the material.