We had two matches up the club as our covid re-opening gathers pace – the Killane confirmed their place in the quarter-finals, while the Branagan B bowed out with three decent draws against Gonzaga B, including one against Ireland’s number 1 ranked player.
The Killane were up against UCD earlier than expected – this was an away game for us, but UCD had asked to play it in Benildus and earlier than the original date to avoid clashes with exams. We didn’t really know what to expect – UCD would be missing their top board, but they have a number of unrated players who are a lottery in terms of real strength. They had won their opening game against Elm Mount 5-1, but here they could have lost 5-1 only for them to rescue a couple of games from much worse positions. Karel took a draw on board 1, Wenle and Dimitri both won, and Finn, Danny and Diana were all material up at some stage but all ended up losing, with Diana’s game going down to the wire and finishing past 11pm. Still, 2½-3½ is a solid score; we avoid being the only one of the nine teams in the competition not to go through to the quarters (a round whose only purpose is to guarantee three games for most sides; the Branagan and Brennan are groups of four, so everyone gets three games). The draw will be made after the final group game on 7th May.
Meanwhile the Branagan B were up against Gonzaga B, and the match took an unexpected twist when Sam Collins IM – who hadn’t played for either Gonzaga side in the tournament so far – turned up. This meant I was renewing acquaintances from the 1996 Irish Novices Championship, in which Sam beat me in our individual game and in which I finished runner-up to Sam. Not unsurprisingly, this was the first game of the evening to finish. As we reset the pieces, Vjeko commented that at least I’d only lost 0.2 rating points – but I hadn’t, because I’d drawn. The game swung on two blunders – one from each of us – and is interesting to play through.
Sam Collins IM (2455) v Kevin Burke (1683)
1. e4 e6 2. d4 d5 3. exd5
Boo! I think this isn’t uncommon given the ratings difference; white can trust to better positional play and minimise any tactical stuff. Though, as was pointed out after, it may be more relevant that Sam recently published a chessable course on the French Exchange. As if I wasn’t up against it enough…
3…exd5 4. Nf3 Bd6 5. c4 Nf6 6. c5 Be7 7. Bd3 0-0 8. 0-0 Bg4 9. h3 Bh5 10. Nc3 c6?!
As Sam rightly pointed out afterwards, b6 is the key break for black in these positions. Now I end up with a fairly static position, giving white easy development.
11. Bf4 Nbd7 12. b4 Re8 13. g4 Bg6 14. Bxg6 hxg6 15. Re1 Nf8 16. Qd3 Ne6?! (D)
I saw the problem with this move as soon as I made it. I didn’t even bother hoping Sam didn’t see it!
17. Rxe6 fxe6 18. Qxg6 (D)
So – what now? This isn’t actually as bad as it might seem – not hard given it seems pretty bad. I didn’t like 18…Rf8 19. Bh6, but I think I missed Ne8 holding everything together. Other white moves allow 19…Qe8, relieving the immediate pressure and leaving me with just the positional weaknesses (the lichess computer scores it +0.6). But if I don’t play that, then white plays Ng5 or even Ne5, both of which look horrible, so I played…
18…Nd7?? (D)
…covering both g5 and e5, and coming to f8 next. But now the computer scores +3.7. 19. Qxe6+ or Re1 are the obvious moves and are both good (around +1.9), but the best continuation is far more subtle and nice. Answer at the end!
Sam afterwards said he’d missed this Nd7 defensive resource. He was worried about 19. Qxe6+ Kh8 and where does the queen go? The immediate threats are Bxc5 winning a pawn, and Rf8 winning a piece. Except 20. Qg6 Rf8 doesn’t win a piece, because of 21. Ne2.
19. Bh6??
This throws white’s edge away entirely (0.0 – and it’d be slightly better for black only that white has perpetual).
19…Bf6
Now white doesn’t have time to kick the bishop with 20. g5 because of the reply 20…Nf8, winning a piece. 19…Bf8 also defends, but not as well (+0.8).
20. Ng5
I’d seen this line – up to the game’s conclusion – when playing 18…Nd7 and figured there wasn’t a way through. In fact with the bishop hanging, I thought white was in trouble if he didn’t take the perpetual – so when Sam played this move, I started trying to find what it was I’d missed.
White’s alternative is just to back away the pieces, be slightly worse, and wait for the player 700 points lower-rated to make another mistake.
20…Nf8 21. Qf7+ Kh8 22. Bxg7+
I still presume there’s a mate that I’ve missed, but again I can’t see it.
22…Bxg7 23. Qh5+ Kg8
Still wondering where the mate is!
24. Qf7+
And at this point Sam surprised my by asking if I would like a draw, to which I replied “Yes. Yes I would.” ½-½
My first game against an IM, and still unbeaten against them! Coupled with my draw against Jacob Flynn in the first group game (when again I was worse before Jacob blundered), it’s technically an FM performance in the Branagan.
Elsewhere, Odhrán also picked up a great draw off Dave Murray on board 2. On 6, we had Advaith, one of Vjeko’s students, subbing in after a late withdrawal – his ICU was only 1046, but his FIDE was 1500+ (which made him eligible for the Branagan) and he had beaten Alan Lane earlier in the month, which was a respectable result. Conor Nolan (1800) was a bridge too far in an entertaining game where Advaith gave up a piece to win the rook on h8, but once there, the knight was never coming back out, and also his king was suddenly under direct attack.
Slavko and Dave lost to Carl Jackson and Gordon Freeman, but we did pick up one last result when William held Peter Carroll CM in a fairly unbalanced game where he was material up at the end, but a mixture of time pressure and a lot of weak pawns meant he wasn’t able to see out the win, settling instead for what I think is a first titled draw.
So a 4½-1½ defeat and while it confirms we won’t be going to the semis, it’s still a great evening’s work given our expected score was 0.28. Top seeds Elm Mount round off the group for us in two weeks.
To the game solution so – white’s best move is the surprising 19. g5! It stops Bf6, and suddenly black’s pieces are struggling to get over to the kingside to defend. So for example 19…Nf8 20. Qh5 e5 (if 20…Nd7 to prevent white bringing the knight to e5, then 21. g6 Nf8 (21…Bf8?? 22. Qh7+ Kf8 23. Qh8+ Ke7 24. Bd6#) 22. Ne5 Bd6 (22…Qc8 23. Nf7 Nxg6 24. Nh6+! Kh7 (24…gxh6 25. Qxg6 and mate in 2) 25. Ng4+ Kg8 26. Qxg6 Qd7 27. Be5 Bd8 28. Nh6+ Kh8 29. Nxf7+ winning the queen) 23. Nf7 Nxg6 (the only way to stop mate on h8) 24. Nxd8 and again the queen is gone) 21. Nxe5 Qc8 22. Qf7+Kh8 23. Re1 (D) and black is close to zugswang.
Neither rook can move (23…Rb8 24. Ng6+ and 25. Bxb8). The bishop’s only square (d8) hangs the e8 rook. The knight can’t move; the check on g6 is lethal. Pawn moves do nothing – well, except g6, which opens the diagonal to the king. The queen has three moves – Qd8, Qe6 or Qxh3 – and each is horrible. 23…Qxh3 24. Re3 Qh7 25. Kg2 and you’ve just opened up another line to the king. If black sits still, 23…Qd8 24. h4 Qc8 25. h5 is a slow but certain way to defeat. 23…Qe6 is the computer’s pick, but it loses a piece after 24. Qxe6 Nxe6 25. Ng6+ Kh7 26. Rxe6.