The Armstrong’s non-quest for a fourth fourth-place finish in five years is on course after a tight win in Kilkenny, which only came about after the visiting side stole three wins in the last half hour.
While Kilkenny lined up with Paul Wallace – playing his first league match in many years – we had been looking at using five subs for a while, until at the last moment, two more regulars opted in. Still, with two of the Heidenfeld subs being 1900, we outrated Kilkenny on all bar board 1.
In the event, though, it was quite an even match. Gerry was first to finish with a win over Colm Quigley, before Stephen added a draw in the battle of the titled players on board 1. Then, as the matches entered the last half-hour, I stole the first win on 8 when, having come under a bit of pressure in a relatively uneventful French Exchange game, my opponent, in time trouble, eschewed a check which I thought landed me in a lot of trouble, and instead took a pawn, which just lost a piece.
Brendan lost shortly later when, into his last minute, taking a pawn back and missing an immediate queen-kind skewer, while Mel lost out in an interesting game in a gambit line of the French Tarrasch. That left the score at 2½-2½ each, with three games remaining and all going into the last ten minutes. Tim had earlier had a bizarre position where all his pieces were on their starting squares except for his king’s knight (swapped off) and his queen (on g1, where it was attacking his opponent’s rook on h2 – Tim was black). After the madness calmed down, with both players ignoring king safety for most of the game, Tim found himself in an opposite-colour bishop ending two pawns up. Those two were connected, passed queenside pawns, but that often isn’t enough to win in these kind of endings. As it turned out, his opponent maybe didn’t defend as accurately as he could have, and Tim’s king and bishop did manage to escort the pawns home.
Meanwhile, Mihailo had built up a big advantage in the middlegame, but eventually found himself in what was, effectively, a Q+P v Q ending. Both players did have three kingside pawns, but these were largely irrelevant to the action as Mihailo’s pawn was on b7, and his opponent was effectively looking for a perpetual – entirely possible in what’s a notoriously difficult ending. Mihailo (correctly) marched his king up the board to join his pawn, but then in this position –
– Mihailo played 1. Qa7?? – and his opponent resigned!
First off so, the correct move. The key to these positions is finding a square for the king where it’s not blocking the pawn, and where it can’t be checked – and here, it’s c7. So if 1. Kb8! Qe8+ 2. Kc7 (or 1. Qf4+ 2. Kc8), and black has no more checks and the pawn promotes. So black has to play 1. … Qd7 (to stop 2. Kc7) – but now white reaches the square by going back the other way around the pawn – 2. Ka7 Qa4+ (the queen has no other moves on the seventh rank, pinning the pawn, and moving one of the kingside pieces is also no good – you can check this with Fritz) 3. Kb6 and the king reaches c7 – 3. … Qd7 isn’t a defence any more as the pawn just promotes.
What of the game continuation? What both players saw – and what they said they would have played – is that white now promotes after 1. Qa7 Qe8+ 2. b8=Q – but neither saw that this is in fact a draw! Black, a queen down, can shuffle back and forth between a4, c6 and e8, and white can never stop the checks. This is a common theme in these kind of endings, and, as both players can now attest, is worth knowing.
In a pure Q+P v Q ending, that would be the end of the story, but of course, there’s six other pawns on the board. So there is one winning try for white – and that’s to promote to a knight!
White can now rearrange his army after 2. … Qe4+ 3. Qb7 Qa4+ 4. Na6 Qe8+ 5. Ka7, and black is out of checks. But is it enough for white to win? Yes, but it’s not straightforward. White now needs first of all to protect his kingside pawns – by playing Qg3 or similar – and then bring his king over to black’s position, using the knight to break the black queen’s cut-off of the various files. The h-pawn can also be used to try crack open the black king’s position. Black, meanwhile, can’t really move any of his kingside pieces without weakening his position – either hanging a pawn or fatally weakening f6 – and white will eventually invade and win. But while the play is safe for white, it’s still at least 20 moves with best play before black’s position finally cracks.
Anyways, Kilkenny stole a point back of their own with the last match, when Constantin – still getting used to the “aggressive” style of Irish players! – found himself up an exchange and two pawns, and although his opponent had an attack in compensation, Fritz shrugged it off, giving +5. Fritz, however, wasn’t playing, and the game reached a position of R, B+2 v 2B+3, where Constantin’s three pawns were c, d and e pawns, while his opponent had a and h pawns. But the two rook pawns proved harder to stop, helped by fork threats, and although again, Fritz notes a draw, it still wasn’t playing, and Constantin’s opponent’s threats proved more dangerous, and won out in the end.
So a 4½-3½ win in the end, which at least avenges Kilkenny’s two 5-3 wins over us last season, but which also does leave us in familiar territory, with two matches against the promoted teams coming up next –