IRLchess

The renowned endgame study composer A. J. Roycroft was born in London in 1929. He’s probably best known for his founding and long-time editing of EG, a long-running journal devoted to endgame studies; he also wrote the book Test tube chess: a comprehensive introduction to the endgame study (1972), among others.

It’s not as widely known as it might be that he has a strong Irish connection. He was an undergraduate at Trinity College, Dublin from 1949 to 1953 (French and German Languages and Literature), and throughout those years he was an active player in the local chess scene. Only one game of his is given in the ICU games archive, though to make up for the sparsity it’s a spectacular win against three-time Irish champion Dónal J. O’Sullivan, from the Armstrong Cup 1952-53.

Here’s one more game, versus J. J. Walsh in the Tormey Cup 1952 (source: J. J. Walsh scorebook). This time it’s a loss: from the diagrammed position play continued 16… Be7, and then Roycroft snatched a pawn with 17. Qb8+ Qd8 18. Qxb7 but quickly ran into trouble.

[Click to replay the full game.]

A. J. (“John”) Roycroft also has one other strong link to Ireland I wasn’t aware of before I read his on-line bio: his father, Benjamin Francis Roycroft, was born in Killarney in 1896.

No sooner had the last post been published, with its statement that only one game was available from the 1964 Irish championship, than David McAlister emailed to say that he had a second in his collection, and a significant one too: it’s the crucial Heidenfeld-Reilly clash in round 4 that effectively decided the championship.

We’re still not sure what the original source is, but this game matches the contemporary descriptions given in newspaper reports. One puzzle from the reports was that one claimed that Reilly had the advantage at the adjournment, but that Heidenfeld won a pawn a move after the resumption and the game a few moves later. Even in those days, without the aid of computers, it seemed surprising that a player of Reilly’s strength and experience could go wrong so soon after an adjournment.

The game score makes it clear that Reilly’s advantage at the break was solely in material terms: he was dead lost, and indeed the only slight surprise is that he bothered resuming.

The critical point came a little earlier, in the diagrammed position, with Black to play.

White had been pressing earlier but Black would now hold some advantage after 36… Qd5. Instead Reilly played the tempting but dangerous 36… Rd1+?, and after 37. Rxd1 Rxd1+ 38. Kh2 followed with the decisive error 38… Ra1? (instead of 38… Rd8, say, when it seems White holds a slight advantage in complications). There followed 39. Qd4! Ra4 40. Qd8+ Qc8 41. Qd6+ Ka8 42. Qc7, where the game was adjourned, with the weakness of the back rank spelling doom for Black.

While Black had some chances, all in all the impression given is of a deserved and reasonably convincing win by Heidenfeld, and in turn this means a deserved and reasonably convincing championship.

[Click to replay the full game.]

The very welcome news that the ICU is offering two norm events in Tralee in April brought back memories of other recent norm events and opportunities. One such event was the City of Dublin IM Norm Tournament 2011. No norms were achieved; Stephen Brady finished on 6/9, ½ point short of recording his first IM norm.

Stephen had one missed opportunity, late in the event, that I have never seen pointed out. In his game in the penultimate round, as Black against the Scottish IM Stephen Burns-Mannion, the following position was reached after White’s 81. Bc6-b5.

Stephen now played 81… Kc3?, after which the Lomonosov tablebases show that there is no win; the game dragged on to move 112 before the draw was agreed.

But in the diagrammed position, there was a momentary opportunity for a knockout blow. After 81… Nd5! (and if 82. Ba6 Nc7 etc.), Black wins a second pawn or liquidates into an easily won pawn ending. White’s best appears to be 82. Kf1, but after 82… Ne3+! it’s hopeless, e.g., 83. Ke2 Nxc4 84. Be8 g2! 85. Kf2 Ne3 and White can resign.

[Click to replay the full game.]

[Update, January 27, 2017: It later occurred to me that the ‘etc.’ above was too casual: Black is indeed winning but it’s not a simple matter of being two pawns up, since the g-pawn will soon drop. After 81… Nd5! 82. Ba6 Nc7, if we placed the white king on g1 it would be a draw, whereas on e1 White is lost. The difference? On e1 White will be one tempo short of being able to stop the b-pawn. So: 83. Bc8 Kxc4 84. Kf1 (84. Bd7 Nd5) 84… b5 85. Kg2 b4 86. Kxg3 b3 87. Kf3 Kd3! (the only move to win; 87… b3? 88. Ke2 b2 89. Bf5 Kc3 90. Kd1; 87… Kc3? 88. Ke2 Kc2 89. Bf5+ Kc1 90. Kd3 b2 91. Kc3) 88. Bf5+ Kd2 89. Ke4 Nb5 90. Kd5 Na3 91. Kc5 b2 92. Kb4 Nc2+. In the game, once the white king got to g1 it was never going to be possible to set up the same opportunity.]