Year Book 2019-20 Contents
Thing of the Day
(Click on underlined link) \/ to end of list \/
Tale of Two Endings
(One being the Tail of the Other)
In yesteryear, and in living memory for most, before incremental time controls, even before rapid-play finishes, there were adjudications. In some ways this was a good thing in that people discussed endings together and perhaps learnt things from each other. The weaker player would hope that, in post ludum analysis, a stronger member of his team could demonstrate what that weaker player hoped: that the inferior ending was in fact “a draw”, or that the apparent advantage could in fact be turned by force into a win. In the absence of agreement, then the position was sent for adjudication.
It is not hard to imagine the “analysis” which would be hammered out in the following position. Perhaps Black had had a knight on a8, and allowed the white king to capture it while the h-pawn broken cover.
White to move. How should play proceed, and with what result?
The players eventually settle on
1. Kxg7 h4 2. Kg6 h3 3. Kg5 h2 4. Kg4 h1=Q.
So, Black wins. No, wait.
5. Kg3! (threatening 6. Ra1#, or winning the queen after 5. … Kf1 6. Ra1+.).
So, White wins after all, as sacrificing the queen with check only buys brief reprieve.
5. … Qa8!
Black prevents the mate and saves the queen. So, the initial position is a win for Black, . . . .or was it?
In the initial position there was a black pawn at g7, which in the “analysis” White eagerly captured, but, with the retention of which, Black’s 5. … Qa8 would be to no avail! So, we have to conclude that White wins with
1. Kh7! h4 2. Kg6 h3 3. Kg5 h2 4. Kg4 h1=Q 5. Kg3!
At this point, but before the win was down on the match result sheet, an onlooker who hitherto had been strangely quiet, being lost in contemplation, might pipe up with, “What if Black under-promotes to a knight?”, causing White to look daggers at the interjector. After 1. Kh7! h4 2. Kg6 h3 3. Kg5 h2 4. Kg4 h1=N (to prevent 5. Kg3) White still wins with 5. Kf3. Faced with losing options 5. … Kf1 6. Ra1#, or 5. … Ng2/f2 6. Rxg3/f2, Black has to move his remaining pawn, but White wins after 5. … g5 (or 5. .. g6 6. Ra5, continuing essentially as the text) 6. Ra4 g4+ 7. Rxg4 Kf1 (or 7. … Kh2 8. Rg8 either mating of winning the knight) 8. Rg2 Ke1 (or 8. … Ng3/Nf2 9. Rxg3/f2) 9. Rg1+ winning the knight.
“But,” says the original interjector, “suppose Black defers promoting the h-pawn and pushes the g-pawn instead?” This turns out better, as follows: 1. Kh7! h4 2. Kg6 h3 3. Kg5 h2 4. Kg4 g5 5. Kg3 (not 5. Kxg4 allowing 5. … h1=Q, drawing) 5. … h1=N+ (delivering the lifesaving check) 6. Kf3 g4+ (otherwise White mates or wins the knight) 7.Kxg4 and now the black knight bolts out of the stable, heading for the wide open spaces with 7. … Nf2+, as in the following diagram.
An ending with K+R v K+2P has now turned in one with K+R v K+N. We all know, or should know, that in the ideal situation, a K+N at the edge of the board can force a draw. Happily for White, this particular position is still a win for White!
After 1. Kh7! Black might instead try pushing the g-pawn giving the following main line:
1. Kh7! g5 2. Kg6 g4 3. Kg5! (leaving Black with a pawn move so that stalemate does not arise later) 3. … g3 4. Kh4 g2 5. Kh3 Kh1 6. Rxg2 (not stalemate because the h-pawn has a move) and White wins.
The initial diagram is in fact a study by Czech composer Josef Moravec (b. 20/03/1882, d. 29/08/1969), which appeared in La Stratégie in 1913.
And now for a humorous postscript:
In a Sheffield congress once, a certain player from the North East had K+R against K+N, the latter pair being on their back rank, but nearer the middle files. The player with the knight, after getting down to less than 2 minutes on the clock, lodged a claim for a draw under the rules as they then were. The controller asked the players to continue while he watched, which they did. The knight-owner able demonstrated his ability to defend in the prescribed way, while the rook-owner demonstrated no new ideas, so the controller upheld the claim for a draw. The rook-owner did not take this well and aggressively told the controller, “This would go down on your record.” (As the controller, who was the present writer as it happens, was not an ECF arbiter as such, he would have no “record”.) For the rest of the congress, when time was available, in the analysis room or wings of the main playing area, the rook-owner continually analysed his ending, yet seemingly failed to find a forced win for the rook-owner! The position was essentially as the following, with White to move.
White to move, but only draw.
(See any elementary book on the endings.)