Cordingley puzzle 72
White to play and win
Solution
First, a digression: note who the player of the black pieces is. A quick google, and I see that in 1936 he was 20, and coming to full strength. His Wikipedia entry though does say (my emphasis added):
At Helsinki 1935, he placed 2nd behind Paulin Frydman with 6½/8 (+6 −1 =1). He won at Tallinn 1936 with 9/10 (+8 −0 =2). Keres' first major international success against top-level competition came at Bad Nauheim 1936, where he tied for first with Alexander Alekhine at 6½/9 (+4 −0 =5). He struggled at Dresden 1936, placing only 8–9th with (+2 −4 =3), but wrote that he learned an important lesson from this setback. Keres recovered at Zandvoort 1936 with a shared 3rd–4th place (+5 −3 =3). He then defended his Estonian title in 1936 by drawing a challenge match against Paul Felix Schmidt with (+3 −3 =1).[7]
This puzzle took me time to solve. There seemed nothing obvious, indeed no way to get the queen to the a1-h8 diagonal and start using the black squares; nor any way to get through on the e file. So I had to go through 'procedures. First thing I noticed was that the Bd7 was LPDO; and all black's other pieces were defended; second thing is that if I could get to e7, then Rf7 is impossible, because after exchanging rooks, Ne5+ forks king and queen: so, f7 is 'mined'.
Almost for want of anything else, I looked at 1 Re4, which turns out to be the winning move. I spent a lot of time on 1…Qc7 2 Qd4+ and thought white was then better, but as can be seen in the notes below, my calculations were imperfect. I looked at 1..de also, of course, and thought that white had enough to at least get his material back, and might still have an edge. Houdini then told me that white is in fact winning, by quite a neat LPDO.
A nice puzzle.
I also had a very quick look at the earlier moves, to see where Keres went astray: one of his earlier moves, and his opponent's reply, do seem very odd.
Posted with Blogsy
allanbeardsworth 