THE MITROFANOV STUDY – PROBLEM
White to play and win
This is probably one of the most famous chess studies ever. Due to its beauty and brilliancy it is also known as the “Study of the millenium”. Although I am not a very big fan of chess composition, I think studying this one is a must for any self respectable chess player.
In this post we will first post the solution. Then afterwards we will also try to explain why the initial version of the study, with the black knight on f3 instead of g2, is flawed.
White has to prevent the blockade of the pawns by Bc7+.
1.. Ka8 (on 1… Kb8 2 g7 is threatening g8 mate. Now however it is not a threat, since Black can defend with Bb8).
This seems pointless, because White can’t stop the black h pawn.
The point of this rook sacrifice will be apparent soon.
Black can also try 3… Nc4+ but it doesn’t help 4 Kb5 h1Q 5 g8Q + Bb8 6 a7!
And Black doesn’t have a check, since the knight on e1 is blocking the path.
Back to the main line however.
4 g8Q+ Bb8 5 a7!
Since the knight on e1 is blocking the queen’s path, it seems that Black is quite dead. But he has one final resource.
Everything else also loses, and faster as well. You can trust me on that one.
6 bxc6 Qxh5+
The climax of the study has been reached. Black starts checking and wherever White moves his king, he can’t avoid further checks.
But White is not forced to move his king.
The most beautiful, most paradoxical, and most amazing move I have ever seen. White simply leaves his queen to be taken, and with check to, only to disrupt the Black’s queen position.
This position deserved a separate diagram.
7 … Qxg5 8 Ka6!
An amazing position. White threatens b7 mate and Black simply doesn’t have an useful check.
His only defence is giving up the bishop
And now it seems that White has overplayed himself. But now follows another stupendous move.
I don’t know how on earth the author arrived at this position, but I think that this ultimate triumph od mind over matter proves that the chess compositions deserves to be proclaimed as an art.
White is threatening both c8Q and b7 mate, and there is no defence against both threats.
The best Black can do is give up his queen.
9… Qa5+ 10 Kxa5
But now both 10… Kb7 11 bxa7 and 10… Bxb6 11 Kxb6 lose for Black.
MITROFANOV’S BAD LUCK
However, the story about the study doesn’t end here. Because he original variation of the study was given by Mitrofanov with the black knight on f3 instead of g2. Although it is ture that he also gave variations with the knight on g2, and even with the White rook already captured on e1.
I will try to give a brief chronology about how the refutation of the diagram position came to be. The most of the story is taken from Tim Krabbe’s excellent monograph about the study.
1970 – Refutation part one
Initially, it was not untill the year 1970 and the effort of the Soviet master Alexander Kuindzhi that any talk about the refutation existed. But in 1970 he had found a refutation.
In the variation
1 b6+ Ka8 2 Re1
Black doesn’t take the rook, but plays the knight check
3 Kb5 Nxb6
And now White has a choice:
A) 4 g7 Ka7 5 Qg8 Nd4+
And amazingly, White can’t avoid the perpetual:
6 Ka5 Nc4 7 Ka4 Nb6
B) 4 Kxb6 Nxe1 5 g7 h1Q 6 g8Q Bb8 7 Qg7
And all seems perfect, but Black has a miracle geometrical save:
7… Qg1+!! 8 Qxg1 Ba7+
1999 – Refutation part two
The dispute didn’t end here. In September of 1999, the Dutch Champion Ron Kuijf found the “refutation of the refutation”.
In the line A above, instead of 5 Qg8, White plays 5 Kc6!
Kuijf’s main variation was:
5…Bb4 6.g8Q Nd4+ 7.Kc7 Nb5+ 8.Kd8 Bxe1 9.Qg7+ Kxa6 10.Qa1+ Ba5
And this endgame should be considered technically winning.
Did it mean that Mitrofanov was lucky after all?
1999 – slightly later – Refutation part three
Alas, it was not meant to happen with the knight on f3. Because a chess endgame composer under the name of the Harold van der Heijden put a final nail to the coffin with his discovery.
Consider once again the following diagram:
Here, instead of 5… Bb4, attacking the rook, van der Heijden discovered an amazing possibility:
This move is almost like a study inside of a study. Black let’s his bishop be captured with check.
6 Rxe7 Kxa6
And now this is a draw whatever White tries:
A) 7 Ra7+ Kxa7 8 g8Q and now 8… Nd4+ draws (but not 8… h1Q when 9 Qh7+ is a forced mate)
B) 7 Re1 7…Nxe1 8.g8Q h1Q 9.Qf7 Qxd5+ draws
C) 7 g8Q Nd4+! 8 Kc5 h1Q 9 Kxd4
And now after the
9… Qa1+ the rook on e7 is lost. 10 Kd3 or 10 Kc5 run into Qa3, while 10 Ke4 runs into Qe1.
That finally concludes this rather long report. It is amazing how deep this study actually is. The modern chess engines has somewhat killed this romantic stories when multiple generations of chess composers sought the ultimate chess truth. Since the modern computer is more than powerfull of finding all the subtleties of this amazingly beautiful, rich and deep piece of chess composition.