Riddle solved: Réti could have held Rubinstein to a draw

R. Réti vs A. Rubinstein, Gothenburg 1920

In this endgame, the black bishop has an edge since White has a number of weaknesses and Black not only has an active king but also a very mobile minor piece. Black indeed won the game, and Rubinstein’s play seems to be a model for this type of positions, in which the stronger side plays on both sides to seek for a win.

However, Réti could have drawn on move 37.

The duel knight against bishop can be very deep. Often the knight wants a static position and control, while the bishop is better in dynamic setups. Here the bishop is better, of course, but Réti could have defended as late as move 37.

Once again, Zoran Petronijevic sent the best solution, with Charles Sullivan adding a few contributions.

Conclusions:

1. The position after 28…Kxe8 is better for Black in practical terms due to his better pawn structure. After proper play, however, White can equalize.
2. The move 30.g4, contrary to Gelfand’s assessment, does not lose, although White should play a lot of subtle moves in order to make a draw (the main line which leads to a draw was found by Charles Sullivan). The game move, 30.Ke3, is better from a practical point of view.
3. 30...Ke6 is an imprecise move — 30…Bd7 is stronger. Still, White can hold.
4. The move 31.g4 is a good defending idea. Another good idea is an old suggestion by Euwe: 31.d4.
5. The game move 32.h3 is okay. Another good move is 32.f5, which gives a relatively easy draw.
6. The move 33.d4 leads to even play, much like the game move, 33.Kd2.
7. The move 34.Nf3 is okay, as well as Gelfand’s suggestion, 34.Ng2.
8. The move 34…h6 is more challenging than the text. Still, Gelfand proved that White can hold.
9. According to Gelfand, 36.Nh2 loses. His suggestion is 36.Nh4, which leads to a draw. Still, 36.Nh2 is not a decisive mistake — according to my analysis, White can still hold.
10. The decisive mistake is 37.Ke2. White has two moves that lead to a draw: 37.Kd2, suggested by me, and 37.Kf2, suggested by Charles Sullivan. The move 37.a3 also leads to draw, transposing to the line with 37.Kd2.

It is really hard to find many of the subtleties in this endgame. Probably new engines will find a new verdict after analysing this game.

Magical Chess Endgames

In over 4 hours in front of the camera, Karsten Müller presents to you sensations from the world of endgames - partly reaching far beyond standard techniques and rules of thumb - and rounds off with some cases of with own examples.

Links

• Endgame riddle: Did Kasparov miss a chance?
• Riddle solved: Kasparov could have drawn
• Historical riddle: Botvinnik's most subtle rook ending
• Riddle solved: Smyslov could have drawn!
• Historical riddle: Petrosian vs Tal, Curacao 1962
• Riddle solved: Tal could have drawn!
• The historical puzzle: Botvinnik vs Bronstein 1951    
• Solution to the historical riddle Botvinnik vs Bronstein 1951: Bronstein could have drawn
• How sound was the endgame Spielmann-Rubinstein?
• Riddle solved: Spielmann could have drawn!
• Could Petrosian have held against Fischer?
• Petrosian could have held!
• Historical riddle: could Kramnik have defeated the computer?
• Kramnik could have defeated Fritz