Dvoretskys endgame manua.., p.24
Dvoretsky's Endgame Manual, page 24
What if Black could not play …h7-h5 in time? We shall analyze two important endings that may serve as landmarks for both sides: the stronger side may pursue them while the weaker side should avoid these situations.
These endings are thoroughly analyzed in endgame handbooks. We skip some less important lines, limiting ourselves to the conclusions.
Botvinnik – Najdorf
Moscow ol 1956
1 Ra5 Rc7 2 Rd5 Ra7 3 e5 fe 4 fe (5 Rd7+! is threatened) 4…Ke7 5 e6 Ra4!
5…Ra6 6 Rd7+ Kf8 7 Kg6! R×e6+ 8 Kh7 is quite bad for Black.
6 g5!
6 Rd7+ Kf8 7 Rf7+ Kg8 8 g5 fails in view of 8…Ra5+! (8…hg? 9 Kg6) 9 Ke4 Ra6=.
6…Ra7!
The best defense, as suggested by Aronin. The rook may return because there is no danger of trading the rooks any longer: 7 Rd7+? R×d7 8 ed K×d7 9 Kg6 hg 10 K×g7 g4=.
The actual continuation was 6…hg?! 7 Rd7+ Kf8 8 Rf7+ Kg8 9 Kg6 g4 10 h6! (the shortest way to a win) 10…gh 11 e7 Ra8 12 Rf6 ( Rd6-d8) Black resigned.
7 Re5!
A key move! White protects the pawn and prepares a king invasion.
An anticipatory pawn exchange is wrong: 7 gh? gh 8 Rb5 ( Rb6) 8…Rc7! 9 Rb6 Rc5+! 10 Kg6 Re5! 11 K×h6 Kf6! 12 Ra6! Rf5!.
In this position, Black must play very precisely in order to achieve a draw, but theory says that this goal is within his reach.
7…hg
7…Kd6 8 gh gh 9 Kf6; 7…Ra6 8 Kg6 Kf8 9 Kh7 hg 10 e7+ Ke8 11 K×g7 g4 12 h6+–.
8 K×g5 (8 Kg6 Kd6 9 Re1 g4 10 h6! gh 11 Kf6 is also strong) 8…Ra1 9 Kg6 Rf1 (9…Rg1+ 10 Rg5) 10 K×g7 Rg1+ 11 Kh6! Rg2 12 Rg5+–.
Capablanca – Yates
Hastings 1930/31
White’s position is winning (the same evaluation is valid with the black pawn on h7 and the white pawn on g5). The winning plan is a rook transfer to the eighth rank followed by f4-f5-f6+. If the black rook aims at the e5-pawn, White defends it with the rook from e8.
Capablanca carried this plan through; however, as renowned rook endgame expert Kopaev demonstrated, both sides made errors along the way.
1 Rb6?!
White should have played 1 Rd6! in order to use the rook to protect against checks from the side. Now Black could reply 1…Ra4! 2 Kf3 (2 Kg3 Ra3+ 3 Kh4 Ra4 4 f5 Ra5 5 e6 fe 6 fe Kf6=) 2…Ra3+ 3 Ke4 Ra4+ 4 Kf5 Rc4 5 Rb7 ( 6 e6) 5…Kg8! (if 5…Rc5, then 6 Rd7 and 7 Ke4).
Kopaev had considered 5…Kf8, after which the Armenian Vardan Pogosyan suggested 6 e6! fe+ 7 Ke5, and came to the conclusion that Black must inevitably lose one of his pawns. The endgame that has arisen is lost for him.
With the king on g8, this idea is less effective in view of 7…Rc5+ 8 Kf6 (if the king were on f8, this move would win) 8…Rc4=, or 8 K×e6 Rc6+ and this version of the “two-pawns-versus-one-pawn” ending is drawn. Let me remind you that we have already dealt with a similar position in the game Smyslov-Keres (diagram 9-85).
6 g5 hg 7 K×g5 Rc5! is not dangerous (with the king on f8, the simplest would be Kopaev’s recommendation 7…f6+!, which is not possible here because of 8 Kg6!+–).
Having occupied the f5-square with his king, White makes the implementation of the strongest plan – the preparation of f4-f5 – more difficult. Now he should have revisited this idea by playing 6 Rb3! Kg7 7 Re3 Rc6 (7…Ra4 8 Re4 Ra1 9 Rd4) 8 Ke4 Rc4+ 9 Kf3 Rc6.
To this point, we have been following Kopaev’s line which demonstrates that 10 f5 Kf8 11 Ra3 Rc1 12 Ra8+ Ke7 13 f6+ Ke6 14 Re8+ Kd5 15 e6 Rc6 leads to a draw. But White’s play can be strengthened. For example, after 10 Ra3 f6 11 Ra7+ Kf8, White does not play 12 Ke4 fe 13 fe Rc1 (Averbakh), but 12 ef! R×f6 13 Kg3 Kg8 14 Ra4 followed by Kh4-h5 (the only way to prevent the seizing of the key h5-square – the transfer of the rook to h1 – is too late here). A similar continuation was suggested by Jonathan Speelman, 10 Kg3 Ra6 11 Rb3: we are already familiar with the follow-up 11…f6 12 Rb7+ Kf8 13 ef, while on 11…Rc6 there follows 12 Rb8 Rc3+ 13 Kg2 Re3 14 Re8+– and f4-f5 is inevitable.
1…Re3?! 2 Rb4
2 Rb8 suggested itself. However, after 2…Re4 3 Kf3 Re1, the straightforward 4 Re8? Allows salvation by 4…h5! 5 g5 (5 gh Rf1+! 6 Ke4 Re1+ 7 Kf5 Rh1) 5…Rf1+ 6 Ke3 h4.
The most precise is 2 Rb1! (temporarily denying the black rook the first rank). Black is in zugzwang. He must either worsen his king’s position or move his rook off the e-file where it is best placed. In both cases, the invasion of the white rook becomes more effective. For example, 2…Re4 (2…Re2+ 3 Kf3 Rh2 4 f5 h5 5 Rb7 hg+ 6 Kg3 Rh5 7 K×g4 Rh1 8 e6+–) 3 Kf3 Ra4, and now time has come for the main plan: 4 Rb8! Ra3+ 5 Kg2 Re3 6 Re8! Re2+ 7 Kf3 Re1 8 f5 Rf1+ 9 Ke2 Rf4 10 Ke3 R×g4 11 f6+ Kh7 12 e6+–.
2…Rc3 3 Kf2? (3 Rb8) 3…Ra3?
Both adversaries missed the fact that after 3…h5!, Black either trades a pair of pawns (4 gh Rh3) or, (in case of 4 g5 h4) obtains enough counterplay to save the game.
4 Rb7?! (4 Rb8!) 4…Kg8?! (4…Ra2+!?) 5 Rb8+! Kg7 6 f5 ( 7 f6+), and White won.
Korchnoi – Antoshin
USSR ch, Erevan 1954
1 h5!
Black, if he was on move, could have considerably simplified his task by advancing his own pawn to h5. If 1 g4?!, then all the same 1…h5!.
1…Ra5!
As Korchnoi noted in his exceptionally deep and far-reaching comments to this endgame, it is useful for Black to force the advance g3-g4.
2 g4
2 Rc8+ Kh7 3 g4 is not dangerous yet because of 3…g5! 4 hg+ K×g6 ( 5…h5) 5 f4 f6!.
The last move is worth special attention. It is vitally important for Black to prevent the pressing advance e4-e5 that leads to the setup from the Capablanca-Yates game. By the way, in that game the white pawn stood on e4 a few moves before the position of the diagram 9-119 arose, and Yates could have had an easy draw by means of …f7-f6.
2…Ra7?
Antoshin had to keep in mind the danger of a check along the eighth rank: his king, when standing on h7, is too far removed from a passed e-pawn if White manages to create it. Therefore here, as well as later on, he should have played f7-f6!. Black could then parry the threat of transposition into the winning position from the Botvinnik-Najdorf game by means of catching the white pawns from behind. A characteristic variation was demonstrated by Korchnoi: 2…f6! 3 Rc8+ Kf7 4 Rc7+ Kg8! 5 Kf3 (5 f4 Ra3 – White’s king is cut off from its pawns) 5…Ra3+ 6 Kf4 Ra2 7 Kf5 Kh7! (this is why the black king pulled back to g8) 8 f4 Rf2!=, as White fails to create the passed e-pawn.
It is time to explain why the move 1…Ra5 was given an exclamation mark. With a pawn on g3, this defensive plan does not work: the f4-pawn is protected, so White can play Ke6, while Black can hardly prevent the penetration of the white king to f5 (via g4) at an earlier stage.
3 Rc6?
After 3 Rc8+! Kh7 4 e5! ( Kg3, f4, Re8+–), Black would have been faced with problems one can hardly tackle over the board.
As Korchnoi showed, almost all defensive methods are doomed to lose: White either creates a dangerous passed e-pawn or transposes to positions from the game Capablanca-Yates. For example, 4…g5? 5 hg+ K×g6 6 f4+–, or 4…Ra4?! 5 Kg3 Re4 6 Re8 g5 (6…Ra4 7 Rf8 Ra7 8 f4+–; 6…f6 7 e6 g5 8 hg+ K×g6 9 f3 Re1 10 Kf2 Re5 11 f4+–) 7 hg+ K×g6 8 f3!+– (8 f4? is not precise; Black holds after 8…h5!). Of course, only basic results are shown here, as a detailed explanation would have been rather complicated and too vast.
4…Re7! (the only defense) 5 f4 f6! 6 Rc5 fe 7 fe, and now Black must prevent the white king’s march to the center, that would transpose to the Botvinnik-Najdorf ending, by 7…Rf7!. Here Korchnoi gives 8 Kg3 g6!= and 8 e6 Re7 9 Rc6 g6 10 Rd6 ( 11 Rd7) 10…Re8 11 Kf3 gh 12 gh Kg7 13 Ke4 Kf6 14 Kd5 Re7=.
The prophylactic move 8 Rd5! is more dangerous for Black. He cannot play 8…g6? on account of 9 e6 Rf8 (9…Re7 10 Rd7) 10 Rd7+ Kg8 11 e7 Re8 12 Rd8 Kf7 13 R×e8 K×e8 14 g5!+–, and 8…g5? 9 e6 Rf8 10 e7 Re8 11 Re5 Kg7 12 Re6+– are also bad. Black also loses in the line 8…Rf8 9 Kg3 Rf1 10 Rd7! (10 Rd3 g5!) 10…Rf8 (on 10…Kg8 the same reply follows) 11 Rd3 g5 12 Rf3 Re8 13 Rf5 followed by 14 Kf3. Nevertheless, as Pogosyan has shown, he manages to save himself by continuing either 9…g5! (instead of 9…Rf1?) 10 Rd3 Kg7 11 Rf3 Re8 12 Rf5 Ra8, or 8…Rf4!? 9 Kg3 Ra4 10 Kf3 Kg8.
3…Ra3? (3…f6!=) 4 f3?
4 Rc8+! Kh7 5 e5! was winning.
4…Ra5!? (4…f6!)
Now White can gradually strengthen his position by means of Rc8-d8, Kg3-f4 or f3-f4, but, as his pawn cannot get to e5, the game will be drawn if Black defends precisely.
Korchnoi decided to force matters and was successful, but only as a result of a new mistake by Black.
5 Rc8+ Kh7 6 f4?! ( 7 e5+–) 6…Ra2+ 7 Kf3 Ra3+ 8 Kf2 Ra2+ 9 Ke3 Ra3+ 10 Kd4
Black can hold the game rather simply: 10…Rf3! 11 Ke5 (11 f5 f6=) 11…f6+ 12 Kf5 Rf1, achieving the position from the note to Black’s move 2.
10…Rg3?
He chases after material gain but lets White create a passed pawn that will cost him a rook.
11 Rf8! f6 12 e5! R×g4
12…fe+ 13 fe R×g4+ 14 Kd5 Rg1 15 e6 Rd1+ 16 Kc6 Re1 17 Kd7 Rd1+ 18 Ke8+– is no better.
13 e6 R×f4+ 14 Kd5 Rf5+ 15 Kd6 R×h5 16 e7 Re5 17 e8Q R×e8 18 R×e8
The fight is almost over. When the white king comes back to his home side of the board, the rook will be stronger than three pawns.
18…Kg6 19 Kd5 Kf5!? 20 Re1
20 Kd4 Kf4 21 Kd3 Kf3 22 Rg8 g5 23 Rf8+– is also strong.
20…h5 21 Rf1+ Kg4 22 Ke4 g5 23 R×f6 h4 24 Ke3 Kg3 25 Ke2 g4 26 Kf1 Kh2 27 Rf4 h3 28 R×g4 Kh1 29 Kf2 h2 30 Kg3 Kg1 31 Kh3+ Black resigned.
Tragicomedies
The two last endings fully fit this category, but I would like to add some new examples, the last of which has some theoretical value.
Bellón – Chekhov
Barcelona 1984
The waiting policy (1 Ra5 or 1 Rb7) gave a rather easy draw, but Bellón decided to chase after the g7-pawn.
1 Rb8+ Ke7 2 Rg8?? Rd8!
White resigned. The pawn endgame is quite hopeless for him, while after 3 R×g7, his rook is lost: 3…Kf8 4 Rh7 Kg8 5 R×h6 Kg7 6 Kg5 Rd5+.
Norri – Svidler
Erevan ol 1996
Here again White did not have enough patience and made an analogous error.
1 f5? (1 Ra7=) 1…gf 2 Rh7 Kg8 3 R×h5??
3 Ra7 could still hold the game.
3…Re5 4 Kf3 f4! White resigned.
Piket – Kasparov
Internet Tournament 2000
White has succeeded in bringing his pawn to e5 (generally speaking, it would have been favorable for Black if he prevented this by playing …f7-f6 at an earlier stage). On the other hand, …h6-h5 has already been played, so reaching a draw should not be a very difficult problem.
Kasparov had to decide how to react should the white king march to g5 via h3 and h4. The simplest method was to play …Kh6 at the proper moment. For example, 1…Kg7 2 Kh3 Ra7 3 Kh4 Kh6!, and one cannot see how White could make any progress.
Moreover, a king invasion to g5 is not too dangerous. Even with the white rook on the seventh rank Black can survive. Averbakh analyzes 3…Ra6 (instead of 3…Kh6) 4 Rc7 Rb6 5 Re7 Ra6 6 Kg5 (6 e6 Kf6! 7 R×f7+ K×e6=) 6…Ra5! (as Bologan says, even 6…Rb6!? 7 e6 Rb5+! or 7 f5 gf does not lose) 7 f5 gf 8 e6 (8 K×h5 Kf8 9…R×e5) 8…f4+! 9 K×f4 Kf6 10 R×f7+ K×e6=.
1…Rd3?!
In many similar situations, to place the rook behind the e-pawn makes some sense. In particular, such a maneuver is not bad when h-pawns are absent. But here this transfer is wrong. Its slightly modified version does not work either: 1…Rd4?! 2 Kh3 Re4? (in case of 2…g5?, White does not play 3 fg Kg6, he has 3 Rc7! instead) 3 Rc7! (3 Kh4 Kh6 4…g5+) 3…Kg7 4 Kh4 Re2 5 Kg5!, and we come to situations that have actually occurred in the game.
2 Kh3 Re3?
2…Rd7 3 Kh4 Kh6= was necessary.
3 Kh4?!
Playing 3 Rc7! Kg7 4 Kh4, Piket could have chained the hostile rook to the e-file and, as we shall see, this was a winning method.
3…Kg7?
He should have tried 3…Kh6! 4 Rc7 Re2!. If 5 Kh3, then 5…Kg7 (5…g5!? is also playable) 6 Rb7 g5! 7 fg Kg6. The line 5 g4 hg 6 R×f7 R×h2+ 7 K×g4 is more dangerous for Black, but after 7…Re2, he seems to be surviving.
4 Kg5?
An erroneous order of moves; again 4 Rc7! Re2 5 Kg5 is correct. Now Black could return to Averbakh’s plan: 4…Ra3! 5 Rc7 Ra5=. However Piket could hardly expect that his opponent would suddenly change his mind and move the rook back.
4…Re1? 5 Rc7 Re2 6 Re7! Ra2 7 f5! gf 8 e6 h4 9 R×f7+ Kg8 10 Kf6 Black resigned.
Let us look at 6…Re4 (instead of 6…Ra2).
This position has occurred in the following games: Stean-Hartston (Great Britain ch, Brighton 1972), Ionov-Karasev (Leningrad 1983) and Matveeva-Rappoport (Baku 1983). In all these games, White found a forced win.
7 e6! R×e6 8 R×e6 fe 9 h3 Kf7 10 Kh6 Kf6 11 g4 h4 (11…hg 12 hg+–) 12 g5+ (12 Kh7? g5) 12…Kf5 13 Kg7 K×f4 14 K×g6 e5 15 Kf6! e4 16 g6 e3 17 g7 e2 18 g8Q e1Q 19 Qg4+ Ke3 20 Qe6+ Kf2 21 Q×e1+ K×e1 22 Kg5+–
Finally, instead of 5…Re2, Black could have played 5…Re4!? at once. The point is to meet 6 Re7 with 6…Ra4!, and 7 f5, as was played by Piket, is not possible anymore, while if 7 e6, then 7…Ra5+ 8 Kh4 Kf6 9 R×f7+ K×e6=.
White must wait: 6 Rb7! Ra4 (6…Re2 7 Re7! transposes to the actual course of the game).
And here Bologan has discovered a brilliant solution: a double pawn sacrifice 7 g4!! hg (7…Re4 8 gh gh 9 h4 is hopeless) 8 f5! gf 9 e6+–.
Winning another pawn is much weaker: 7 e6 Ra5+ 8 Kh4 Kf6 9 ef Kg7 10 Re7 Kf8! 11 Re5 Ra2, or 10 h3 Kf8 11 g4 hg 12 hg Rc5 13 f5 Rc6! (rather than 13…gf? 14 g5+–). It looks like Black holds in both these lines.
Balance on One Wing and an Extra Pawn on Another
Situations with an extra distant passed pawn occur now and then, therefore it is very important to learn their correct evaluation and handling. The decisive factor in this sort of endgame is the position of the rook of the stronger side. In the majority of cases, the rook is placed best “à la Tarrasch,” behind its own passed pawn; sometimes its sideways position is preferable.
Quite often, however, we lack free choice, so the rook mostly stands in front of the pawn in practical games. Therefore we shall pay more attention to these cases.
The Rook Behind Its Own Pawn
Botvinnik – Boleslavsky
Leningrad/Moscow 1941
1 Rb1!
The rook has occupied its correct position, behind the pawn. After 1 h3? Rb2! 2 Re4, Black could have achieved a draw.
1…Kf7?
The passed pawn should be blocked as soon as possible. Black had to play 1…Rc6! 2 b5 Rb6. I do not think this was enough for a draw but, anyway, his opponent would have then been faced with more complicated problems. After a king’s march to the queenside, Black removes his rook from b6 either for protecting his own pawns or for attacking the hostile ones.
2 b5 Ke6 3 b6 Rc8 4 h3
4 b7? Rb8 is wrong because it allows Black to eliminate the b-pawn and thereafter to bring his king back to the kingside in time. For example, 5 Kg1 Kd6 6 Kf2 Kc6 7 Ke3 R×b7 8 R×b7 K×b7 9 Ke4 Kc6 10 Ke5 Kd7=.
4…Rb8 5 Kh2 Kd5
If the black king remains with his pawns, his adversary heads to the b-pawn. Black cannot prevent this by means of the opposition because White can make a waiting rook move. Black will then be obliged to give way to the white king because his rook has no waiting moves. This clearly demonstrates the difference between the rook positions.
6 Kg3 Kc6 7 Kg4 Kb7
A capture on b6 is impossible now. Therefore, Black blocks the pawn with his king, releasing the rook from this duty. A standard and often quite useful method, but alas, it does not bring any relief to Black in this particular case.
8 Re1!
Excellently played! While the rook was pinning the black rook down it was superbly placed on b1, but now it will be more active when placed sideways. In case of 8…K×b6 9 Rb1+, Black loses the pawn endgame.
8…Rg8 9 Re6 Ka6 10 Kg5 Kb7 11 h4
The rest is simple. White attacks on the kingside, having an extra piece there.
11…Ka6 12 h5 Kb7 13 g4 Ka6 14 Kh4 Kb7 15 h6 gh 16 R×h6 Rg7 17 Kh5 ($ g5, Re6, Kh6+–) 17…Ka6 18 Rc6 Re7 19 Rc7 Re5+ 20 g5 K×b6 21 R×h7 Kc6 22 Kh6 Kd6 23 g6 Re1 24 Rf7 Ke6 25 Rf2 Ra1 26 g7 Rh1+ 27 Kg6 Rg1+ 28 Kh7 Rh1+ 29 Kg8 Ke7 30 Re2+ Kd7 31 Re4 (“bridging”) 31…Rh2 32 Kf7 Black resigned.
Botvinnik – Borisenko
USSR ch, Moscow 1955
1 Ra4!
Botvinnik blocks the pawn immediately. If he had allowed …a5-a4, he would have had no chances at all. Alekhine won a similar ending from Capablanca in the last, 34th, game of their match for the world championship in 1927: it can be found in almost every book on endgames.
1…Kg5
The king heads for the queenside, but a safer road was via g7. The text move was regarded as an error by all the commentators. Why? The point is that the best chance for a successful defense in this sort of position is counterplay on the kingside by creating a passed pawn or weakening the opponent’s position. And the position of the king in front of the pawns makes it easier to create counterplay.
Let’s see how the fight takes shape after 1…Kg7!?:
(a) 2 f3 Kf7 3 g4 h4–+;
(b) 2 Kf1 Kf7 3 Ke2 Ke6 4 Kd3 Kd5 5 Kc3 Kc5 6 Kb3 Kb5 7 Re4 a4+ 8 Ka3 Ra6 9 h4 Kc5 10 Re2 Kd4 11 Re3 g5–+ (Marin);
(c) 2 h4 Kf7 3 Kf1 Ke6 4 Ke2 Kd6 5 Kd3 (5 g4 hg 6 R×g4 a4 7 R×g6 a3 8 Rg1 a2 9 Ra1 Ke5 10 Kf3 Ra4–+) 5…f5! (wrong is 5…Kc6? 6 g4 Rd7+ 7 Kc3 Rd5 8 Rf4 f5 9 gh gh 10 Kc4=) 6 f3 Kc5 7g4 Kb5 8 Rd4 a4 9 Kc2 a3 10 Kb1 Ra4! 11 Rd6 hg 12 R×g6 gfo (Kopaev);
