Euwe vs Geller, Zurich 1953

Here is a position from the game Euwe – Geller, Zurich (ct) 1953. It is on the menu today because it shows that the late middlegame is often characterised by inaccurate play, especially when the time control is approaching.

Black has the initiative and is trying to turn it into a decisive attack on White's king, which is the more exposed of the two. First, let's see how the game went:

55...Be3 56.Rf1 Qd2 57.Rf7 Qxe2+ 58.Kg3 Qe1+ 59.Kf3 Qh1+ 60.Kg3 Qg1+ 61.Kf3 Qf2+ 62.Ke4 Re8+ 63.Re7 Qh4+ White resigned (0-1).

The first impression is that the game took a normal course and ended in victory for Black. David Bronstein, for example, writing in his book Mezhdunarodny Turnir Grossmeisterov, does not query any of the last eight or nine moves. However, a closer examination reveals a dramatic exchange of blunders.

Peelback No. 1

Instead of 56.Rf1?, White should play 56.Rf7!, setting up the threat of 57.Rxb7 mate. Now Black is forced to work with checks; there is no time for a quiet move. Thus: 56...Qg1+ 57.Kf3 Qf1+ 58.Kg3 Bf4+ 59.Kg4. Bronstein reaches this position in his analysis and gives 59...h5+! “et cetera,” with the clear implication that Black is winning.

I always get suspicious when I see a note ending in “et cetera.” I like to carry on the analysis until the win is obvious. In this case there is a problem. After 59...h5+ 60.Kh4 Qf2+ 61.Kxh5 Qxe2+ 62.Kh4 Qe1+ 63.Kh5! there is no forced mate, no forced win of material, and no time for a quiet move. Black can even go wrong: 63...Qe5+?? 64.Qxe5 Bxe5 65.Nd7+ and wins. The first conclusion: 56.Rf7 draws. The second conclusion: 56.Rf1? is a weak move. The third conclusion: 55...Be3 is probably a weak move too.

Peelback No. 2

There is no good reason for Black to allow the move Rf3-f7 until he is ready with a winning continuation. Black can play a move that takes the square g4 away from the White king, so that if White does nothing significant, the attack beginning with Bf4-e3 will be far more dangerous. The move is:

55...h5!

Now White has:

A. 56.Qf7 Bc7! (changing direction in order to give Black's queen access to g5 in some lines) 57.Nd7+ Ka7 58.Nf8 Qg5+ 59.Kf2 Bh2 (also possible is 59...d3 60.Qxc7 Qh4+ 61.Qg3 Qd4+ 62.Kg2 d2 63.Rf1 d1Q 64.Rxd1 Qxd1 and Black is winning) 60.Qg6 Bg1+ 61.Kf1 Qc1+ 62.Kg2 Rd5 63.Ne6 Re5 64.Nf4 Be3 65.Rf1 Qd2 66.Qf7 Bxf4 67.Qxf4 Rxe2+ 68.Kg1 Qxb4 with a winning ending for Black;

B. 56.Qg6 Qd2 57.Rf2 Bc7 58.Qg7 Qe3 59.Qg6 Ka7 60.h4 Qe7 61.Nc4 (unavailing is 61.Qxh5 Rg8+ 62.Kf1 Qe4 and wins) 61...Qxh4 62.Nd6 Qe7 63.Qf6 Bxd6 64.cxd6 Qxd6 65.Qxd6 Rxd6 66.Rf5 Rd5 67.Rf4 d3 68.exd3 Rxd3 69.Rc4 Rd5 70.Kh3 b6 71.Rxc6 bxa5 72.bxa5 Rxa5 and White will have to give up his rook for Black's a-pawn.

I believe the moves 55...Qd2 and 55...Re8 also lead to a win for Black, but the pawn move is less complicated. The conclusion is that Black's move 55...Be3? threw away the win, but White's move 56.Rf1? gave it back.

When preparing his book The Application of Chess Theory for publication, Efim Geller found Bronstein's error, but did not notice his own error.

Marshall vs Alekhine, Bradley Beach 1929

A couple of weeks ago, Eduardo Moura and I were analysing some tournament games played by Alexander Alekhine. One in particular – a queen ending – caught our eye because of the way Alekhine managed to fashion a win out of very thin material. The opponent was Frank Marshall, and the venue was the Bradley Beach tournament of 1929. (This, by the way, was Alekhine's first official event following his defeat of Capablanca in 1927.) Here is the game with annotations by Soviet GM Alexander Kotov, taken from Volume 2 of his book Shakhmatnoe Nasledie A.A. Alekhina:

In queen endings, as in other types of endings, an outside passed pawn sometimes plays an important role. Thus in the following game of Alekhine's, White's queen was deflected by a strong Black passed pawn on the a-file while White's own passed pawn on the d-file was easily held in check by the centralised Black king.

1.d4 Nf6 2.Nf3 e6 3.e3 c5 4.Bd3 d5 5.c3 Bd6 6.O-O Nbd7 7.Nbd2 O-O

A well-known variation of the Colle System. This opening was once considered rather threatening, but after several equalising methods were found for Black it was eventually reassessed as insufficient for advantage.

8.e4

White has to acquiesce to an isolated pawn on d4 because the preliminary exchange on c5 does not work. It was precisely for this reason that Black developed his knight on d7.

8...cxd4 9.cxd4 dxe4 10.Nxe4 Nxe4 11.Bxe4 Qb6!

This move is considered strongest. Black attacks the b-pawn and prevents the development of White's bishop on c1.

12.Ne5

This advance is tactically justified by a double attack: 12...Bxe5 13.dxe5 Nxe5 14.Bxh7+ followed by 15.Qh5+. However, stronger was the natural 12.Qe2 followed by 13.Be3 or 13.Bg5.

12...f5

Decisively preventing any White attempt to attack on the kingside. In the sequel Alekhine skilfully defends the resulting weakness on e6.

13.Nxd7 Bxd7 14.Bf3 Bc6 15.Re1 Kh8! 16.Bg5

Not falling for the trap 16.Rxe6? Bxf3 17.Qxf3 Bxh2+.

16...Rae8 17.Rc1

This enables Alekhine to seize the initiative and obtain the better position. More promising was 17.Bxc6.

17...Bxf3 18.Qxf3 Qxb2 19.Rb1 Qxa2 20.Rxb7 Rb8 21.g3

Marshall makes an escape square for his king so that he can follow up with 22.Reb1, which at the moment is not possible because of 22...Qxb1+! However, this pawn move gives Alekhine the opportunity to exchange bishops and at the same time inflict serious damage on the pawn cover in front of Marshall's king.

21...f4! 22.Bxf4

Capturing on f4 with the pawn is worse because of 22...h6.

22...Bxf4 23.gxf4 Rxb7 24.Qxb7 Qd2

The heavy piece ending is favourable for Black. All four of White's pawns are isolated and weak, and his king is exposed. However, the reduced material does not allow Alekhine to decide matters in the so-called “fourth stage” of the game [i.e., a mating attack in the endgame – tr.]. Eventually he has to agree to an exchange of rooks.

25.Qe4 h6 26.Qe3 Qb2 27.Qe5 Rf6 28.Re3 Qb1+ 29.Kg2 Qb7+ 30.Kg1 Qf7 31.Qb8+ Kh7 32.Qb1+ Rf5 33.Qe4 Qf8 34.Re2 a5

After some extensive probing manoeuvres, Alekhine finally sets his passed pawn in motion. This forces Marshall to start thinking about creating his own passed pawn.

35.Qxe6 Rxf4 36.d5 Rd4 37.Re4 Rd2 38.Re2 Rxe2 39.Qxe2 Qf5

On the surface the queen ending appears to be absolutely equal, but this is not the case at all. There is no doubt that Black stands better. In the first place, his king is much more secure than White's, although this is not the decisive factor. The main thing is that Black's passed pawn is the cause of many worries for White's queen, distracting her from any sort of attack on Black's kingside. Meanwhile, White's pawn on d5 will be held in check by Black's soon-to-be centralised king. It is interesting to follow how the World Champion is able to realise an apparently insignificant advantage against such a strong player as the American grandmaster.

40.Qd1 Qg6+ 41.Kf1 Qa6+ 42.Kg2 Qd6 43.Qd4 Kg8

The king sets off toward the pawn on d5 in order to assist the queen in either holding up this dangerous pawn or even capturing it.

44.Qc4 Kf7 45.Qb5 Qd8 46.f4

In order to deprive the Black king of the square e5 in case he decides to settle on d6.

46...Qc7 47.Kf3 Qc3+ 48.Kg4

Here is the first indicator of the power of the outside passed pawn. The White king cannot go to e4 because then 48...Qb4+ would lead to an easily-won pawn ending.

48...Ke7 49.Qb6

Marshall does not notice the World Champion's clever tactical trick. He had to play 49.h4, preventing the following “hammer blow” by the Black pawn. But even then 49...Kd6 would leave Black with reasonable winning chances.

49...h5+!

This forces White's destruction within a few moves.

50.Kxh5

White has to take; otherwise Black forces the exchange of queens with a check on f6.

50...Qh3+ 51.Kg5 Qg2+ 52.Kf5

Or 52.Kh4 Qxh2+ 53.Kg4 Qg2+ 54.Kh4 Qxd5 and with his extra pawn Black must win.

52...Qxd5+ 53.Kg6

White cannot avoid the pawn ending in any case.

53...Qd6+ 54.Qxd6+ Kxd6 55.f5 a4 56.Kxg7 a3 57.f6 a2 58.f7 a1=Q+ 59.Kg8 Qg1+

White resigned.

With this we conclude our extract from GM Kotov's book.

I will only add the observation that Alekhine was particularly skilled in his conduct of queen and pawn endings. Anyone who looks up his games will find many similar examples.

A late tactic

Here is rook and pawn ending of a type that occurs frequently in blitz chess. With both sides about to promote a pawn, the general result is a draw. One side will give up his rook for the opponent's pawn, and will then force his opponent to do the same.

I am White in the diagram position and my last move was 57.g7. Black has several ways to force a draw, the simplest of which is 57...Rh4+ 58. Kg8 b2 59.Rb8 Rb4 60.Rxb4 Kxb4 61.Kf8 b1Q g8Q and there is no play left for either side.

However, my opponent apparently believed there was absolutely no danger and that he could play anything he liked. That's how he came up with:

1...Kc4?

A golden opportunity for White!

2.Rf4+! Rxf4

Unfortunately forced.

3.g8Q+

Black is now on the horns of a dilemma. If 3...Kb4 then 4.Qb8+ picks up the rook with an easy win. The same goes for 3...Kc3 4.Qg3+. All other moves (except 3...Rf7) allow 4.Qxb3, and White reaches the winning ending of queen vs rook. There was nothing left for Black but resignation.

The lesson here is that no matter how harmless a position may appear, it is still possible to fall into a trap!

J.Yoos vs L.Davies, Vancouver 2009

Today we will analyse an interesting ending from the game J.Yoos-L.Davies, played in the sixth round of this year's B.C. Championship tournament in Vancouver.

In the first diagram White is about to make his 39th move. Although White has the nominal advantage of bishop vs knight, Black's position does not present any cause for alarm. He is well-centralized and has the entry points under control. Indeed, White's bishop is somewhat short of targets, and will have to content itself with restricting Black's movements in the hope that his colleague the king can make something happen. But at the moment that does not look too likely.

39.Bf5 Ne8 40.Be4+ Kd6 41.Bg2 Nc7 42.Bb7 Ne8 43.Be4 Nf6 44.Bg6 Kc6

After some minor sparring we are back where we started. White tries a different tack.

45.Bf7 Kd6!?

After 45...a6 46.b5+ axb5+ 47.axb5+ Kd6 the draw could be agreed fairly soon.

46.Kb5 Ne4!?

This aggressive move suggests that Black is playing for a win, or at least keeping that possibility open. Otherwise he might have dug in with 46...Kc7 and if 47.Ka6 then 47...Kb8 48.Bg6 Nd5 49.Kb5 Kc7 50.Kc4 Kd6 51.Bf7 Nf6, when things are looking familiar.

It should be noted that at this point both players were running short of time. This largely accounts for the reversals of fortune that occur over the next few moves.

47.Ka6

White's only chance for active play is an attack on his opponent's queenside pawns. He could have prefaced this move with the interesting 47.g3!?, trying to prevent Black from creating a passed pawn, but then 47...Nc3+! 48.Ka6 Nxa4 49.Kxa7 Kc7 leads to yet another drawish position.

47...Nxf2 48.Kxa7 Kc7?

A serious error that should have cost Black the game. Correct was 48...Nd3! when there are two main variations:

A) 49.Kxb6 Nxb4 50.a5 e4 51.Bc4! (not 51.a6? Nxa6 52.Kxa6 e3 53.Bc4 Kc5 and Black wins) 51...e3 52.a6 Nxa6 53.Bxa6 Kd5 54.Ka5 h5 55.Kb4 Ke4 56.Be2 h4 57.g4 (or 57.gxh4 gxh4 58.Kc3 Kf4 59.Kd3 Kg3 60.Bg4 Kf2 and draws) 57...Kf4 58.Kc3 Kg3 59.Kd3 Kxh3 60.Kxe3 Kg3 61.Bf3 h3 62.Ke4 h2 63.Ke3 h1Q 64.Bxh1 Kxg4 with an immediate draw;

B) 49.b5 e4 50.Bg6! (other moves lose) 50...Nc5 51.Bxe4 Nxe4 52.Kxb6 Kd5 (not 52...Nxg3 53.a5 and White wins) 53.a5 Nd6 54.a6 Nc4+ 55.Kb7 Kc5 56.a7 Nb6 57.Ka6 Na8 and the draw is obvious.

(second diagram)

49.a5?

A serious error in return. White is perfectly placed to queen his b-pawn, needing only to lever Black's b-pawn out of the way. But he has to do it accurately. He should prevent Black's next move with 49.b5! when the likely continuation is 49...Nd3 50.a5 (only now!) 50...Nc5 (or 50...bxa5 51.b6+ etc.) 51.axb6+ Kd6 52.b7 Nxb7 53.Kxb7 and wins.

49...b5!

Correctly blockading the dangerous White b-pawn. Completely wrong would be 49...bxa5? 50.b5! as in the previous note.

50.Ka6 e4 51.Kxb5 e3 52.Bc4 Ne4 53.Bd3?

This is no time for waiting moves. White can still save the game with 53.Ka6! Kb8 (if 53...Nxg3 54.b5 e2 55.Bxe2 Nxe2 56.Ka7 Nd4 57.b6+ Kc6 58.b7 Nb5+ 59.Ka8 and Black must give perpetual with 59...Nc7+, etc.) 54.Kb6 Nc3 55.a6 e2 56.Bxe2 Nxe2 57.a7+ Ka8 58.g4 Nc3 59.b5 Na4+ 60.Ka6 Nc5+ 61.Kb6 Na4+ and again Black has nothing better than a draw.

53...Nxg3 54.Kc5 e2 55.Bxe2 Nxe2 56.b5 Nf4 57.b6+ Kb7 58.Kb5 (third diagram)

White's connected passed pawns give him a semblance of counterplay, but that's about it.

58...h5!?

Black wins cleanly after 58...Nxh3! 59.a6+ Kb8 60.Kc6 Nf4 61.a7+ Ka8 62.b7+ Kxa7 63.Kc7 Nd5+ 64.Kc8 Nb6+ 65.Kc7 Nd7! But the text move does not spoil anything.

59.a6+ Kb8 60.Kc6 Ne6?!

Black is afraid of the advancing White pawns and does not notice the winning knight manoeuvre. After 60...g4 61.hxg4 hxg4 62.a7+ Ka8 63.b7+ (if 63.Kc7 then 63...Nd5+ 64.Kc6 Nxb6, etc.) 63...Kxa7 64.Kc7 Nd5+ 65.Kc8 Nb6+ 66.Kc7 Nd7 Black wins easily.

61.Kd5 g4?

The final error, after which Black can no longer win. He can still transpose to the previous note with 61...Nf4+ 62.Kc6 Nxh3.

62.hxg4 hxg4 63.Ke4!

White's king correctly heads east in order to deal with Black's pawn.

63...Nc5+

Neither side can make progress and the draw was agreed here.

½–½

Defensive tactics

Here is a late middlegame position of a type that arises so often in mixed tournaments. One side clearly has the initiative because of greater piece activity, but a decisive tactic has not quite materialised. In this particular case both players have pawn weaknesses but Black is on move and should be thinking about putting White to the test. But what's the best way forward?

This was the challenge facing Efim Geller (Black) in his game with Yuri Balashov from the 1969 Soviet Championship in Moscow. If it were White's turn to play, he would oppose rooks with 1.Re1 and take a large step towards neutralising Black's initiative. Geller understood this very well and was ready with an active idea:

1...Re2

The attack on the g-pawn makes things difficult for White, as the following variations show:

a) 2.Be1 Rxc2+ 3.Kxc2 Rxa2+ followed by 4...Rxg2 and Black wins easily;

b) 2.Rg1 Bd4 3.Re1 Rxg2, etc.;

c) 2.g3 Rf2 3.f4 (3.Re1 Rxf3) 3...b5 4.Re1 h5! 5.Re7 Rf1+! 6.Re1 Rxe1+ 7.Bxe1 h4 8.Kb1 ( 8.Rc5 Rxa2 9.Rxf5 Ra1+ 10.Kd2 b4–+) 8...Re4 9.Bd2 h3 and White is in continuing difficulties.

But there is a hidden resource that saves the game for White, and Balashov finds it:

2.Re1!! Rxg2 3.Rc8+ Kf7 4.Rc7+ Kf8

Unfortunately for Geller, Black can no longer play for a win. If 4...Kg6 5.Re6+ Kh5 and now White has the surprise shot 6.Rxg7!, which wins immediately because 6...Rxg7 runs into 7.Rh6#.

After the further moves

5.Rc8+ Kf7 6.Rc7+

the players agreed to a draw.

In the starting position Black has two other tries:

a) 1....Kf7 2.Re1 Rxe1+ 3.Bxe1 Be5 4.g3 f4, but White can defend with 5.gxf4 Bxf4+ 6.Kb2 Be5+ 7.Bc3 Rb4+ 8.Kc1 Bf4+ 9.Bd2 Ke6 10.Bxf4 Rxf4 11.Rc3;

b) 1...b5 2.Re1 Rxe1+ 3.Bxe1 Rc4 4.Rxc4 bxc4 5.Kc2 Kf7 6.g3 and a draw is the correct outcome.

The conclusion is that Black has a temporary initiative but no real advantage.

Despite his reputation as the scourge of world champions, Efim Geller had his hands full with Yuri Balashov, losing four games, drawing ten, and scoring just a single win against the Moscow grandmaster.

Barcza vs Haag, Tallinn 1969

Last time we looked at one of Grandmaster Barcza's rook endings. Here is another less-complicated one (see the first diagram).

Most endings of rook and pawn vs rook and pawn are drawn with best play from both sides. The exceptions occur when one of the pawns is much closer to its queening square. If the inferior side cannot set up a blockade, the correct defensive strategy is to give up the rook for the pawn (or new queen) and then try to promote one's own pawn. The superior side will of course try to prevent this, and the game becomes a race in which the outcome can depend on a single tempo.

In the starting position White has the advantage but with correct play it will not be enough to force a win.

51.g6 Rh3+ 52.Kg4 Rh1 53.Rf7+ Kd6 54.g7 Rg1+ 55.Kh5

(second diagram)

55...Ke5?

It isn't necessary to calculate variations to see that this move must be wrong. White is queening by force, so Black has to be ready to advance his own pawn as quickly as possible. After the self-blockade, Black will have to spend another move getting his king out of the way, which in this case is enough to turn a draw into a loss.

Correct was 55...Kd5! 56.Kh6 (after 56.Rd7+ Black can safely change plans with 56... Ke5 because his king gets to f6, where it interferes with White's king: 57.Kh6 Kf6 58.Kh7 e5 59.Kh8 (definitely not 59.g8Q? Rh1#) 59...e4 60.g8Q Rxg8+ 61.Kxg8 Ke5 and Black draws easily) 56...Rh1+ 57.Kg6 Rg1+ 58.Kh7 e5 59.g8Q (or 59.Rf6 e4 60.Rg6 Rh1+ 61.Kg8 Rf1 and White cannot make progress) 59...Rxg8 60.Kxg8 e4 61.Re7 Kd4 and White will have to give up his own rook in order to stop Black's pawn.

56.Kh6 Rh1+ 57.Kg6 Rg1+ 58.Kh7 Rh1+ 59.Kg8 Rg1

(third diagram)

60.Rf8?

This is a rather serious error at the grandmaster level. Instead of taking two moves to win Black's rook, White takes three moves to do the same thing. In this way he returns the tempo that Black wasted earlier, which is enough turn the game back into a draw.

The winning line is 60.Kf8! Kd4 61.Rd7+! (an important technical device that gains a tempo for White) 61...Kc4 (Black also concedes a tempo if the king moves in front of the pawn) 62.g8Q Rxg8+ 63.Kxg8 e5 64.Re7! Kd4 65.Kf7 e4 66.Kf6 Kd3 67.Kf5 e3 68.Kf4 e2 69.Kf3 and White successfully captures the pawn.

60...Kd4 61.Rd8+

Or 61.Kf7 e5 62.g8Q Rxg8 63.Rxg8 e4 and the e-pawn will cost White his rook.

61...Ke4 62.Kf7 e5 63.g8Q Rxg8 64.Rxg8 Kf3 65.Ke6 e4 66.Rf8+ Kg2 67.Kd5 e3 68.Re8 Kf2 ½–½

Barcza vs Rossolimo, Vrsac 1969

This ending appeared in Informant 8 with analysis by the winner, GM Gedeon Barcza of Hungary. White has just exchanged minor pieces on c6 and is ready to take over the b-file. Having the move is usually a valuable asset when both sides have pawn weaknesses, and that is certainly the case here.

36.Rb3 Ke7 37.Rb5 e4

Black could play 37...Ra6 with the idea that a passed c-pawn is less dangerous than a passed a-pawn, but that would give his rook less mobility than in the game. With the text move he is trying to prevent White's king from joining the action.

38.Rxa5 Ke6 39.e3 h5

If 39...f5 40.Ra7 Kf6 41.a5 (Barcza) and White is threatening Ra7-b7-b6.

40.Ra8 Rb6 41.Re8+

Worth consideration was 41.a5!?, for example: 41...Rb4 42.Re8+ Kf5 43.Re7 Kf6 44.Rxe4 Ra4 45.f4 Rxa5 46.Re5 g6 47.Kf3 and the simple plan of attacking the c-pawn will put Black in serious trouble.

41...Kd6 42.Rxe4 Ra6 43.Kh3 g5 44.f4 f6

(second diagram)

45.fxg5?

Rossolimo must have breathed a big sigh of relief after this move, a serious error giving away most if not all of White's advantage. The game is going to turn into a pawn race, and in these situations the race is usually to the swift. Much stronger was the immediate 45.g4! If Black continues 45...h4 as in the game, then White plays 46.f5!, fixing a weak pawn on f6. A likely continuation then is 46...Rxa4 47.Re6+ Kd7 48.Rxf6 Rxc4 49.Rg6 Re4 50.Rxg5 Rxe3+ 51.Kxh4 c4 52.f6!, when it is clear that White is winning easily. Black fares no better after 45...hxg4+ 46.Kxg4 gxf4 47.Rxf4 Ke5 48.Rf5+ Ke4 (or 48...Ke6 49.Rxc5 Rxa4 50.h4 Kd6 51.Rd5+ Ke6 52.Rd4) 49.Rxc5 Rxa4 50.Rf5 Kxe3 51.Rf3+! Kd2 (51...Ke4 52.Rxf6 and Black cannot capture the c-pawn) 52.Rf4 and White's h-pawn is the decisive factor. It is strange that a player of Barcza's calibre did not see these ideas when analysing the game.

45...fxg5 46.g4 h4 47.Re8 Rxa4 48.Rg8 Rxc4 49.Rxg5 Re4 50.Rf5 c4 51.Rf1 c3 52.g5

Barcza gave the line 52.Kxh4 Rxe3 53.g5 Kg5 c2 54.h4 Re2 55.h5 Rd2 56.Rc1 Ke7 57.Kh6 Kf8 58.g5 Kg8! and Black draws. If 59.g6 Kh8 60.Ra1 then 60...Rd8! and White cannot make progress.

52...c2 53.g6 Rxe3+ 54.Kxh4 Re2 55.h3 Ke6?

Black can draw easily with 55...Rd2 56.Rc1 Ke7 57.Kg5 Kf8 as given by Barcza, or in much trickier fashion with the immediate 55...Ke7!?, for example, 56.Kh5 Rd2 57.g7!? Rd5+!! (the point of this check is to drive White's king to one of a number of awkward squares) 58.Kg6 (other moves are no better, as the reader can verify) 58...Rd1! 59.g8Q Rxf1! 60.Qg7+ Ke6! and White must repeat moves after 61.Qg8+ Ke7! since 62.Qh7+? loses after 62...Kd6!

56.Kh5 Rd2

Black could try 56...Rf2!? but after 57.g7 Rxf1 58.g8Q+, he will inevitably lose either his rook or his pawn.

57.g7 Rg2

There is no time left for 57...Rd1 because White will simply queen with check.

58.Kh6 Ke5 59.h4 Kd4 60.h5 1–0

An endgame concealing some interesting subtleties.

Alekhine on New York 1927

Alexander Alekhine's book of the New York 1927 international tournament is an acknowledged classic in European countries but is almost unknown in North America. The reason is simple: the book was published in German, and a proper translation has never been made available in English. Today we are going to remedy that situation to a small degree by presenting Alekhine's annotations to one of the more interesting games of the tournament: the first-round encounter between Frank Marshall and Aron Nimzowitsch.

Before this game, Marshall's lifetime score against Nimzowitsch stood at 2 wins, 3 losses, and 4 draws. He had won their last encounter (at Marienbad 1925) and was probably eager to level the scores, especially since he had the White pieces here. His choice of the Exchange Variation of the French Defence was a strange one for this purpose, although he gave it a typically aggressive interpretation. Nimzowitsch handled Marshall's attacking ideas in his own deeply positional style, and the result was an impressive victory, at least on the surface...

The commentary in standard font is by Alekhine from his book Das New Yorker Schachturnier 1927, while the commentary in italics is by Nimzowitsch from his book Chess Praxis. There are some interesting points of disagreement!

Marshall -- Nimzowitsch
New York 1927 (1)
French Defence C01

1.e4 e6 2.d4 d5 3.Nc3 Bb4 4.exd5 exd5 5.Nf3

With this move White voluntarily gives up any pretensions to an opening advantage; moreover, by exchanging on c3 at a convenient moment, Black obtains chances of creating an enduring weakness in White's position in the form of doubled pawns. The immediate 5.Bd3 is therefore played more often.

5...Ne7 6.Bd3 Nbc6 7.h3

If 7.0-0 immediately, then 7...Bg4 would not be very pleasant for White. The asymmetrical development of the king's knights has not turned out in White's favour.

7...Be6

Nimzowitsch is emboldened by his opponent's unsuccessful handling of the opening and he therefore avoids the natural exchanging manoeuvre 7...Bf5 in favour of a plan that is complicated and interesting, but also not completely correct.

8.O-O Qd7 9.Bf4

Simpler was 9.Ne2 with approximately equal play. However, the move in the game is also not to be criticised – it actually gives good practical results because it provokes the opponent into some risky experimentation.

9...Bxc3

Over the past few years, play against weak doubled pawns has become one of Nimzowitsch's favourite strategic motifs; and he plays such positions, which he knows how to obtain in the most varied openings, with particular virtuosity. However, in the present case he is not on the right track because the insecure position of his own king is bound to interfere with the exploitation of his opponent's weak points. Of course, with the move 9...Bd6 he could have easily equalised the game.

10.bxc3 f6

In order to safeguard the bishop against a possible Nf3-g5 – the prophylactic meaning of 7...Be6 is thus clear. White now tried an attack on the open b-file, but it is not surprising that it failed, on account of the dynamic weakness of the double complex.

A necessary consequence of his previous move. Because the king bishop has disappeared from the board, the dark squares must be defended as far as possible by pawns – but now the square e6 has been weakened.

11.Rb1 g5 12.Bg3 O-O-O (first diagram)

Looks risky, but is part of the plan initiated on the 9th move.

The king takes on the defence of the pawns on b7 and c7.

Generally speaking, the role of the king in defence has been seriously underestimated for a long time, particularly after the sad fiasco suffered by the efforts of the older Steinitz to use the king – with many other pieces still on the board – as an inducement for his opponents to attack. It is only the period since the Great War that has brought changes in this regard; see, for example, the games of the match in Buenos Aires, where even in the middlegame the kings often served to protect certain squares where a breakthrough was threatened; in other words, they played an active role well before the endgame.

Of course, Black's last few moves – by their directness and, in a certain sense, their strategic novelty – make a very favourable impression on anyone who believes in the evolution of chess strategy and its extension to new depths. Therefore it is almost vexatious to discover – after a very close examination of the position – that this plan is not only not the best, but that the correct response by the opponent – and not a difficult one to find at that – would actually put Black in a highly unenviable situation.

In view of all this it was better to forgo 11...g5 and play instead the more modest 11...Nd8 followed by 0-0, etc.

13.Qe2?

Marshall's equilibrium has been disturbed by his opponent's boldness, and he commits – both here, and later on – a series of inaccuracies that are difficult to remedy. For example, instead of the queen move – which carries only the crude and easily-parried threat of 14.Ba6 – a less-stereotyped plan of attack was appropriate: 13.Nd2! If, say, 13...Nb8; then 14.Nb3 b6 15.Qe2 – this time with really unpleasant threats. It would therefore be better for Black to reply to 13.Nd2 with 13...Na5 with the aim of exchanging off the enemy knight on its way to c5. However, notwithstanding that this exchange would restore White's pawn position on the queenside and would therefore demonstrate the insufficiency of the plan begun by the move 9...Bxc3, White would in any case be under no obligation to play the immediate 14.Nb3; very strong instead would be Qd1-c1 (followed by Qc1-a3 when appropriate) and only then the move of the knight. It is not difficult to convince oneself that by adopting such a plan of play White would obtain an enduring initiative. Now, on the other hand, he gradually runs into difficulties.

13...Rde8!

Not 13...Rdg8 because a flank attack is best undermined by a concentration in the centre and not by a counter-attack on a wing.

Defence (by clearing the square d8 for the king) and counterattack at the same time.

14.Rfe1

If 14.Ba6 then 14...bxa6 15.Qxa6+ Kd8 16.Rb7 Nf5! 17.Rxc7 Qxc7 18.Bxc7+ Kxc7 with an easy win.

14...Nf5 15.Bxf5(?)

The move 15.Ba6 proves insufficient after 15...bxa6.

It is only through the absence of this bishop that White's queenside pawns become really weak – and this promises White a difficult endgame. In spite of the obvious danger, more chances were offered by 15.Bh2.

15...Bxf5 16.Qb5 Nd8 17.Qc5?

The decisive mistake, giving Black an opportunity to set up – with tempo – a solid defensive position on the queenside. By playing 17.Qa5! Marshall could have prevented this, since if 17...Kb8 then 18.c4! and the resulting complications are not unfavourable for White.

17...b6 18.Qa3 Kb7 19.Qb3

The beginning of a tragedy: the square c2 can hardly be defended otherwise.

19...Nc6

Already a blockader makes for c4 where it will demonstrate the weakness of the doubled pawn.

20.Nd2

On the 13th move this would have been the start of a very promising attack, but now it is just a modest defence against the threat of a knight invasion on c4. The storm clouds are gathering.

20...Na5 21.Qb2 Rxe1+ 22.Rxe1 Re8

The double exchange of rooks is connected with a purely tactical idea (to say nothing of the fact that Black does not need these pieces in order to realise his advantage): Black eliminates the threat of Nb3-c5+ (after Nd2-b3) and at the same time he avoids the exchange of knights, which would otherwise be forced.

23.Rxe8 Qxe8 24.Qb1

What else could White do? If 24.Kf1, for example, then 24...Qa4 would be very strong.

24...Kc8(?)

Here 24...Qe2 was also good.

Time pressure, obviously; otherwise one cannot understand why Black did not play 24...Qe2! Then after 25.Qc1 White would be practically stalemated, while in case of 25.Nb3 Nc4 26.Nc5+ Kc8 27.Nd3 Nd2! Black would win material and also retain the attack.

25.Qd1 Qe6

Also strong was 25...Qc6.

26.Nb3 Nc4 27.Nd2 Na3 28.Nf1 Nxc2

Now Black has an ending with a pawn plus but bishops of opposite colours, and many of the onlookers prognosticated a draw.

Because of the presence of opposite-coloured bishops, White could perhaps reconcile himself to the loss of this pawn, if not for the structural weaknesses on his queenside. But now he can only wait while his opponent prepares the fatal blow.

29.Qh5 Bd3 30.Qd1 Qe4

Not 30...Qe2 at once because of 31.Qxe2 and 32.Ne3.

31.Nd2

If 31.f3 then 31...Qe2 is sound.

31...Qe2 32.Qxe2

After 32.Qc1 Ne1! White would gradually perish from suffocation.

32...Bxe2 33.f4 Na3 34.fxg5 fxg5 35.Kf2

Otherwise Black plays 35...Nb5 followed by Be2-d3, etc.

35...Bh5 36.Be5 g4 37.hxg4

Black can only welcome pawn exchanges because on the other wing he has enough material for a win. Therefore from the practical point of view it was better to play 37.h4.

37...Bxg4 38.Ke3 Bf5 39.Bg7 (second diagram) 39...Be6!

Threatening to win a pawn with 40...Nb5, which if played immediately would be premature on account of the possible reply c3-c4. White, being obviously in time pressure (the 40th move!) does not notice the threats, and after this the endgame does not present his opponent any difficulties whatsoever. However, Black would also win after 40.Kd3!, for example: 40...Kd7 41.Bf8 Bf5+ 42.Ke3 Nc2+ 43.Kf4 Bg6 44.Ke5 Ne3, etc.

40.Bf8? Nb5 41.Nb1 a5

Here 41...Bf5 was also playable: 42.a4 Bxb1 43.axb5 Ba2 44.Kf4 Bc4 45.Ke5 Kd7 46.Bb4 c6 47.bxc6+ Kxc6 and the king migrates to b3.

Also possible was 41...Bf5, since if 42.a4 Bxb1 43.axb5, after which the simplest way for Black is Kc8-d7 followed by c7-c6 (in case of Ke3-f4-e5). After the exchange of pawns his king gets to c4.

42.Kd2

A winning line, not unlike that shown in the preceding note would be: 42.Kf4 Bf7 43.a4 Bg6 44.axb5 Bxb1 45.Ke5 Ba7 46.Ke6 Bc4 with 47...Kb7 and 48...c6, etc.

42...Bf5 43.Na3 Nxa3 44.Bxa3 Bb1 45.Bf8 Bxa2

Marshall could have quietly spared himself the following 15 moves.

46.Bg7 Bc4 47.Ke3 Kb7 48.Bf6 Ka6 49.Kd2

If 49.Bd8, Black has a win, the king penetrating to b3, e.g., 49.Bd8 Kb5 50.Bxc7 Kc4 51.Bxb6 a4, with 52...Kb3, and wins as the a-pawn cannot be stopped. This variation shows the enduring weakness of the dead and gone double complex. For in the passed a-pawn is mirrored the weakness of the defunct White a-pawn, and in the blocked long diagonal f6-a1 is manifested, in memoriam, the obstructive effect of the pawn formation c3 and d4. White might have resigned here.

49...Bf1 50.g3 Kb5 51.Kc1 Kc4 52.Kb2 c5 53.Be5 cxd4 54.Bxd4 b5 55.Bb6 a4 56.Ba5 d4 57.cxd4 b4 58.Bb6 a3+ 59.Ka2 Kb5 60.Bc5 Ka4 White resigned. 0-1

Black to play and... lose

This rook ending arose in the game Wolff – Fishbein, USA Jr. Ch. 1988 and was analysed by Fishbein in Informant 47. Black will soon have to surrender his rook for White's advanced pawn, and his drawing chances will turn on whether he can promote his own pawn safely. Because it is still on its home square and White's pieces are not too far away, indications are that the race will be a close one. In the game Fishbein managed to achieve the draw as follows:

1...Kg5
Unavailing is 1...Rc1+ 2.Kd8 Kg5 3.Ra5+ Kg4 4.Rd5! (making it more difficult for Black to control the queening square) 4...Ra1 5.Ke7 Ra8 6.d8Q+ Rxd8 7.Rxd8 h5 8.Kf6 h4 9.Rd4+ Kg3 10.Kg5 h3 11.Rd3+ Kg2 12.Kg4 h2 13.Rd2+ Kg1 14.Kg3 h1N+ (or 14...h1Q 15.Rd1 mate) 15.Kf3 and wins.
2.d8Q+ Rxd8 3.Kxd8 h5 4.Ke7 h4!
Of course not 4...Kg4? 5.Kf6! h4 6.Ra4+ and White wins as in the previous note. In these positions the technique of “shouldering off” the opposing king always plays an important role.
5.Ke6 h3 6.Ke5 Kg4 7.Ke4 h2 8.Rh7 Kg3 9.Ke3 Kg2 10.Rxh2+ 1/2-1/2

Based on this rather straightforward result, the diagrammed position is labeled by Informant 47 as “Black to play and draw.” But it turns out there is a large and surprising improvement for White:

1...Kg5 2.Ra1!!
Black must of course avoid the capture, but the result is that White's rook gains crucial checking distance against Black's king.
2...Rd2 3.Rg1+! Kf5 4.Rh1! Kg6 5.d8Q
After some accurate play to restrain Black's king, White now cashes in his pawn for Black's rook. Meanwhile Black's pawn has still not moved.
5...Rxd8 6.Kxd8 h5 7.Ke7 Kg5 8.Ke6
White is obviously somewhat ahead of the game Wolff-Fishbein, and the difference is enough for a decisive result. Here is one possible conclusion:
8...h4 9.Ke5 Kg4 10.Ke4 Kg3 11.Ke3 h3 12.Rg1+ Kh2 13.Rg8! Kh1 14.Kf3 h2 15.Rb8 Kg1 16.Rb1 mate.

As they say, it just ain't over till it's over!

Endgame study

The importance of studying the endgame was underlined once again by the outcome of a recent blitz game of mine.

I am White in the diagrammed position and it is Black to play. My opponent saw that he could not go forward without moving his knight and surrendering his last pawn in the process. So he offered me a draw, and of course I accepted it immediately.

This outcome seems reasonable because in the overwhelming majority of cases the ending of knight vs pawn is a trivial draw. But a little analysis shows this particular case to be an exception. Let's have a look:

1...Kd7! 2.Ka8 Nd6!
The idea Black had missed. He can sacrifice his last pawn because despite the greatly reduced material White's king is caught in a mating attack.
3.Kxa7 Kc7 4.Ka8 Nc8!
Forcing White's reply.
5.a7 Nb6 mate!

On his third move White can avoid capturing the pawn but it makes no difference:
3.Kb8 Kd8 4.Ka8 Kc7 5.Kxa7
Now this is forced.
5...Nb5+ 6.Ka8 Kc8 7.a7
Unfortunately this too is forced.
7...Nc7 mate!

You all know the moral of the story so I won't bother repeating it.

Endgame tactics 2

Last time I showed a rook ending in which a tactical device simplified the task of converting an extra pawn. Today there is a similar example on the menu.

The position in the diagram arose in one of my games from the 1976 B.C. Championship. After an unusual opening a weak pawn had appeared on e4, and in the middlegame that followed I managed to exchange off every piece that could reasonably defend it. We have now arrived in a rook ending with the pawn about to be captured.

34.Rxe4 Kd6 35.Ke3 Rf7
No better is 35...Rb5 36.f4 h5 37.Kd3 g6 38.Ke3 and Black is in a mini-zugzwang; for example, 38...Rd5 39.Rd4 or 38...Kd7 39.Re5.
36.Rd4+!?
Of course the natural move here is 36.f4, so my actual move 36.Rd4+ requires a bit of explanation. After 35...Rf7 it didn't take me too long to work out my opponent's defensive strategy, which in simple terms consists in defending the b-pawn laterally with his rook and keeping his king centralised in order to prevent the advance of White's king. With these thoughts in mind I worked out a few variations and in so doing spotted an unusual tactical trick.
36...Ke5
As predicted!
37.f4+ Kf5 38.g4+!
This appeared to take Black by surprise.
38...Kxg4 39.f5+! Kxf5
This loses immediately, but after 39...Kh5 40.fxe6 Re7 41.Re4 Black cannot hold the position; for example, 41...g5 42.hxg5 Kxg5 43.Kd4 Kf6 44.Kd5, etc.
40.Rf4+ Kg6 41.h5+!
The main point of White's play. Black's king is deflected away from his rook, which is then captured by White for an easy win.
1-0

Being able to foresee and make use of tactical devices in order to simplify the task of converting an advantage is part of what is known collectively as endgame technique. It is very important that in playing for such tricks we analyse carefully and do not allow the opponent any surprise defences that may bring him back into the game. For example, in this game it was very important that the alternative 39...Kh5 did not bring any relief to Black.

Endgame tactics

Picture this: after a difficult middlegame you have managed to transpose into an endgame with an extra outside passed pawn. Smooth sailing, right? No, not always. If your opponent has blockaded your extra pawn you could still have a lot of work left to do. Before casting around for a second target, it is often worth another look to see if the blockade is actually all that sturdy.

Here is an example from one of my games in the 1983 B.C. Open. I am Black and I would like to promote my advanced b-pawn, but there doesn't seem to be an easy way forward at the moment. Enter the tactician... it turns out that the geometrical situation is perfect for an unusual trick:
1...Qc4
Attacking the f-pawn. If White lets it go his position will deteriorate rapidly.
2.Rf3 Qc2+!
White's reply is more-or-less forced because the alternative 2.Rf2 runs into 2...Qxf2+! 3.Qxf2 b2 and Black will either win White's queen or get a new one of his own.
3.Qxc2 bxc2 4.Rc3
My opponent had banged out his last two moves quickly and was now looking quite satisfied with himself. Unfortunately for him, I have seen a bit further:
4...Rb2!
Although this is not a check, it has similar force. I am threatening the discovery 5...c1Q+, to which 5.Kg1 is no defence because of 5...Rb1+ followed by 6...c1Q and wins. So my opponent grabbed his king and went to play it up a rank... and then froze in mid-move... and then turned somewhat red... With his time running down he eventually realised he had to move the king that was in his hand, so he quietly put it on g3, pressed the clock, and waited...
5.Kg3 Rb3!
The point of Black's manoeuvre. White's rook is now pinned against his king and he has lost control of the important c-file. My opponent did not want to see any more and resigned the game here.

Quo vadis

During a break at the recently-concluded Grand Pacific Open in Victoria I happened to be in the hotel lobby watching a blitz game between two juniors. When the diagrammed position was reached both players commented that they had no idea how the play should continue or who, if anyone, was winning.

Does Black have the advantage because his rook is supporting an outside passed pawn from behind and tying up White's rook in the process? Or does White have the advantage because of his pawn duo and the opportunity to advance it and perhaps attack Black's king?

It is one of the paradoxes of chess that the fewer pieces remain on the board, the more difficult the game becomes for human players. A major reason is that many endgame positions cannot be assessed properly without a deep and accurate calculation of variations. When faced with a board consisting largely of empty space it can actually be difficult to keep track of where the pieces are, to say nothing of where they should be going or what they can accomplish.

From the practical side there is just no substitute for extended analytical work on as wide a variety of endgame positions as possible. Without absorbing the general principles and technical devices that apply to the endgame a player will never achieve even average proficiency in this phase of the game. Based on my own difficult experience I always advise players of all ages and playing strengths to dedicate a part of their chess experience to a study of the endgame.

Let's clear up the mystery of the diagrammed position: White is winning very easily. Yes, his rook is tied up. But if the blockade is lifted, Black's pawn will not queen for at least four more moves. In the meantime by combining the action of his forces White can carry out an enveloping manoeuvre against Black's king.

Here is how the game might continue:
1.g5 Kf7 2.h5 Kg7 3.Kg4 Kf7 4.Kf5 Kg7 5.h6+ Kh7 6.Kg4! Kg6
It looks as if Black has managed to blockade White's pawns, but here is where White's rook enters the picture.
7.Rd4! Kf7
If 7...b4 then 8.Rd7! and Black's king is caught in a mating net.
8.Kf5 Kg8 9.Rd7 Rf8+ 10.Kg6 Rb8 11.h7+ Kh8 12.Kh6 b4 13.g6
Black is faced with the threat of mate by 14.g7, and 13...Rb6 is no defence because of 14.Rd8 mate.

Fighting for the draw

There were some interesting moments in the game Wu-Pechisker, British Columbia Championship 2006. In the first diagrammed position Black is to play his 52nd move. He is a pawn up and the natural question is whether this gives him any real winning chances. In effect there is only one meaningful plan: create a passed pawn and try to queen it. Before executing this plan Black could have tried improving the position of his other pieces but in the end he decided not to wait:

52...d5 53.exd5+ Kxd5
White must now decide on a plan of defence. The most natural move is 54.Rg4 aiming for a blockade on the square e4. After 54...Ra8 55.Nd2 Ra3+ 56.Ke2 Be7 57.Nc4 Ra2+ 58.Kd3 Bc5 59.Re4 Black appears to have no way of breaking White's blockade; for example 59...Bd4 60.Rg4 Rh2 61.Ne3+ Kc5 (if 61...Bxe3 62.Kxe3 Rh3+ 63.Ke2 and Black cannot win) 62.Nf5 Rh3+ 63.Ke4 Rh1 64.Nxd4 exd4 65.Rg8 Re1+ 66.Kd3 and again Black can make no progress. White must have seen some problems with this plan because instead he chose the radical
54.Nxe5!??
sacrificing his knight to reach the theoretically drawn ending of rook vs. rook and bishop. I have given this move one exclamation mark for hubris and two question marks for impracticality. Unless White is completely certain of his defensive technique, taking on this endgame is a very risky enterprise.
54...Bxe5 55.Rh3 Ra8 56.Rh5 Ra3+ 57.Kc2 Rg3 58.Rf5 Ke4 59.Rh5 Rc3+ 60.Kd2 Rc6 61.Rh7 Bf4+ 62.Ke2 Rc2+ 63.Kd1 Rd2+ 64.Ke1 Rb2 65.Kd1 Be3
White's king has been driven to the edge of the board, and he must now play absolutely perfectly in order to survive.
66.Rd7!
The only move to draw; for example, 66.Re7+ Kd3 67.Rd7+ Bd4 68.Rc7 Ra2 69.Rc8 Rf2 70.Re8 Bf6 71.Re6 Bc3 72.Rd6+ Bd4 73.Re6 Rd2+ 74.Ke1 Ra2 75.Kf1 Rf2+ 76.Ke1 Rf5 77.Rd6 Rg5 and White must give up his rook to avoid immediate mate.
66...Rh2 67.Rd8 Bd4
(second diagram)
68.Kc1?
White has defended accurately until now, but this move is a mistake because the king will run into trouble in the restricted space on the queenside. White could maintain a defensible position with 68.Rf8!; for example, 68...Kd3 69.Ke1 Rg2 70.Rf3+! Be3 71.Rf8 and Black has not made any progress.
68...Kd3 69.Kb1 Rb2+ 70.Kc1 Ra2 71.Rb8 Rc2+?!
Black does not notice the winning manoeuvre 71...Rf2! 72.Kb1 Rf1+ 73.Ka2 Ra1+ 74.Kb3 Rb1+, picking up White's rook.
72.Kb1 Kd2?
A second error that allows White a clever drawing resource. It was not too late to get back on track with 72...Rf2 73.Rc8 ( 73.Rd8 Rb2+ 74.Kc1 Ra2 75.Rb8 Rf2 transposes to the previous note) 73...Rb2+ 74.Kc1 Ra2 75.Kd1 Rf2 76.Re8 Bf6 and wins as in the note to White's 66th move.

73.Rb4! Bc3 74.Rd4+!
With this rook sacrifice White forces stalemate, and so the players agreed to a draw here.
½-½

Artillery

Positions in which all the minor pieces have been exchanged and the players are left with a queen and two rooks each were examined by Alex Angos in a fascinating 1982 monograph entitled Endgame Artillery. Such positions present an interesting blend of middlegame and endgame features. If there are no passed pawns or serious pawn weaknesses and the two kings are relatively safe the outcome is often an uneventful draw. But when there are significant imbalances the initiative becomes a key factor and the play can be extremely complicated. In this regard the classic encounters Rubinstein-Alekhine, Dresden 1926; Alekhine-Euwe, Nottingham 1936; and Korchnoi-Spassky, Belgrade (m/7) 1977/78 are recommended for further study.

In a recent team match I managed to win a pawn but at the same time had to give my lower-rated opponent dangerous kingside pressure. After some exertions I managed to exchange off the minor pieces and blunt the attack, arriving at the “endgame artillery” position shown in the first diagram. My opponent's last move had been 42...Reg8 and I could sense that he was now looking for a draw by repetition in the variation 43.Rg1 Re8 44.Re1 Reg8, etc. In unhappily typical fashion I had left myself five minutes for all the rest of my moves; meanwhile my opponent had nine minutes on his clock. I briefly considered acquiescing to the draw because the match score at that moment was 4-0 for our side with three games to go including this one. But I rejected this notion almost immediately – I was a pawn up and I wanted payback for the difficult defence I had conducted! On top of it all there was an interesting counterattacking line available, so without further ado I plunged ahead...

43.e4! fxe4
After 43...Rxg3 44.Rxf5+ Black is forced to play 44...R8g5 because 44...Kh7 loses immediately to 45.Rxh5+ while 44...R3g5 allows the simplifying operation 45.Rxg5 Rxg5 46.Rg1 Qe7 47.Qxg5+ Qxg5 48.Rxg5 Kxg5 49.Kg3 with an easily won pawn ending for White. After 44...R8g5 White must play 45.Re2 (definitely not 45.Rxg5? Qh3#!) and now there are two lines:
a) 45...Kg6 46.Qf4! R5g4 (the alternative 46...R3g4 47.Rxg5+ Rxg5 48.Rg2 leads to a queen ending that is very difficult for Black to defend) 47.Qb8! Rd3 48.Qd8! (taking control of key squares) 48...Rgg3 49.d5 Rh3+ 50.Kg2 Rhg3+ 51.Kf1 Rd1+ 52.Kf2 and Black is losing at least a rook;
b) 45...h4 46.Qf4 Kh5 47.Rxg5+ Rxg5 48.d5 Qd7 ( or 48...Qe5 49.Qxe5 Rxe5 50.d6 Re8 51.e5 and wins) 49.Rf2! and White will either mate, win material or promote a pawn (and then mate!)
44.Rfxe4
Stronger was 44.Rexe4; for example, 44...Qd7 45.Qe3! R8g6 46.Re8 Qg7 47.Rff8 Kh7 48.Re7 and wins.
44...Qf5 45.Rf4 Qd3?! (second diagram)
Perhaps Black was afraid for his king but in the circumstances he should not be the one forcing off the queens. More tenacious was 45...Qg6 although after 47.Re7! Rg7 48.Rxg7 Kxg7 49.Qe3! White should still win with proper technique.
46.Qxd3?!
The time factor... Black's last move had surprised me and I was forced to spend one of my remaining minutes seeing if I could win after exchanging queens. This left me with two minutes to play the rest of the game and no time to look at the alternative 46.Rf6+! which is in fact much stronger. The main line then is 46...R8g6 47.Rxg6+ Kxg6 48.Re6+ Kf7 49.Qxg5 Kxe6 50.Qe5+ Kd7 51.Qxb5+ Kc8 52.Qe8+ Kc7 53.Qe5+ Kd7 54.d5 Qf1 55.a4 with two passed pawns for White and no visible counterplay for Black.
46...cxd3 47.Rf6+ Kh7?!
This loses quickly. Black should have abandoned the attack on g3 and played 47...Kg7! After 48.Rf3 Rc8! 49.Rxd3 Rc2+ 50.Kh3 Rxb2 51.d5 Rf5 52.d6! (not 52.g4!? immediately because of 52...hxg4+ 53.Kxg4 Rf8 54.d6 Rg2+ 55.Kh3 Rg5 56.Re4 Rh8+ 57.Rh4 Rd8 58.Rg4 Rxg4 59.Kxg4 Kf6 60.Rd5 Ke6 61.Rxb5 Rxd6 and White cannot win) 52...Rff2 53.g4! Rh2+ 54.Kg3 Rbd2 and now White must find 55.Ree3! (stronger than 55.Red1 because it creates a safe passage for White's king) 55...Rdf2+ (or 55...hxg4+ 56.Ke3 Rxd3 57.Kxd3! Rg1 58.Kd2! Rg2+ 59.Kc3! and wins) 56.Ke4 Rxg4+ and now after 57.Kd5! it is clear that Black will be resigning soon. The problem with the text move 47...Kh7 is that White can immediately force off a pair of rooks. The resulting single rook ending can then be won with almost no thought, a pleasantly untroubling task when one has only two minutes left on the clock.
48.Re7+! R8g7
Much weaker is 48...Kh8? 49.Rh6#!
49.Rxg7+ Rxg7 50.Rf3 Rc7 51.Rxd3 Rc2+ 52.Kh3 Rxb2 53.d5 (third diagram) 53...Rc2 54.d6 Rc8 55.d7 Rd8 56.Kh4 Kg6 57.Rd5!
Winning the h-pawn and bringing White's g-pawn into play. The rest requires no comment.
57...Kf6 58.Kxh5 Ke6 59.Rd3 Rh8+ 60.Kg4 Rg8+ 61.Kf4 Rf8+ 62.Ke4 Rd8 63.g4 Rg8 64.d8Q Rxg4+ 65.Kf3

Black resigns 1-0

Study time

In a recent issue of Chess Today GM Alex Baburin looked at some endings in which a lone bishop must contend with a knight plus two pawns that are separated by one file. If the defender's king is on scene a surprisingly large proportion of these endings cannot be won for the stronger side.

Baburin's article caused me to recall a 1929 study by the Soviet composer A. Gurvich, which is shown in the diagram. This study has been republished in a number of places, most notably Volume 5 of the Encyclopaedia of Chess Endings from Chess Informant as well as the Chess Endgame Training CD from Convekta.

Endgame studies are a fascinating branch of chess creativity. The stipulation – White to play and win or White to play and draw – usually if not always looks impossible of achievement but closer examination will reveal a tactical solution that eventually overcomes all resistance. In the best studies a forcing line of play will appear to hit a dead end several moves along but then a new tactic that has arisen as if by chance will allow further progress to be made. There is no better way of increasing one's powers of combination and calculation than solving endgame studies. For best results I recommend setting up the position on a board but not allowing yourself to move the men. In fact I suggest you do that here before reading further. To help with this I have not set the moves of the solution in bold type.

In Gurvich's study White is called upon to win against any defence. This looks typically impossible because Black is threatening to capture the h-pawn and then give up his bishop for the f-pawn. A lone knight cannot deliver mate so White must prevent this plan. The only way forward is 1.Nh5+ since 1...Kxh7 loses the bishop after 2.Nf6+. Black must therefore play 1...Kh8.

White must now be careful. Both 2.f6 and 2.Kf6 fail to 2...Kxh7 since the knight check is now blocked. 2.Kg6 is met by 2...Be4! 3.Nf6 Bd3 (or 3.Bc2) and White cannot make further progress since any move of the knight runs into ...Bxf5 followed by capturing the h-pawn with either king or bishop according to circumstances. So White must play 2.Kh6, protecting his pawn. This has the added virtue of threatening 3.Nf4 followed by 4.Ng6 mate.

Black has two moves to counter this plan: 2...Bf7 and 2...Be4. The first one is overcome in prosaic fashion after 3.Nf4 Be8 4.f6 Bf7 5.Nd3! Bg6 6.Ne5 Be8 (or 6...Bxh7 7.f7 Bg8 8.Ng6 mate) 7.f7 Bxf7 8.Nxf7 mate. The second move is more difficult to meet but a little reflection reveals that White must push his f-pawn with 3.f5. Everything else allows Black to force a draw with 3...Bxf5. This includes 3.Ng3 Bxf5! since 4.Nxf5 is stalemate.

After 3.f6 Black of course must take on h7 – otherwise White wins as in the line with 2...Bf7 that we looked at earlier. So 3...Bxh7 and now 4.f7 looks quite deadly. But Black is not done yet: 4...Bg8! threatens to take the pawn and force a draw, and if White promotes to a queen (or rook) Black is again stalemated.

We have reached the culminating point of the study. White must promote his pawn to either a bishop or a knight. It is well known that two knights can deliver mate only in exceptional circumstances, which do not apply here since Black has a bishop standing ready to prevent the mating move. So the only move left is 5.f8=B. Black is now free to move his bishop anywhere he likes, but in every case White will reply with 6.Nf6! This takes away the king's flight square on g8 and threatens an unstoppable 7.Bg7 mate, thus satisfying the stipulation. All of White's pieces are on dark squares so without the stalemate factor Black's bishop may as well be on another board!

To those interested in delving further into the wonderful world of endgame studies I can recommend two books: 1234 Modern End-Game Studies by Harold Lommer and Test Tube Chess by John Roycroft. Both are available in Dover reprints and should be relatively easy to obtain.

No draw

The semi-endgame position shown in the first diagram arose an open tournament held in Italy a number of years ago. White is on move. As always the key question is: who stands better and why? I think the most accurate assessment is that White has an edge but Black will manage to neutralise it. With careful play on both sides one could easily expect a handshake and a draw within the next ten moves or so. All very reasonable except that this is the game D.Gurevich-Korchnoi, Bratto 1998. Both players are in the running for first place and both players are grandmasters so today there will be no agreement to a draw until the position is a draw. And with time running short on both sides, things are going to get bumpy...

17.Rc2 Nb8
The immediate 17...a5 was also possible: 18.Na4 (not 18.b5 Nb4! 19.axb4 axb4 20.Rac1 Rxc3 21.Rxc3 bxc3 22.Rxc3 Rd5 23.Rb3 and White is now on the shady side of the draw) 18...axb4 (or 18...Na7!? 19.Rxc8 Nxc8 20.bxa5 bxa5 21.Nd4 Bf8 with equality) 19.Nxb6 Rc7 20.axb4 (definitely not 20.Rac1 b3! 21.Rxc6? Rxc6 22.Rxc6 b2 and Black wins) 20...Rb8 21.Na8 Rcc8 22.Rac1 Rxc8 23.Rxc6 Rxc6 24.Rxc6 Bxb4 25.Rc7+ Kg6 and neither player will make much progress from here.
18.Rac1 a5 19.Na2
White is playing for a win. He could practically force a draw with 19.Na4 Rxc2 20.Rxc2 axb4 21.axb4 Bxb4 22.Nxb6.
19...Rxc2+ 20.Rxc2 axb4 21.Nxb4 Bc5!?
Black also seems to be playing for a win. He can achieve a compact defensive position after 21...Bxb4 22.axb4 Na6! 23.Rc6 (23.b5 Nc5=) 23...Nxb4 24.Rxb6 Nd5 25.Rb7+ Ne7 26.Nd4 Rd6. The next few moves see White play on the slight insecurity of the bishop's post on c5.
22.Rc4 Nd7 23.a4 Ra8 24.Nd3 e5 25.Nd2 Ke6 26.e4 Kd6 27.Nb3 Ba3 28.a5!?
The point of White's preceding manoeuvres, but it is all bluff.
28...Bc5?!
There is nothing wrong with 28...bxa5; for example, 29.Ra4 Bc5 30.Nxa5 Kc7! and Black has no problems.
29.Rc2
White cannot achieve anything by exchanging on c5 because Black's king is well-placed to deal with White's a-pawn. So he marks time and waits for a better opportunity.
29...g6 30.Rd2 Ke7? (second diagram)
Black can defend his position with 30...Kc6!; for example 31.axb6 Bxb6 32.Nb4+ Kc7 33.Nd5+ Kc6 34.Rc2+ Kd6. Perhaps Korchnoi was keeping the tension as a way of playing for a win. But after the text move the game takes a big turn in White's favour.
31.Ndxc5! bxc5
No better is 31...Nxc5 32.Nxc5 bxc5 33.Ra2! c4 (or 33...Kd6 34.Kd3 Kc6 35.Kc4 Ra6 36.g4! and Black will eventually succumb to zugzwang; for example, 36...Ra7 37.a6 Kb6 38.Ra3 h6 39.Rb3+! Kc6 40.Rf3, etc.) 34.Kd2 Kd6 35.Kc3 Kc5 36.Rd2! Rb8 37.Rd5+ Kc6 38.f3! Rb3+ 39.Kxc4 Rb2 40.Rc5+ Kd6 41.Rb5 Rxg2 42.Rb6+ Ke7 43.a6 Ra2 (43...Rxh2? 44.Kb3!) 44.Kb5 Kd7 45.Rxf6 and the rest will just be a matter of technique.
32.Kd3 Rb8 33.Kc4 Rb4+ 34.Kc3 Rb7 35.a6
Also possible was 35.Rd5 Rc7 36.Kc4 winning the c-pawn.
35...Ra7 36.Kc4 Ra8 37.Kb5!
Definitely not 37...Ra2 Kd6 38.a7 Kc6 39.Ra6 Nb6+ and Black is suddenly better.
37...Ke6
Or 37...Rb8+ 38.Kc6 Rc8+ 39.Kb7 Rb8+ 40.Kc7 Rxb3 41.Rxd7+ Ke6 42.Rd6+ Ke7 43.Rc6! Ra3 44.Kb6 Rb3+ 45.Kxc5 Ra3 46.Kb6 Rb3+ 47.Kc7 and White wins.
38.Na5 Rc8
If 38...Nb8 White wins neatly with 39.Rd8! Nxa6 40.Rxa8 Nc7+ 41.Kc6 Nxa8 42.Kb7! trapping the knight.
39.Nb7?!
In some time pressure White does not notice that he wins quickly after 39.a7 Ra8 40.Ka6. But the text does not spoil anything.
39...c4 40.Rc2?!
On the last move of the time control White commits another inaccuracy, and a more serious one this time. He could still decide matters quickly with 40.a7! c3 41.Rd6+ Ke7 42.Rc6 Ra8 43.Nd6! Kd8 44.Nc8! However this was not easy to calculate, especially in time pressure.
40...c3 (third diagram)
Both sides have reached time control and Korchnoi has typically managed to scrape up counterplay in the form of his passed c-pawn. But with more time available for reflection Gurevich finds a beautiful winning line that involves the immediate surrender of his biggest asset: his passed a-pawn!
41.a7! Ra8 42.Rxc3 Rxa7 43.Nd8+!
The beginning of a forced manoeuvre to drive Black's king offside. This enables White's king to take over.
43...Kd6 44.Rd3+ Kc7 45.Ne6+ Kb8
No better is 45...Kc8 46.Rc3+ Kb8 47.Kc6 Rb7 48.Kd6 Nb6 49.Nc5 Nc8+ 50.Ke6 Rb6+ 51.Kf7 with a decisive attack on Black's kingside pawns.
46.Kc6 Kc8 47.Rc3 Ra6+?
Losing immediately, but everything else lost eventually.
48.Kb5+ 1-0

Dmitry Gurevich is not in the first rank of grandmasters but he has a respectable record against Viktor Korchnoi: +2 -2 =7.