FIDE Surveys – Georg Mohr

1

Georg Mohr:

Capablanca and the endgames

There is a general belief that the third world
champion,

grandmaster

Jose

Raoul

Capablanca was a genius in the endgames.
His play in the final part of the game was
thought to be perfect and practically
infallible – if you wanted to defeat the
Cuban, you needed to outplay him earlier.
Aleksander Alehin was aware of that and he
was closely studying the endgames before
their match (Buenos Aires 1927). It is
interesting that Aljehin, after a couple of
months preparing, did not mention the
endgames anymore. He obviously came to a
conclusion that his knowledge of endgames
was no weaker than Capablanca’s.
Analysis after decades, when a legend about
Capablanca was fading, put things in place.
Capablanca was, without a doubt, a master
of endgames, but in spite of all that, there
were mistakes present in his games.
Something is true for certain: a general
playing level in the endgames was very low
at that time.

Capablanca J.R. : Marshall F.
Havana 1913

XIIIIIIIIY
9-+-+-vL-+0
9+-+-+-+-0
9p+-+kvl-zp0
9+-+-+-zp-0
9-+KzP-+P+0
9+-+-+p+P0
9P+-+-+-+0
9+-+-+-+-0
xiiiiiiiiy

1.d5?
Where are you going? After 1.Kd3 Ld4
2.Lh6 Kf6 3.Lf8 the bishop returns to g3 and

the game would finish with a draw.
1...Ke5 2.Kd3
There is no solution left. After 2.Lc5 Le7
3.Lg1 Ke4 4.Lf2 Ld6 5.a4 Lb8 the game
would finish with, for this kind of endgames,
typical zug zwang.
2...Kf4 3.Ld6 Le5 4.Lc5 Kg3 5.Ke4 Lf4
6.d6 f2 0:1.

It is interesting that Capablanca’s opponents
in most cases played badly in the endgames;
they played worse than they knew. They
played like that probably out of fear. The
Cuban managed to save more lost and equal
positions that anyone before.

Blanco J. : Capablanca J.R.
Havana 1901

XIIIIIIIIY
9-+-+-+-sn0
9+-+-+-+P0
9p+-+-mK-+0
9+p+k+-+-0
9-zP-+-+-+0
9zP-+-+-+-0
9-+-+-+-+0
9+-+-+-+-0
xiiiiiiiiy

1.Kg7??
White would draw after simple 1.Ke7! Ke5
2.Kd7.
1...Ke6! 2.Kf8 Kf6 3.Ke8
3.Kg8 Kg6! 4.Kh8 Kf7 and after 5.a4 ba4
Black will mate from a1.
3...Kg7 4.Kd7 Sg6 5.Kc6 Se5 6.Kb6 Sc4
7.Ka6 Sa3
The knight cannot be attacked, Black’s king
will first deal with the h pawn and later
move to the queenside.

FIDE Surveys – Georg Mohr

2

Marquez Sterling M. : Capablanca J.R.
Havana 1901

XIIIIIIIIY
9-+-+-+-+0
9+P+-+-+-0
9-+k+-sNp+0
9+-+-vl-+p0
9-+-+-+-+0
9+-+K+-+P0
9-+-+-+P+0
9+-+-+-+-0
xiiiiiiiiy

50.g4
After 50.Ke4 Lg3 51.Kf3 (Khalifman
analyzed 51.Sd5? g5 52.Sf6 g4!, Draw)
51...Ld6 52.h4!! Kb7 53.Sd5 Kc6 54.Sf4
Le7 55.Sg6 Kd6 56.g3, then Sf4 and the
final result would be clear.
50...Kb7 ½.

Capablanca J.R. : Echevarria C.
Havana 1901

XIIIIIIIIY
9-+-+-+-+0
9zp-+-+-+-0
9-+-tr-+-zp0
9+-zP-mk-zp-0
9P+-tRp+P+0
9+-+-zP-+-0
9-+-+-mK-+0
9+-+-+-+-0
xiiiiiiiiy

Now let us see what is possible in chess!
Capablanca’s opponent played brilliantly for
40 moves and instead of drawing, after
44…Tc6 45.Tc4 Kd5, he began to play
completely without a sense for direction.
44...Td4?? 45.ed4 Kd5 46.Ke3 a5
And now a small triangle.
47.Kf2 Kc6 48.Ke2! Kd5 49.Ke3 1:0.

In 1901 Capablanca was crowned as Cuban

champion, after he won against Corzo.
Completely undeservedly: at that time Corzo
was still a much better player than
Capablanca in every aspect of the game
except the endgames …

Capablanca J.R. : Corzo y Prinzipe J.
Havana 1901

XIIIIIIIIY
9-+-+r+k+0
9+R+-+-zpp0
9-+-+-+-+0
9vL-zPp+-+-0
9-+-zP-sn-+0
9+-+-+K+-0
9-+-+-+-+0
9+-+-+-+-0
xiiiiiiiiy

42...Se6 ?
After 42...Se2 a pawn would be lost d4 and
White could just resign.
43.Lc3 Tc8
Black waited without a reason - after 43…h5
his position would probably still be winning.
44.Td7 Td8 45.Te7 Sg5 46.Kg4 Kf8
47.Tb7 Se4 48.Lb4 Tc8 49.c6 ½.

Corzo y Prinzipe J. : Capablanca J.R.
Havana 1901

XIIIIIIIIY
9-+-+-+-+0
9zPR+-+-+-0
9-+-+-zpk+0
9+-+p+-+-0
9-+-zP-mK-+0
9+P+-+P+-0
9r+-+-+lzP0
9+-+-+-+-0
xiiiiiiiiy

After
40.Tb8?
The games quickly ended with a draw. After

FIDE Surveys – Georg Mohr

3

simple 40.h4! Ta3 41.h5! Kh5 42.Kf5 Kh6
43.Kf6 Lf3 44.b4 Le2 45.b5 White would
have a chance for winning.
40...Ta7,
And the opponents agreed on a draw. ½.

Capablanca J.R. : Corzo y Prinzipe J.
Havana 1901

XIIIIIIIIY
9-+-+-+-mk0
9+-mK-+-+-0
9-+-zP-+-zp0
9+-sN-+-zp-0
9-+P+-+-+0
9zp-+-+-+-0
9-+-+-+-zP0
9+-+r+-+-0
xiiiiiiiiy

This was a third games of the match. Corzo
was leading with 2:0. If he would win this
game too, Capablanca certainly would not be
able to put himself together anymore. Who
knows, what the history books would say in
that case? The question is simple: with
which pawn we should proceed?
53...a2?
After 53...g4! the knight on c5 would block
the way of his own passed pawn.
54.Sb3 g4 55.d7 h5 56.d8D Td8 57.Kd8 h4
58.c5 g3 59.hg3 hg3 60.c6 g2 61.c7 ½.

Capablanca J. : Ward W.
New York 1906

XIIIIIIIIY
9-+-vL-+-+0
9+P+K+k+p0
9-+-+p+p+0
9+-+-zPp+-0
9-+-+-+-zP0
9+-+-+-+-0
9-tr-+-+-+0
9+-+-+-+-0
xiiiiiiiiy

Capablanca did not find a winning plan:
51.Kc6! (with the idea Lb6)
51...Tc2 52.Kb5 Tb2 53.Ka6, with a win for
White.
In the game was played:
51.Kc8? Ke8
At this point Black drew. After 51...f4!
52.b8D Tb8 53.Kb8 f3 54.Lb6 h5 55.Lf2 g5!
56.hg5 Kg6 the final result would be the
same.
52.Lc7 h6 53.b8D Tb8 54.Lb8 g5 55.hg5
hg5 56.Lc7 f4 57.Ld8 g4 58.Lg5 f3 59.Lh4
Kf7 60.Kd7 1:0. .

Marshall F. : Capablanca J.R.
New York 1909

XIIIIIIIIY
9-+-+-+-+0
9+-+-+p+-0
9-+-+-mK-+0
9zp-+kzP-+-0
9P+-+-tR-+0
9+-+-+-trP0
9-+-+-+-+0
9+-+-+-+-0
xiiiiiiiiy

White is winning simply: 46.h4 Tc3 47.Tf5
Ke4 48.h5 Tc6 49.Kg5 Tc1 50.Tf7 etc.
Marshal wanted to win the game faster.
46.Tf5? Ke4 47.Th5?
The move 47.Tf1 would still be better.
47...Tf3 48.Ke7 Tf4 49.Tg5 Kd4 50.Th5
Kc3 51.Th7 Tf5 52.Kd6 Kb3 53.Th4 Tf3
54.Ke7 Te3 55.Kf7 Te5 56.Tg4
Even better would be 56.Tf4!?
56...Tc5 57.Tf4 Tc7 58.Kg6 Tb7! 59.h4
Tb4 60.Kg5 Ka4! 61.h5 Ka3! 62.h6 Tb8
63.h7 a4 64.Th4 Th8 ½.

FIDE Surveys – Georg Mohr

4

Burgess E. : Capablanca J.R.
New York 1906

XIIIIIIIIY
9-+R+-tr-mk0
9+-+-+-zp-0
9R+-+L+-zp0
9+-+P+-+-0
9-+-+psn-+0
9+-+-+-+P0
9-+-+r+PmK0
9+-+-+-+-0
xiiiiiiiiy

Capablanca played:
38...Se6
His opponent unexpectedly resigned! After
39.Taa8 the game would end in draw!
We have to mention that Capablanca had a
winning position a move earlier: 38...Tg2
39.Kh1 Tc8 40.Lc8 e3 41.Te6 Se6 42.Kg2
Sf4 43.Kf3 Sd5 44.La6 g5!
0:1.

Chajes O. : Capablanca J.
New York 1911

XIIIIIIIIY
9-+-+-+-+0
9+-+-+-+-0
9p+k+-+-zp0
9+p+p+L+-0
9-zP-zP-zP-+0
9+-sn-zP-+P0
9-+-+-+-+0
9+-+-mK-+-0
xiiiiiiiiy

In the game followed:
61.Le6?
White would keep his winning chances after
61.Lc8! Sa2 (61...Kb6 62.Kd2! Sa2 63.f5
and White is winning)
62.La6 Sb4 63.Lc8.
61...Sa2 62.e4?
White had a better possibility: 62.f5!? Sb4

63.f6 Kd6 64.f7 Ke7 65.Kd2 and he would
keep some advantage.
62...Sb4 63.Ld5 Sd5 64.ed5 Kd5 65.f5 b4
½.

Capablanca J.R. : Reti R.
Vienna 1914

XIIIIIIIIY
9-+-+-+k+0
9+-+-+pzpp0
9p+-zpr+-+0
9+-zp-+l+-0
9-+-+-+-+0
9+-zP-vL-+-0
9PzP-+-zPPzP0
9+-+R+-mK-0
xiiiiiiiiy

For a defeat in this kind of endgame, soviet
trainers took away the categories from their
pupils …
25.b4 cb4 26.cb4 h6
It was possible also 26...Lc2 27.Tc1 La4
28.Tc8 Te8 etc.
27.a4 Lc2 28.Ta1 Te4 29.b5!
An attempt of penetration.
29…Ta4 30.Tc1 Le4 31.b6 Kf8?
Who knows why? After the preventive
31...Tb4 White would not win the game.
32.f3 Ld5 33.Tc8 Ke7 34.b7 Lb7 35.Tc7
Black is left without a piece and after a
couple of moves he resigned.

There is a famous statement in the boxing
world: »In boxing there are no invincible
boxers, there are only those who fall rarely«.
Capablanca could relate to this statement.
Even the endgames in which the Cuban
“fell” are instructive …

FIDE Surveys – Georg Mohr

5

Rubinstein A. : Capablanca J.R.
San Sebastian 1911

XIIIIIIIIY
9-+-+-+L+0
9+-+-+-+-0
9-mkn+-+-+0
9+-+-+-+R0
9pzp-+-+-zP0
9+-+P+-zP-0
9Ptr-+-zPK+0
9+-+-+-+-0
xiiiiiiiiy

White has a huge advantage and Capablanca
is left with only one attempt of counter play.
He sacrificed all the pawns on the kingside
and created threats on the queenside.
If Rubinstein would recognize the danger, he
would play a preventive 38.Lc4!, This would
end the battle immediately. But the great
Akiba obviously thought that all moves were
winning:
38.Ld5?
Incredibly, Capablanca did not recognize his
opportunity.
38...b3?
Instead of 38...Ta2!! 39.Th6! (the only
move, which can guarantee a winning point.
After 39.La2 b3 Black would be the one
celebrating; after 39.Lc4 Tc2 40.Tb5 Kc7
41.Lg8 a3 42.h5 a2 43.La2 Ta2 44.h6 Ta6!
The game ends with draw!) 39...Tc2 40.h5
b3 41.Te6 b2 42.La2 Tc1 43.h6 Ta1 44.h7
Ta2 45.h8D b1D 46.Db8.
39.ab3 a3 40.Lc6 Tb3 41.Ld5 a2 42.Th6
1:0.
The counter play with pawns is too late.

Capablanca J.R. : Reti R.
New York 1924

XIIIIIIIIY
9r+-+-+k+0
9zpp+-+-zpp0
9-+n+-+-+0
9+-+-+-+-0
9-+-zp-zP-+0
9+-+P+N+-0
9PzP-+-+-zP0
9+-+-tR-mK-0
xiiiiiiiiy

It is hard to believe that Black lost in this
position after around ten moves. He has
some trouble with the d pawn, but the right
solution is simple: 24...Td8 25.Te4 h6!, then
Kf7, Td5 and Black certainly does not have
to lose.
Let us see how did Reti continue.
24...Kf8 25.Te4 Te8 26.Kf2 h6?
This move is unnecessary. After 26...Te4
27.de4 Sb4 28.Sd4 Sa2 29.Sc2 Sc1 the
knight would successfully return to the play.
27.Se5! Se5
Black already has troubles, which are not
easy to get rid of. From the first sight the
transposition to the rook endgame seemed
like the best idea.
28.fe5 Kf7 29.Kf3 Td8 30.Tg4 g5 31.h4!
The creation of weaknesses!
31… Kg6
Even 31...Ke6 cannot help Black, for
example 32.hg5 h5 33.Th4 Ke5 34.Th5 Kf5
35.Th7.
32.hg5 hg5 33.Ke4 Kh5 34.Tg1 Kh4 35.e6
g4 36.e7! 1:0.

FIDE Surveys – Georg Mohr

6

Capablanca J.R. : Fine R.
AVRO 1938

XIIIIIIIIY
9-+-+-+R+0
9+-+-+p+-0
9-+-+pmk-zp0
9+-+-+-+-0
9-+-+p+PzP0
9+-+-zP-mK-0
9-tr-+-zP-+0
9+-+-+-+-0
xiiiiiiiiy

Sledilo je:
39.g5 hg5 40.Tg5
Capablanca did not see (or he miscalculated)
the known maneuver: 40.h5! Tb1 41.Kg2!
g4 42.h6 Tb5 43.h7 and Black can stop the
pawn only with giving up his rook.
40...Tb8 41.Kh3 e5 42.Tg1 ½.

Capablanca J.R. : Alekhine A.
Buenos Aires 1927

XIIIIIIIIY
9-+-+-+-+0
9+-+-+p+-0
9-+N+-mkp+0
9+-vlP+-+-0
9-+-+-+P+0
9+-+-+K+-0
9-+-+-zP-+0
9+-+-+-+-0
xiiiiiiiiy

Aljehin could choose 55...Ld6! And Black’s
defense cannot be broke down. But he chose:
55...Kg5? 56.Se5!!+- Ld4?
Or 56...La3 57.d6 Kf6 58.d7 Ke7 59.Sf7
Kd7 60.Se5+-.
57.Sf7 Kf6 58.Sd8 Lb6
58...Ke5? 59.Sc6+-.
59.Sc6 Lc5 60.Kf4!+- Lf2 61.g5 Kf7 62.Se5
Ke7 63.Sg6 Kd6 64.Ke4
Black lost another pawn and resigned after a

couple of moves.
1:0.

Eliskases E. : Capablanca J.R.
Semmering 1937

XIIIIIIIIY
9-+-+-+-+0
9+k+-+-+-0
9-zP-+-+-zp0
9mK-+-+-+-0
9-+-+-+-zP0
9+-+L+-+-0
9-+-+-+l+0
9+-+-+-+-0
xiiiiiiiiy

In this simple position Capablanca’s
reaction was wrong:
61.La6 Kc6?
Black would draw after 61...Kb8! 62.Kb4
Lb7! 63.Le2 (the pawn ending after 63.Lb7
Kb7 64.Kc5 h5 ends with a draw) 63...Lg2
64.Kc5 Kb7 65.Kd6 Kb6 66.Ke6 Kc6
67.Kf6 Kd6 68.Kg6 Ke7 69.Kh6 Kf8, with a
transposition to a theoretically drawn
position.
62.Lc8! Lf1 63.Lg4 Ld3 64.Lf3 Kd6
65.Lb7 Le2 66.La6 Lf3 67.Lf1 Lb7
68.Lh3! Ke7
Ali 68...Kc5 69.Lg4 Kc4 70.Le2 Kc5 71.La6
Lf3 72.Lc8+-, Averbah.
69.Kb5 Kd6 70.Lg4! Ke7 71.Kc5 Lg2
72.Lc8 Kd8 73.La6
Ali 73.Le6 in 74.Ld5.
73...Lf3 1:0.

A conclusion: Capablanca was a master of
queen and combined endgames. In the
endgames with light pieces he played
weaker.His weak points were the rook
endgames, which were for a long period of
time believed to be his strong points.