MICHIGAN CHESS ASSOCIATION
Analysis by Vasik Rajlich
Todd Rowland and I have taken a good
look at this game and as a result these notes are much improved. When I
use Todd's notes directly they are
cited in blue. There's a lot here on
the subject of piece vs pawns imbalances.
Rajlich,V.-Nickoloff,B.(2540 CFC) (Toronto
Open, Round 5)
Sicilian: Najdorf [B96/09 (3rd ed.)]
1998.04.12. CAN Toronto [TC:40/120; G/60]
1.e4 c5 2.Nf3 d6 3.d4 cxd4 4.Nxd4
Nf6 5.Nc3 a6 6.Bg5 e6 7.f4 Qc7 8.Qf3 b5 9.f5
Black
to play
Even though this move is the most direct and natural
it is in fact not the most topical since the resulting complications peter
out into a position which is somewhat drawish. 9.Bxf6 and 9.0-0-0 are most
frequently played.
9...Nc6
9...b4 is the critical move. Nunn then gives 10.Ncb5
an "!" and doesn't even consider anything else. Before I deal with his
analysis, I will give my analysis of the move which I intended to play,
10.Nd1. I have not been able to find a way for Black to equalize against
it. He can try:
A) 10...Nc6 11.Nxc6
Qxc6 12.Bxf6 gxf6 13.Bd3 when White should stand better.
B) 10...e5 11.Bxf6
gxf6 12.Ne2 when White's plan of Ne3 and Bc4 and perhaps Ne2-g3-h5 should
give him a solid positional edge.
C) 10...Be7 (Daring
White to try 11.fxe6 fxe6 12.e5) 11.fxe6 fxe6 12.Ne3 (12.e5 dxe5 13.Qxa8
exd4 would be tremendous for Black.) when Black has:
C1) 12...Nxe4? 13.Qxe4
+-
C2) 12...Qc5? 13.e5
C21) 13...Qxd4 14.exf6
gxf6 15.Rd1 +-
C22) 13...dxe5 14.Qxa8
exd4 15.Qxb8 +-
C23) 13...Nd5 14.Bxe7
Kxe7 15.Ndf5+ +-
C3) 12...Nc6?! 13.e5
C31) 13...Nxd4 14.Qxa8
Ng4 (I can't imagine that 14...dxe5 would leave Black with sufficient compensation.)
15.exd6 +-
C32) 13...Nd5 14.Nxc6
Qxc6 (14...Bxg5 15.Qh5+ g6 16.Qxg5 +-) 15.Bxe7 Kxe7 16.Bc4 Bb4 (16...Rf8
17.Nxd5+) 17.0-0-0 +-
C4) 12...0-0
13.Bc4
C41) 13...Kh8 14.0-0
and there will be no useful discoveries, for example 14...Nfd7 15.Bxe7
should be good for White.
C42) 13...Nxe4 14.Qxe4
Bxg5 15.Bxe6+ Bxe6 (15...Kh8 16.Nd5 should be excellent for White.) 16.Nxe6
Re8 (16...Qe7 17.Nd5 +-) 17.Nxc7 +-
C43) 13...Nd5 14.Nxd5
+-
Nunn's 10.Ncb5 (in response to 9...b4) enters
a fairly forcing line: 10...axb5 11.Bxb5+ Bd7 12.fxe6 Bxb5 13.Nxb5 Qc5
14.Bxf6 Qxb5 15.Bxg7 Bxg7 16.Qxf7+ Kd8 17.Qxg7 Re8, leading to the following
position:
White to play
Two moves - 18.Rf1 and 18.0-0-0 - have been heavily
analyzed.
18.Rf1 defends the e-pawn, since 18...Rxe6 19.Rf8+
Re8 20.0-0-0 would be awesome for White; but it runs into 18...b3, when
after 19.cxb3 Black will choose either 19...Qb4+ or 19...Qa5+. According
to Nunn, White will have a hard time winning for two reasons: 1) His king
is very exposed to threats of perpetual. 2) Even when White gets four or
five pawns for the piece the endgame is often a relatively easy draw since
White's pawns are scattered. White can also respond to 18...b3 with 19.0-0-0
Qc5 (20...bxa2 21.Rxd6 wins for White.) 21. axb3, when black will still
have play against the white king.
18.0-0-0 does avoid the problems with king safety
which 18.Rf1 introduces but on the other hand it allows Black to play 18...Rxe6.
After 18.Rf1 Rxe6 Nunn gives 19.Qg8+ Qe8 as comfortable for Black without
further analysis, dismisses 19.Rhf1 Nd7 as better for Black without further
analysis, and suggests 19.Rd5 as White's best. I want to concentrate on
the position arising after 18.0-0-0 Rxe6 19.Rhf1 Nd7:
White
to play
White's last two moves are quite natural - 18.0-0-0
safeguards the king, while 19.Rhf1 brings the last piece into what seems
like a possibly promising attack against the black king - so if Black stands
better it is noteworthy. It is also noteworthy that Nunn doesn't even bother
to justify his claim that Black stands better. Is it really so obvious?
After all, White can claim the following as compensation for his piece:
1) He is about to
collect a fourth pawn. (i.e. 20.Qxh7, since then 20...Rxa2 would lose to
21.Qg8+) for the piece.
2) He will have two
connected passed pawns on the king's side.
3) Black's king seems
to be more exposed than white's.
Is it possible that this is not enough?
I think that this position needs to be looked
at differently. Pawns definitely can provide compensation for a piece,
as several positions arising later in the game will show, but in order
for them to do so one of the following conditions needs to be met:
1) The pawns dominate
the center.
2) There are few
or no remaining pieces or the side with the pawns can force a position
in which there are few or no remaining pieces.
3) There exists some
other form of compensation.
4) The side with
the pawns has a ton of them.
As long as there is a certain volume of material
on the board, the extra piece will provide constant attacking chances,
and this is true even when the king of the side with the extra piece appears
to be more exposed. In the lines arising from the diagrammed position it
is repeatedly White's king which is in greater danger, 20.Qxh7 Kc7 (20...Rxa2
21.Qg8+ +-) and now:
A) 21.Rf7?! Rxa2
22.Kb1 Ra7 with 23...Qa4 next when Black should be much better.
B) 21.Kb1 Qa5 22.a3
bxa3 23.b3 Qc3 -+
C) 21.Rf5 Qc6 accomplishes
nothing for White.
D) 21.Qf7 (Awkward
but it defends a2 and thus seems best.) and now:
D1) 21...Qc4? and
then both 22.Rf6 and 22.Rxd6 give White a ton of play
D2) 21...Rf8? 22.Qxe6
Rxf1 23.Qxd6+ +-
D3) 21...Rxe4? 22.Qg6
(22.Rxd6 Kxd6 23.Qg6+ Kc7 -+) 22...Qe5 23.Rxd6 Qxd6 24.Qxe4 and now 24...Rxa2
can't be played due to 25.Qa2+.
D4) 21...Qe5 is best.
It leaves white with two ways to trade queens. Unfortunately, Black's attack
is relentless even after a trade of queens:
D41) 22.Qf4 Rxa2
23.Qxe5 Rxe5 24.Kb1 (24.Rf4 Ra1+ 25.Kd2 Rxd1+ 26.Kxd1 Nc5 -+) 24...Ra8
25.Rf4 (25.Rd4+ Rea5 -+) 25...Rea5 26.Kc1 (26.c3 bxc3 27.bxc3 Ra1+ 28.Kc2
K8a2+ -+) and now:
D411) 26...Ne5? 27.b3
and all of a sudden Black's attack is stuck
D412) 26...Rea5 and
White doesn't have a good move:
D4121) 27.h4 (Show
me!) 27...Ra1+ 28.Kd2 R1a2 is awesome for Black: 29.Kc1 Na4 30.b3 Nc3 and
White has been shown
D4122) 27.b3 Ra1+
28.Kd2 Nxb3+ 29.cxb3 R8a2+ -+
D4123) 27.Rf7+ Kc6
28.Rf6 Ra1+ 29.Kd2 Nxe4+ -+
D4124) 27.e5 dxe5
28.Rxb4 Ra1+ 29.Kd2 Rd8+ -+
D42) 22.Kb1 Qxe4
23.Qf3 Nc5 (This knight will be awesome on e4.) 24.Qxe4 (Else 24...b3 25.axb3
Nxb3) 24...Nxe4 and Black's attack looks awesome: 25.Rf4 Ra5 with the idea
of ...d5 and ...Ra6, or 25.Rf5 Kc6 with the idea of ...Re7 and ...Rea7
The themes running through these variations are:
1) Black has a better
attack than White because of his extra piece despite his more exposed king.
2) Even the presence
of only a pair of rooks gives Black enough firepower to make serious threats.
Hence:
1) White shouldn't
bother to count as a plus the slight exposure of Black's king as with Black
in posession of the extra piece White's attempts at attacking will be ineffective.
2) Until White forces
wholescale simplification (as opposed to merely forcing a trade of queens)
he should expect to come under heavy fire.
As will be seen shortly, these themes persist
in the piece vs pawns positions which arise in this game.
Back in the game, after 9...Nc6, the following
position arose:
White
to play
10. Ncxb5?!
9...Nc6 is dismissed by Nunn as inferior due to
10.Nxc6 Qxc6 11.Bxf6 gxf6 12.Bd3 (12.fxe6 fxe6 13.Qxf6 Rg8 would give Black
excellent compensation for the pawn), when White stands better since he
will castle king's side and have a strong attack against Black's king regardless
of where Black chooses to place it.
Nunn incorrectly dismisses 10.Ncxb5 with the variation
10.Ncxb5 axb5 11.Bxb5 Bd7 12.fxe6 Nxd4 13.Bxd7+ Ke7 -+. In fact, the correct
result of this move is a draw by perpetual, as will be shown below.
10...axb5 11.Bxb5 Bd7 12.Bxf6
Necessary, since inadequate were:
A) 12.fxe6? Nxd4
13.Bxd7+ Ke7 -+
B) 12.e5? Qa5+ (From
White's perspective, this is an unfortunate resource, since 12...Nd5 would
get crushed by 13.Bxc6 Bxc6 14.0-0) 13.Bd2 Nxd4 -+
12...gxf6
White to play
Now, White has:
A)
13.e5? Qa5+! (13...Rc8 14.fxe6 fxe6 [14...Nxd4 15.Bxd7+ Ke7 16.Qxf6+ +-]
15.Qh5+ Ke7 [15...Kd8 16.Bxc6 Bxc6 17.Nxe6+ +-] 16.exf6+ +-; 13...Nxd4
14.Qxa8+ Ke7 [14...Qd8 15.Bxd7+ Kxd7 16.Qa7+ +-] 15.exf6+ Kxf6 16.0-0 and
White has excellent compensation for his small material deficit; 13...d5
might also be good for Black) 14.b4 Nxd4 15.Bxd7+ Kd8 and everything holds
together, since 16.Qb7 is met with 16...Qxb4+ 17.Qxb4 Nc2+ etc.
B) 13.fxe6? Nxd4
14.Bxd7+ (14.Qxf6 Bxb5 -+) 14...Ke7 -+
C) 13.Qc3? Rc8 14.fxe6
fxe6 and White has no way to take advantage of the fact that he has not
had to play Bxc6 in this variation.
D) 13.Bxc6 (clearly
correct), whose consequences are analyzed below.
13. Bxc6 Bxc6 14. fxe6
Black
to play
This is the critical position for the evaluation
of 10.Ncxb5. Black has:
A) 14...fxe6, as
in the game (and thus analyzed below), leading to an endgame in which White
has a significant advantage.
B) 14...0-0-0 (Also
apparently inaccurate.) 15.exf7 d5 (15...Be7 16.Ne6 Qa5+ 17.b4 Qxb4
18.c3 Qb2 19.0-0 Rd7 20.f8Q+ Bxf8 21.Qxf6 +-) 16.Qxf6 and Black has no
suitable continuation: 16...Bd6 17.Ne6 Be5 18.Qxd8+ , or 16...Kb7 17.Ne6
Qa5+ 18.c3 and 19.Qxh8 is next.
C) 14...Be7 15.exf7+
Kxf7 16.Qh5+ Kg8 when White should accept the draw with 17.Qg4+ Kf7 18.Qh5+.
Attempting to play for more with 17.Nf5 would again bring about the piece
vs pawns imbalance discussed above and the themes that come with it:
Black
to play
Again, White could look at this position optimistically,
since along with three pawns (which in theory should alone almost compensate
for the piece) he has exposed Black's king and locked his KR out of play.
Unfortunately, as before, the position needs to be viewed differently:
1) White is down
material.
2) White's pawns
don't dominate the center and thus don't do enough to contain Black's pieces.
3) White is nowhere
near an endgame in which his pawns would be relatively strong .
4) Black's inactive
KR and exposed king are only temporary problems and because White is down
a piece he is in a poor position to exploit them.
5) White has weaknesses.
White's two forms of compensation for the piece
- his three pawns, and Black's exposed king and sidelined KR - do not cooperate
with each other.
Black's options are:
A) 17...Be8? 18.Nxe7+
Qxe7 19.Qd5+ +-
B) 17...Kf8 18.Qh6+
(18.Nh6 Be8 and white's pieces are misplaced; 18.0-0 Bxe4 White has no
compensation.) 18...Ke8 19.Qg7 Rf8 20.Qxh7 Qa5+ 21.c3 Qe5 and now:
B1) 22.Nxe7 Qxe4+
23.Qxe4 Bxe4 and Black keeps the bishops
B2) 22.Qxe7+ Qxe7
23.Nxe7 Bxe4 etc.
B3) 22.0-0 Bxe4 and
Black is safe and better
B4) 22.Ng7+ Kd8 and
Black is also safe and better
C) 17...Bf8 18. 0-0
and now:
C1) 18...Qb6+? 19.Kh1
Qxb2? 20.e5! is a pretty awesome tactic.
C11) 20...fxe5??
21.Nh6+ Bxh6 22.Qf7mate or 21...Kg7 22.Rf7 mate.
C12) 20...dxe5? 21.Qg4+
Kf7 22.Qc4+ Kg6 23.Qxc6 is pretty strong.
C13) 20...Qxe5 21.Rae1
Qc5? 22.Qg4+ Kf7 23.Nh6+ Bxh6 24.Qe6+ mates.
C14) 20...Qxc2 21.Qg4+
Kf7 22.Nxd6+ Ke7 (22...Bxd6? 23.Rxf6+) 23.Rxf6 is also pretty strong
for White.
C15) 20...Bxg2+ is
Black's last resort.
C2) 18...Be8 (Also
not best but only because white gets another chance to force a draw.) and
now:
C21) 19.Qd1 Rc8 20.Rf2
Bg6 21.Qd4 Bxf5 22.Rxf5 Bg7 should be good for Black. He will play 23...h5
and 24...Rh6 unless White disrupts this with 23.Raf1 in which case he will
play 23...Kf7 (23...h5 24.Qd5+ Kf8 [24...Kh7 25.Rxh5+ +-] 25.Rxf6+ Bxf6
26.Rxf6+ Ke8 27.Qe6+ Kd8 28.Rf7 Qb6+ 29.Kh1 Rc7 30.Qf6+ +-) 24.Qd5+ Kf7
and now if 25.Rg5 Rg8 when Black should again stand better.
C22) 19.Ne7+ Kg7
(19...Bxe7 20.Qd5+ +-) 20.Nf5+ (20.Qd5 Qa7+ 21.Kh1 Bxe7 -+; 20.Qg4+ Kf7
21.Nd5 [21.Rxf6+ Kxe7 22.Qe6+ Kd8 -+] 21...Qc5+ 22.Kh1 Be7 when Black is
fine) 20...Kg8 21.Ne7+ etc., ½-½.
C3) 18...Qf7 (Seems
best.) 19.Nh6+ Bxh6 20.Qxh6 and now:
C31) 20...Bxe4 (Seems
to lead to too many trades.) 21.Rxf6 Qg7 22.Qxg7+ Kxg7 23. Rxd6 and White
might draw.
C32) 20...Rf8 (This
seems best.) 21.Rf4 (21.Rf3 Bxe4) 21...Qg6 and 22...h5 and Black has easily
ended White's "attack" and is ready to use his extra piece to begin his
own.
It's the same story as with the 18.0-0-0 Rxe6
19.Rhf1 Nd7 line analyzed previously. With significant material on the
board and with the pawns not dominating the center, Black's extra piece
enables him to easily deal with the slight initial problems to his king
and eventually obtain excellent pressure. On the other hand, with endgames
with queening themes far away, White's extra pawns contribute very little
to his cause.
In playing 14...fxe6 my opponent follows these
principles. He avoids the draw which 14...Be7 would allow, and he is not
concerned about his exposed king, as White is down a piece. He expects
that his temporarily disorganized pieces will eventually coordinate an
attack against the white king. Unfortunately, White has a tactic which
forces Black to choose between allowing a massive set of trades (and making
White's pawns very powerful) and giving White a fifth pawn for the piece.
14...fxe6?! 15.Qxf6 Bg7 16.Qxe6+ Kd8 17.Nxc6+
Better than 17.0-0-0 Bd7 18.Qd5 (18.Qf7 Be5) 18...Ra5
when Black has consolidated and should stand no worse.
17...Qxc6 18. 0-0-0 Kc7
18...Ra6 19.e5 +-
19.Rhf1 Rhf8!
So far, the pawns have not done well in this debate.
Are three or even four pawns just generally less valuable than a piece?
It's time for them to speak up.
At first glance, 19...Rhf8 appears to be a
mistake since with 20.e5 White can force Black to part with yet another
pawn. However, regardless of the assessment of the position arising after
19...Rhf8, 19...Rhf8 is at least more stubborn than the alternative, 19...Rhd8.
After 19...Rhd8 20.Rf7+ Rd7 21.Rdf1 Rad8 22.Rxd7 Rxd7 23.Rf7 Be5 24.Rxd7
Qxd7 25.Qxd7+ Kxd7 26.g3, the following position would arise:
Black
to play
In this endgame, White wins.
Hopeless for Black are:
A) Marching his king
straight to the kingside, i.e.
26...Ke6 27.c3 Kf6 28.Kc2 Kg5 29.a4 and Black
is already in the void.
B) Marching his king
straight to the queenside, i.e.
26...Kc6 27.c3 Kc5 28.Kc2 Kb5 29.Kd3 Kc5 30.b3
Kb5 31.a4+ Kc5 (31...Ka5 32.Kc4 h6 33.Kd5 Kb6 34.c4 Ka5 35.c5) 32.b4+ Kb6
33.c4?! (33.Kc4 may be simplest.) 33... Bf6 34.a5+ Ka6 35.Kc2 Be5 36.Kb3
Bd4 37.Ka4 Be3 38.b5+ Kb7 39.g4 Bf2 40.h3 h6 41.a6+ Kb6 (41... Kc7 42.Kb3
Kd7 43.h4 Ke6 44.g5 hxg5 45.hxg5 Kf7 46.c5! Bxc5 47.Kc4 Bg1 48.Kd5 Bc5
49.e5 and White wins.) 42.Kb3 Bd4 43.Kc2 Bf6 44.Kd3 Bg7 45.Ke3 Be5 46.Kf3
(zugzwang) 46...Ka7 47.h4 Bf6 48.g5! hxg5 49.h5 Bg7! 50.Kg4 Bh6 51.Kf5
Kb6 52.e5 dxe5 53.Kxe5 g4 54.Ke4 Bc1 55.Kd3 Bh6 56.Ke2 Bc1 57.Kf2 Bf4 58.Kg2
(zugzwang) 58...Ka7 59.c5 Kb8 60.b6 Be3 61.h6 Bxc5 62.h7 Bd4 63.Kg3 and
White wins by bringing his king to g8.
Or 61...Bxh6 62.c6 Bf4 63.a7+ Ka8 64.c7 wins.
So Black needs to play 26...Bd4 27.c3 Bg1 28.h3
Ke6 29.Kd2 Ke5 30.Kd3 Bf2, leading to the following position:
White
to play
When White has 31.a4! Bxg3 32.Kc4! when Black
can't hold the position:
A) 32...Ke6 33.Kb5
Kd7 34.Kb6 Kc8 and White will win easily since after an eventual c4-c5
he will have passed a, b, and e pawns.
B) 32...Kxe4 (This
is less logical than 32...h5 but Black has a few tricks that need to be
considered.) 33.a5 and now:
B1) 33...d5+ 34.Kb5
Bf2 (34...d4 35.a6 Bb8 36.cxd4 Kxd4 37.b4 +-) 35.Kc6 is indeed winning
for White as claimed: 35...h5 36.b4 h4 37.b5 Kf4 38.b6 Kg3 39.a6 Kxh3 40.a7
+-.
B2) 33...Bf2 34.Kb5
h5 (34...Kd5 35.c4+ and White gets into c6 anyway; 34...Kd3 35.b4 Kxc3
36.b5 Kxd5 +-) 35.Kc6 h4 36.b4 Kf4 37.b5 Kg3 38.b6 Kxh3 39.a6 Kg2 40.a7
h3 41.a8Q h2 45.Kxd6 +-
C) 32...h5 33.a5
h4 34.b4 Bf2 and now White has two ways to win:
C1) 35.Kb5 Kf4 36.a6
Kg3 37.c4 Ba7 (37...Kxh3 38.c5 Kg2 39.a7 +-) 38.c5 dxc5 39.bxc5 Kxh3 40.c6
+-
C2) 35.b5 Kf4 36.b6
Kg3 37.Kb5 Kxh3 38.a6 Kg2 39.a7 +-
When playing the side with the piece against
three or four pawns one should probably almost categorically avoid trading
all of the remaining pieces, regardless of the cost.
20.e5 Rxf1 21.exd6+
Best. 21.Rxf1 Bxe5 22.Rf7+ Kb6 23.Qb3+ Ka6 leads
to nothing.
21...Kb6
White
to play
22.Rxf1?
White should clearly have played 22.Qb3+ Qb5 (Having
the extra piece doesn't make your king completely immune: 22...Ka6 23.Rxf1
Qxd6 23.Qa4+ Kb7 24.Qe4+ [24.Rf7+ Kb6 is unclear.] 24...Kb8 25.Qe8+ Kb7
26.Qf7+ Qc7 27.Qxc7+ Kxc7 28.Rf7+ +-) 23.Qxb5+ Kxb5 24.Rxf1 Ra7 when whether
Black can draw is unclear. Clearly, however, this is the endgame with which
he should take his chances, as the presence of rooks makes White's job
far more difficult than in the endgame arising after 19...Rhd8.
22. Rxf1+ leaves White with nothing better than
a draw. While this was pretty surprising to me at the time, it really shouldn't
have been. White is banking on taking advantage of an exposed black king
while he has one piece less to throw at it than Black has to defend or
counterattack with. Furthermore, it is horribly incongruous for White to
be looking for play against Black's king when he should be looking to take
advantage of his five extra pawns by exchanging queens.
22...Qb5 23.d7+
Harmless or bad were:
A) 23.Qe3+ Ka6 24.Qa3+
Kb6 ½-½.
B) 23.Qb3 Qxb3 24.cxb3
Ra7 =
C) 23.Rf4? Re8
The main alternative was:
D) 23.Rf7 when Black
has two ways to draw:
D1) 23...Bxb2+ (To
give Black escape squares on the queenside.) 24.Kd2 Re8 25.d7+ Rxe6 26.d8Q+
Kc5 and now:
D11) 27.Rc7+ Kb4
and Black's king will be safe on a3 and a4.
D12) 27.Rf5+ Re5
28.Qc7+ Kd4 29.c3+ Ke4 and Black is again OK since 30.Rxe5+ Qxe5 31.Qb7+
Qd5+ holds everything.
D2) 23...Qxb2+ 24.Kd1
Qb1+ (24...Qa1 [or 24...Qd4+] 25.Ke2 Qe5+ 26.Qxe5 Bxe5 27.d7 is awesome
for White.) 25.Kd2 Qb4+ 26.Ke2 (Else the position repeats.) 26...Qb5+ and
now:
D21) 27.Kf2 Bd4+
28.Kg3 (28.Kf3 Qf1+ 29.Kg3 Be5+ transposes to "B") 28...Qg5+ 29.Qg4 (White
should actually play 29.Kh3 Qh5+ 30.Kg3 and take the draw here. I'm only
trying to show that White can't win.) 29...Ra3+ 31. f3 Bf2+ 32.Kh3 Qh6+
-+
D22) 27.Kf3 Qf1+
28.Kg3 Be5+ 29.Kh4 (29.Kh3 Ra3+ 30.Kh4 Ra4+ transposes.) 29...Ra4+ 30.Kh5
(Again, 30.Kh3 Ra3+ 31.Kh4 ½-½ is better.) 30...Qe2+ 31.Kg5
Bf4 -+
23...Kc7 24.Rf4
24.Rf7 Rd8 would be harmless while 24.d8Q+ Rxd8
25.Rf7+ Kb8 would win for Black.
24...Qxb2+ 25.Kd1 Qa1+ 26.Ke2 Qe5+ 27.Qxe5+ Bxe5 28.Rf7
Rxa2 29.Rxh7 Rxc2+ ½-½
After 30.Kf3 either 30...Rd2 or 30...Bxh2 would
be simple draws.