T. Erdelyi - V. Rajlich
[A24/07] English: Carls (Bremen)
10 Jan 2000
|
1.Nf3 Nf6 2.c4 g6 3.Nc3 Bg7 4.g3 0-0 5.Bg2 d6 6.0-0 e5 7.d3
| My opponent and I had reached this exact position
in the previous tournament, but in that game I misplayed it with 7...Nh5
8.Rb1 c6 9.b4 d5 10.Nd2 d4?! 11.Nce4, etc., and White is better. |
| I did not prepare for this move. Ehlvest-Topalov
from the ’96 Olympiad continued instead 9.e4 a5 10.a3 Na6 11.b4 axb4 12.axb4
h6 13.Be3 Nc7 14.b5 Ne6 with rough equality. |
| I am not sure what I think about this move.
My opponent, and IM who has a lot of experience with these systems, felt
that 10...d4 was slightly inaccurate, claiming that White’s subsequent
play on the queenside seems a bit easier to carry out than Black’s play
in the center and on the kingside. |
| Logical but seemingly inferior would have been 13.Bg5
Qc7 14.Nf6+ Bxf6 15.Bxf6, as I don’t see White’s response to 15...h6. |
| I rejected the immediate 15...Nd5 because I felt
that the trade of knights which would occur after 16.Nd6 Nc3 17.Qc2 Nxb1
18.Nxe8, etc., would favor White. Neither of Black’s bishops is very good
in this position. |
| An interesting move. It looks a little bit
awkward, but the Nb6 tactic is annoying to deal with. Black could
just play 16...a6 or 16...Bd7, but neither of those moves is very impressive
and in both cases I think that it’s pretty clear that White stands at least
somewhat better. Or, Black can turn the game into a mess. |
| According to my analysis, the complications resulting
from 16...Nd5 17.Bxd5+ peter out into a draw, in which case 16...Nd5 is
best. |
| Possibly White should refrain from this move and
try to claim a positional advantage after 17.Bd2 Nc7, but it’s probably
just roughly equal in that case. |
| So far Black hasn’t had to calculate much but now
the situation has become very concrete. 21...bxc5 22.bxc5 Qxc5 23.Rxb7
seems to be good for White, i.e. 23...Qc2 24.Rb8. 21...bxc5 22.bxc5
Bxc5 23.Rxb7 is less clear, but White also seems to emerge on top: 23...Qe6
24.Rc7 (not 24.Rb8 Bd6) 24...Qh3 25.Qxc6, etc. |
| So, how does Black continue? |
21...Qe6
| This funny-looking move seems to be the only way
for Black to demonstrate compensation. |
| And this in turn seems to be the only adequate defense
(though 22.Bd2 Qh3 23.Kf2 would transpose). 22.g4 is met with 22...h5
when Black’s attack is very strong, while 22.Rf2 Qh3 23.Bd2 Qxg3+ 24.Rg2
Qxg2+ 25.Kxg2 Bh3+ leaves Black with a much better ending. |
| Forced, Black threatened 23...Qxf1+ and 23.Bg5+
lost to 23...Qh2+ 24.Ke1 Qxg3+ & 25...Qxg5 while 23.Qa3 is met with
just 23...Qh2+ 24.Ke1 Bh3. |
| Now, Black has another critical decision.
It seems hard to believe that White can survive this, but that is in fact
what appears to be the case. |
| 24...Bh3 25.Qxe8 Bxf1 26.Qe6+ is at least a draw
(and probably more) for White, so Black will need to continue his attack
with ...Bd7 first and only then 25...Bh3. The critical question is,
should Black play ...bxc5 (when the reply bxc5 is forced) and/or ...e4
(when the reply dxe4 is forced) before continuing with ...Bd7 and ...Bh3.
It seems to be a draw in all cases. |
| Ok, Black could also have had this position without
the two b-pawns (by first playing 24...bxc5) and/or with his e-pawn missing
and White’s d3 pawn on e4 (by first playing 24./25...e4). In all
of these cases, he has five winning tries, none of which seem to work.
The first three are straightforward: |
| A) 26...Bxf1 27.Kxf1 Qh1+ 28.Kf2 Qxb1 29.Qxe8
and regardless of whether 24...bxc5 or 24...e4 were thrown in White is
just winning. |
| B) 26...Qg2 27.Kd1 Qxf1+ 28.Kc2 Qxe2 29.Qxe8
and White is winning, though with a 24...e4 25.dxe4 sequence thrown in
Black would have the extra possibility of 29...d3+ (well, 30... with the
extra move pair) but then 30.Kc3 and White is safe. |
| C) 26...Re6 27.Qd5 Qg2 28.Kd1 Qxf1+ 29.Kc2
Qxe2 (29...Qg2 and if Black didn’t play 24...bxc5 then 30.Bxc5 here would
be good while if he did then 30...Rb8 [actually 31...] would win)
30.g4 Kf2 31.Re1 and White wins. |
| However, as it turns out, Black has another winning
theme, based on the fact that he actually has two seperate unconnected
threats to win a rook (...Bxf1 & ...Qh1+, and ...Qg2 & ...Qxf1).
So, can he just move the e8 rook out of danger, or protect it? Unfortunately,
these too seem to fail. |
| For starters, there is no good square for the rook.
e7 is best but after 26...Re7, then if 24...bxc5 hadn’t been played then
27.cxb6 draws, i.e. 27...Bxf1 28.Kxf1 Qh1+ 29.Kf2 Qxb1 (29...Qh2+, etc.,
is the draw) 30.b7 and White wins, or 27...Qg2 28.Kd1 Qxf1+ 29.Kc2 Qxe2
30.b7, etc. If on the other hand 24...bxc5 had been played, then
26...Re7 (again, actually 27...Re7 in this case) 27.Rb8 Qg2 28.Bh6 wins. |
| Along the same lines, though, is Black’s toughest-to-crack
winning try: 26...Kf7 (Of course this must be played without 24...bxc5).
The point is the following tactical resource: 27.cxb6 Qg2 (27...Bxf1 still
only draws since 28.Kxf1 Qh1+ 29.Kf2 Qxb1 30.b7 with 31.Qxe8+ next leaves
White with the only chances) 28.Kd1 Qxf1+ 29.Kc2 Rc8 30.Qxc8 Qxb1+ 31.Kxb1
Bxc8 and though this ending isn’t completely clear it is at least definitely
Black who is playing for the win. However, White has the following
counter-resource: 27.cxb6 Qg2 28.Qe5+ Kg7 29.Bh6+ Kxh6 30.Kd2 and if 30...Bb4+
then 31.Rxb4 Qxf1 32.b7 wins. So, this line too fails to improve
on the draw. |
| Whew! Let me take a step back. I find
things interesting at a higher level. Of course when a game which
you play explodes into practically unfathomable complications, you try
to figure them out. At the same time, though, you don’t see everything,
so as you make your decisions you’re also managing risk. Sometimes
it’s kind of scary in retrospect, like some angel is watching over you.
For example, when I played 16...Nd5, I didn’t even come up with my 21st
move. I could have, I guess, and some people would have, it’s all
forced if White opts for 17.Bxd5+. But sometimes I like to trust
my instinct and wait to have the position sitting on the board, if it seems
promising enough. Besides, White may end up deciding against 17.Bxd5+. |
| At any rate, even after I played 21...Qe6, I still
wasn’t completely sure that White couldn’t consolidate and win. An
exchange is an exchange, who knows, maybe 22.g4 h5 and White can hold on.
In fact I am still not completely sure. Basically there is no way
around the fact that this sort of play is speculative. Because you
can’t calculcate it all out, you’re always to some extent guessing.
And you could be wrong. But for your risk there is a reward.
For example, in this game, at the board I did most of my calculating after
White had already played 24.Ke1, and by then Black can already force an
immediate draw, so there it is White with the risk. Of course according
to his calculations everything holds together, but it’s complicated and
Black then has a chance to find a flaw. Anyway, this risk tradeoff
is what needs to be assessed. When you sac material and you know
you’re seeing only a fraction of the lines, are the chances on your side?
This is how the question needs to be phrased. |
| At any rate, back to the game... or what little
is left of it. |
|
|
