Wednesday, November 23, 2011

Whatever happened to the Polugaevsky Variation?

1.e4 c5 2.Nf3 d6 3.d4 cxd4 4.Nxd4 Nf6 5.Nc3 a6 6.Bg5 e6 7.f4 b5

When I was young, fresh and foolish, as opposed to old, tired and cynical, I used to play the Sicilian Najdorf. In fact I seem to have played it almost right the way through my teenage years, something which surprised me when I looked it up in the course of writing this piece. I'm not sure I learned anything from those years playing the Najdorf other than I'd learned nothing from those years playing the Najdorf, so after I started playing chess again, a couple of years after university, I dropped it. But, during those teenage years, my problem was, as with any other Najdorf player, what to do against 6.Bg5, since of the many possible alternatives, none seemed to be quite secure.

I seem to have mostly played the Poisoned Pawn with 6...e6 7.f4 Qb6, although eventually I moved over to the old main line with 6...e6 7.f4 Be7. But I looked at more or less everything at one time or another. Including, although I never actually played it, the Polugaevsky Variation with 6...e6 7.f4 b5.

I was, almost certainly, first attracted to the line by the long discussion of its development that takes up more than a quarter of Lev Polugaevsky's book, Grandmaster Preparation, which was issued by Pergamon in an English translation (by Ken Neat) in 1981 and swiftly recognised as a classic. The work the author had put into its creation
Sometimes you see books that have been written in one month. I don't like that. You should take at least two years for a book, or not do it all
was evident, and in some ways mirrored by the twenty years, off and on, he had been prepared to put into the variation that took his name, trying always to keep alive a line too tactical in nature to be anything but fragile.

In his long chapter The Birth Of A Variation, Polugaevsky* went from its birth almost right up to the time of publication, to the point where two extra short sections had to be added just before the book came out, one, by Neat, giving these two games, and an addendum by Polugaevsky giving this one (as did Neat) as well as this one from the same match, and this one from later in the year.

Anyway, if I try and give too many of the games that are referred to in the book, I will never finish this piece and you will never read it, so onwards. Polugaevsky devotes most of his time to what he considered the "most critical" continuation, 8.e5 dxe5 9.fxe5 Qc7

after which there are two lines which most concern him, 10.exf6 and 10.Qe2. After the first of these, the game normally continues 10...Qe5+ 11.Be2 Qxg5 12.O-O (there are alternatives, of course, but we're skimming rather than swimming here) after which Black looked to be in a lot of trouble

until Polugaevsky came up with 12...Qe5!

which (after many games and much analysis) looked satisfactory for Black.

However, scarcely had that resource been discovered than Polugaevsky had to find a reply to 12.Qd3 instead

and although he found a response, it proved inadequate: the blow being sufficiently severe that he abandoned the variation (or, as he has it, The Variation) for more than a decade.

I won't go into the whys and wherefores, or what he discovered that brought him back to his variation, but rather observe that he was also discouraged for a long while by the other major line he discussed after 9...Qc7, which was 10.Qe2.

I mean why not keep more pieces on, if almost all of yours are developed and almost none of the opponent's? Play then usually went 10...Nfd7 11.O-O-O Bb7 12.Qg4 (though other ideas are given, notably 12.Qh5 and 12.Nxe6). Then Black's two ideas - both resulting in "colossal complications" - were 12...Qb6 and 12...Qxe5.

Never mind the colossal complications, one of those options always looked ludicrously dangerous to me, while the other looked to be another move by a piece already developed without the virtue of grabbing any material for it.

To be honest, I don't recall whether I ever seriously intended to play the Polugaevsky, and if so, whether I'd opted for 12...Qb6 (I am sure I was never in a million years going to take on e5) or for another line that Polugaevsky tried, in the 1979 Riga Interzonal. This involved, instead of 11...Bb7, Black playing 11...Nc6

which he did in his game against Grunfeld (given above, and analysed in the extra section added by Ken Neat).

Who knows? It never happened. I never played it, not even once. I took much of the Eighties off, as far as playing chess was concerned, and when I came back to the board at the end of the decade, although there was a new world champion who championed the Najdorf, he wasn't one to play the Polugaevsky. And, although I didn't really think about it at the time (I was on to safer, Scheveningen ground by then) I don't recall seeing any other leading players try their luck with 7...b5. Not then, not since and not right now.

So it's been more than twenty years since I had any real knowledge of contemporary Najdorf theory - thank God - give or take a brief reacquaintance with it during the Kasparov-Short match, though not the 6.Bg5 line (unless you count this).

Where did it all go, eh?

So, as far as I am concerned, it's as if I fell asleep when the Polugaevsky Variation was alive and thriving: then, when I woke up, having forgotten all about it, it was no longer there. And now I've remembered it, and I'm wondering what happened and where it went. Was it refuted? Did it go permanently out of fashion? Or is it, in fact, still being played quite happily by grandmasters even today, but as I'm neither a Najdorf player nor a reader of databases, I just don't know about it?

[* I've chosen to render the surname thus, rather than Polugayevsky, even though the latter variation is the one given in the book. Both have been used previously on this blog!]

[Polugaevsky photo: Perluka Farinn]


Anonymous said...

A database user writes:- There are about twenty games a year that make it to TWIC, none of them played in British events. You don't normally recognise the names of players using it, although Van Wely gave it a go in the recent ETCC.

I could believe that it's regarded as too dangerous to play against opponents who had studied critical positions with the aid of databases and chess engines, but I don't know of an actual refutation. Perhaps there doesn't need to be, if all lines lead to positions which at best are slightly worse for Black, it's not going to be popular. The forum may have more insight. There most recent discussion is here

I don't think anyone has written a book in recent years advocating it. All the recent books would be written from White's view just containing lines trying to kill it.

David R said...

Remarkable coincidence. A number of us were discussing this matter only last week. No one knew why, but everyone agreed that it had 'disappeared'. A quick skim of the databases suggests that something happened in 2001. Thereafter, B96 slides away quickly, and rarely features in games by major GMs. Did the engines do for it? I've no idea frankly; and I don't find anything conclusive in the sample of games I've looked at. Mystery.

abdooss said...

Anybody going to tell me, they missed this one? ;

Alexey Shirov vs Viswanathan Anand
Morelia-Linares 2008 · Sicilian Defense: Najdorf Variation (B96) · 0-1

Michael Yeo said...

Leko's win against Ivanchuk in 2001 was possibly the critical game. Solak -Van Wely from the ETCC does suggest there is no outright refutation, although there are a couple of places Solak might have improved.
I had a nasty shock when someone played it against me in a County match earlier this year - the first time I have faced it in 15 years.
The article reminded me of my first overseas tournament in Bern in 1973. Grant Kerr, a New Zealander who spent a couple of years in the UK in the early seventies,was a Polugayevsky enthusiast. Playing a Swiss junior,he produced an improvement on move 19 (19....Qf6!! if you look it up) on a game that had appeared in Informator 14 and won very quickly thereafter. The Swiss junior was crestfallen. "I don't understand. This is supposed to be a forced draw from Dueball - Kerr from the 1972 Skopje Olympiad." He then looked down at his scoresheet and went "Oh..." as he had only then realised that his opponent was the same player! The joys of pre-database days.

Anonymous said...

"database user" again

The Shirov - Anand game is a relatively normal Najdorf, featuring as it does, 7 .. Nbd7 rather than 7 .. b5. I expect you could reach the same position from 7 .. b5, but only if White declines to enter the complications by 8 e5.

7 .. Nbd7 is more popular, with around 100 examples a year. They classify as B96, B98 or B99 depending on what happens next.

Anonymous said...

From NIC Yearbook 61, 2001 (Tibor Karolyi, Survey SI 7.4):

"I once had a pupil, Adrien Leroy, who went on to train with Polugaevsky. Adrien told me that the great Polu once admitted that his variation had a hole and that the line was incorrect. He told Adrien he would never tell anyone what was wrong with it."

I originally purchased that Yearbook for several other opening surveys, but found that one fascinating, largely because of the above quote.

Anonymous said...

So has this "hole" been discovered since? With computer analysis as it is these days, you would have thought it would have been picked up by now - if it indeed does exist.

It says something, surely, that "The Variation" was never busted when it was a regular in top GM tournaments.

Unknown said...

no idea how i came across this old post but IIRC Nakamura played it against Wang Hao some months back and it didn't work out well for him

ejh said...

2012 but no, it didn't

Ed Townsend said...

Since Nakamura blundered at move 26 in a position that Kevin Go Wei Ming thought favorable for Black (THE Sicilian Najdorf 6Bg5, Everyman 2014)

Anonymous said...

Kasparov in his Najdorf serie, discuss in length the variation; and seemed to suggest (in around year 2005 now I think) that tha variation was still a potent one.