Friday, October 05, 2007

World Championship Openings

Over the past couple of days (here and then here) we've been looking at the some stats from Mexico 07. Today I want to turn the spotlight on the openings rather than the results of the games played.

Individual World Championships often come to be associated with one, or at most a limited range of openings. Just as Kasparov-Kramnik, London 2000 with the Berlin Defence and San Luis 2005 with the Najdorf, so it was at Mexico City. Just two openings, the Marshall Gambit and the Semi-Slav, were played in 32% of all games. If we throw the net a little wider and include the Petroff (seven) as well as the Catalan and the Queen’s Indian (six each), we've accounted for over two thirds of the entire tournament.

Consulting Chessbase’s download of the games, it can be seen that out of the 56 games played the opening move was,

1. e4 .. 26
1. d4 .. 22
1. c4 ... 4
1. Nf3 .. 4

In fact these figures are slightly misleading given that when White opened with 1. Nf3 it was usually to avoid Svidler’s Grunfeld. When transpositions from the flank openings are taken into account we find:-

king’s pawn openings .... 26
queen’s pawn openings ... 27
others ................... 3


The list of openings and variations breaks down like this:-


KING'S PAWN OPENINGS

Petroff (7)
Nc3 .............. 4
Classical ........ 2
Modern ........... 1


Marshall Gambit (4)
12. Re1, 13. g3 .. 1
12. Re1, 13. d3 .. 1
13. Re2 .......... 1
15. Qe2 .......... 1

Anti-Marshall (5)
8. h3 ............ 3
8. a4 ............ 2


Spanish Other (2)
Exchange ......... 1
5. … Bc5 ......... 1


Scotch (2)
4. ... Bc5 ....... 2

Italian (1)
d3

Sicilian (4)
Najdorf .......... 3
Classical ........ 1


Caro-Kann (1)
4. … Nd7


QUEEN’S PAWN OPENINGS

Semi-Slav (9)
Anti-Moscow ...... 5
Moscow ........... 2
Meran ............ 2

Catalan (6)
6. …dxc4 ......... 5
4. …dxc4 ......... 1


Queen’s Indian (6)
4. ...Ba6 ........ 3
4. ...Bb7 ........ 2
4. a3 ............ 1

Queen’s Gambit Declined (2)
...Bb4 ........... 1
Nbd2 ............. 1

Benoni (2)
g3 ............... 1
e3 ...............
1

Grunfeld (1)
Anti-Grunfeld

Nimzo-Indian (1)
4. Nf3


FLANK OPENINGS

English (3)
...e5 ............ 2
...c5 ............ 1


I'll come back to the Semi-Slav, Marshall, Petroff and Catalan in coming days (if I get around to it). In the meantime it's worth taking note of those openings that stand out through their relative absence.

The Najdorf, after being so dominant a couple of years ago. only appeared here three times - and not at all until Leko-Svidler in the 9th round. Also, not long ago Garry Kasparov described the Queen's Gambit Declined as "the opening for the world championship match" but in this tournament it only got a run out twice - and sidelines at that.

So just a small range of openings on display in Mexico and the field could have easily been even more restricted. Consider Morozevich and Leko using less popular open games to avoid the Marshall and Petroff respectively. It's also the case that both Svidler and Grischuk only abandoned the Marshall for the more doubled edged Najdorf towards the end of the tournament when both of them were desperate points.

I wonder if Mexico 07 will trigger a burst of enthusiasm for the Marshall, Semi-Slav, Petroff, Catalan and Queen's Indian amongst club players or whether the gap between what's popular at super-GM level and that at amateur level (as pointed out by Justin) will continue to grow.

6 comments:

Tom Chivers said...

I predict we'll carry on playing the same old rubbish...

Anonymous said...

All these openings are just incredibly unsatisfying for club players.

Semi-Slav - what's the point in learning enormous amounts of theory, only to discover your opponent playing 3. cxd5.

Marshall - loads of theory and then draw.

Petroff - don't really need to say much.

My impression is that Queens Indian is fairly popular anyway, although I'm not sure what you're qualifying as "club player".

Catalan is obviously rare because it requires Black to play a rare opening as well.

Richard

ejh said...

Semi-Slav - what's the point in learning enormous amounts of theory, only to discover your opponent playing 3. cxd5.

I made a similar point to Angus last night - I suspect that we'll see a growth in the use of the Exchange Slav at club level for much the same reasons that cause people to play 2.c3 against the Sicilian.

It doesn't put me off playing it as Black, and besides, I take the view that if you're going to play a mainstream, sound opening, there's going to be loads of theory anyway.

What it does perhaps do is put me off playing it in correspondence chess. I've just been playing a game in which my opponent deviated from theory at move 23 and it's basically a draw already....

Anonymous said...

Many years ago I noticed Shirov playing into the Semi-Slav via ...d5 and ...c6 when he could just has easily have got there with ...e6 first and only later playing ...c6.

I know that leaves white with other options but it seems that he didn't mind White playing cxd5 on the grounds that he'd then equalised by move 3 and could set about out-playing his opponents in the resulting simplified position at little or no risk to himself.

Anonymous said...

I've heard Shirov called some things, but never a club player! ;)

Richard

Tom Chivers said...

Rowson's "Zugzwang Lite" migth be interesting vis a vis the exhchange slav, like with that Seirawan-Beliavsky game chessbase.com feature a while back.