Friday, January 30, 2009

A Scandinavian Speciality in the Closed Ruy Lopez

I still have an unfinished manuscript on the 9...a5 variation of the Closed Ruy Lopez. It was intended to be a low-theory alternative to the Zaitsev in 'The Ruy Lopez: A Guide for Black'. But even before my GM co-author had had a look at it, we decided that 9...Qd7 was a better companion move - mainly because of the many possible transpositions. However, I recently had a new look at the line because of this game:

Ivanchuk - Carlsen
Corus (Wijk aan Zee) (5) 2009
1.e4 e5 2.Nf3 Nc6 3.Bb5 a6 4.Ba4 Nf6 5.0–0 Be7 6.Re1 b5 7.Bb3 0–0 8.c3 d6 9.h3 a5 (D)
This somewhat exotic move seems to first have been played by the Swede Gösta Stoltz and has later been championed by the Dane Lars Bo Hansen and the Norwegian Simen Agdestein, so it doesn't seem unreasonable to name it 'The Scandinavian variation'.
If I ever complete the manuscript I will consider including a chapter on the even rarer 9...Bd7!? which can be treated as a 'sister variation' with the connecting line being 10.d4 a5!?.
10.a4 is White's other major option.
The oldest game in my database went 10...exd4 11.cxd4 a4 12.Bc2 Nb4 13.Nc3 Nxc2 14.Qxc2 c6 15.d5 cxd5 16.exd5 b4 17.Nxa4 Bd7 18.b3 Nxd5 =+ Thomas-Stoltz, Warsaw 1935.
11.Bc2 Bd7 12.Na3
This seems logical and after getting Ivanchuk's approval I assume it is the new main line. 12.Nbd2 has been played more frequently.
12...Rb8 (D)

This looks more natural than 12...Qb8 which was Agdestein's preference: 13.Bd3 exd4 14.cxd4 Nb4 15.Bb1 Qb7 16.Bg5 Rad8 17.Nc2 Na6 18.Ne3 Rfe8 19.Qd3 g6 20.a3 c5 21.e5 dxe5 22.dxe5 Bc6 23.Qc3 Nd5 = Renet-Agdestein, Lyon 1988. However, it must be said that the rook returns quickly to a8 so who knows?
This must be more critical than 13.Bd3 when Yagupov-I.Zaitsev, Moscow 2000 continued 13...b4 14.Nc4 bxc3 15.dxe5 Nxe5 16.Nfxe5 dxe5 17.bxc3 Bd6 18.Bc2 Bc6 19.Bg5 a3 20.Qf3 h6 21.Bc1 Qe7 22.Ne3 Bd7 23.Bb3 Kh8 1/2–1/2.
One of Black's big challenges in most Closed Ruy Lopez systems is to activate his queenside knight. The alternative obviously was 13...Na5 which also seems satisfactory. One reasonable line suggested by Rybka goes 14.Qe2 Qc8 15.b4 axb3 16.axb3 c5 17.dxc6 Qxc6 18.b4 Nc4 19.Nxc4 Qxc4 20.Qxc4 bxc4 21.Be3 Ra8 22.Nd2 +=.
14.c4 Ra8 15.Be3 b4 16.Nb1 c5 17.a3 b3 18.Bd3 (D)

Typically Carlsen grabs the first opportunity to active play.
19.Bxe4 f5 20.Nfd2
Rybka initially prefers 20.Bd3 e4 21.Nc3 but after the further moves 21...Bf6 22.Rc1 Nc8 23.Bf4 exd3 24.Qxd3 Re8 25.Rxe8+ Qxe8 26.Re1 Qh5 27.Re3 h6 chances seems balanced.
20...fxe4 21.Qh5 Be8 22.Qe2 Bd7 23.Qh5 Be8 24.Qe2 Bd7 ½–½
If it wasn't for the repetition, the position still would have been fairly equal.

Thursday, January 29, 2009

Dangerous Weapons - Dutch

For the moment I find the Dutch defence one of the most interesting opening subjects. Nonetheless I was not exactly ecstatic to see that Everyman is announcing a new book in their series Dangerous Weapons: The Dutch.

The three authors are all very competent so of course my slight feeling of disappointment has to do with timing. I would much have preferred to have the book available before completing "Win with the Stonewall Dutch". The book is being set and is announced for March so there is not a lot to be done anyway. But there might just have been room for some small but valuable updates.

One advantage of writing on the Dutch is that a complete repertoire can still be fitted within a reasonably sized book. But the anti-Dutch lines are constantly expanding and in the not too distant future they will demand a separate volume.

I hope the book will be balanced in the way that it doesn't concentrate too much on White's minor systems. There are a lot of minor systems for Black that would fit perfectly in the Dangerous Weapon series.

Wednesday, January 28, 2009

He That Seeketh Findeth

He who seeks shall find ... but not always what he seeks for. When looking for games in the 6...a5 line that I wrote about yesterday, I came across this game. It has little relevance to the theme but still is food for thought:

Dautov - Maier, Garmisch Partenkirchen 1991
1.d4 f5 2.Nf3 Nf6 3.g3 e6 4.Bg2 Be7 5.0–0 0–0 6.c4 a5!?
This may look strange if you don't know that 6...d6 7.Nc3 a5 and 6...d6 7.b4 are two quite critical lines in the Classical Dutch. See yesterday's entry for some more thoughts.
If there is a way to take advantage of Black's rare 6th move, it most likely is 7.d5.
Black presumably is happy to have avoided the 6....d6 7.b4 variation and enters a position more frequently reached by the move-order 6...d6 7.Nc3 a5. Normally I would have gone on to the next game in my database but I noticed that the unrated player held a draw with Black against a strong GM so I had a closer look:
8.Re1 Ra6!? (D)

An amazing move that I would never have considered in a serious game. I have seen ...Ra6 played at a later stage in this kind of positions - after Black has played ...e5 and accepted an isolated e-pawn. But then there has always been a clear way to the kingside.

9.Qc2 Kh8
I am not quite sure whether Black is preparing ...e5 with this move or if he is mainly waiting for White to play e4.
10.e4 fxe4 11.Nxe4 Nxe4 12.Qxe4 e5!?

In his book Play the Classical Dutch, Simon Williams states "If Black can achieve the ...e5 advance he will generally be at least equal". This advise is not at all hard to remember. The tricky part is to decide when Black can do this and when he cannot. It may be a question of calculation but also of evaluation. Here ...e5 certainly is a legal move. But after seeing the game, I am still not quite certain whether Black here can play ...e5 or not.
13.dxe5 dxe5
The Isolated King's Pawn (IKP) occurs far less frequently than its cousin, the Isolated Queen's Pawn (IQP). Normally Black is doing fine with the IKP in the Dutch as he is well placed to attack White's kingside fianchetto. But isn't the pawn just for taking?

One of two critical lines. The other starts 14.Qxe5 and one possible continuation is 14...Re6 15.Qxa5 Nc6 16.Qd5 Rd6 and Black has at least some practical compensation. However, in a practical game one may wonder whether a developing move like 14...Be3 hadn't been better.
14...Bb4 15.Rf1 Re6 16.Be3 Qe7 17.f4 Nd7 18.Bd4 Re8
White is pretty much tied up to the defence of his central knight.
19.Bh3 Nxe5 20.fxe5 Rd6 21.Bxc8

Isn't Black clearly better after 21...Rxd4 22.Qxb7 Qxe5?

I am not sure how well Black played in this game. But I think that for the next few months I will be looking a little harder for ...Ra6 ideas in the Dutch.

Tuesday, January 27, 2009

Delaying the Decisions

Sometimes you have to wonder why certain opening lines are well explored while others are ignored. Usually there is a good reason but it's not always easy to find and sometimes fashion seems the only explanation.

Generally the critical cross-road in the Classical Dutch is considered to arise after these moves: 1.d4 f5 2.Nf3 Nf6 3.g3 e6 4.Bg2 Be7 (the Modern Dutch Stonewall is characterized by the move 4...d5 followed by ...Bd6; the advantage is a more active employment of his dark-squared bishop and the disadvantage is that Black s reveals his central formation earlier) 5.0–0 0–0 6.c4 d6 7.Nc3 (Dia)

In this position Black's main moves are 7...Ne4, 7...Qe8, 7...a5 (all examined in Simon Williams' Play the Dutch played by strong players and in general quite well researched) and 7...c6 (which has been played by Saemisch among others but seems unfashionable). None of these moves are developing moves in a strict sense but they all to some extent prepare queenside development, prepare ...e5 or prevents e4.

What's interesting is that if we retrace two half moves, to this position (Dia):

the picture is entirely different. Black's main options are 6...d5 - Botwinnik's Stonewall which sets up a relatively static centre and 6...d6 - the Classical Dutch (aka. the Ilyin Zhenevsky variation) which keeps the central pawns fluid (but generally prepares ...e5).

However there is also another relatively popular option -
Alekhine's idea 6...Ne4!? which delays this decision. This system too is examined in Williams' book. Black's main idea is to meet some of White's developing schemes with ...d5 and others with ...d6. Another benefit is the fact that Black doesn't allow b4 (6...d6 7.b4!? is quite an important little line, scoring considerably better than 7.Nc3).

Another respected 6th move with similar motives is
Bellin's 6...c6!? which has slowly increased in popularity for at least 15 years. This line has been played by Short, Larsen, Smyslov and Bronstein among others and is discussed in Giddins' excellent 'How to Build Your Chess Opening Repertoire'.

These by no means are easy lines to play for
Black. Not only does he delay development but he also needs a good general understanding of the Dutch. In order to play the position well, you not only must know when to go for a Stonewall set-up with ...d5 and when to go for a fluid centre with ...d6. You also must be able to handle both kinds of position well!

The obvious question is: What about
6...Qe8!? (Dia) and 6...a5!? (Dia).

None of the moves have been tested in modern grandmaster clashes. Actually there are relatively few practical examples altogether - I find only 99 games with 6...Qe8 and 30 games with 6...a5 in MegaBase 2009. Obviously both moves can transpose to standard lines after 7.Nc3 d6 so the two principal questions are now:
  1. Can and should White avoid 7.Nc3 d6 which would transpose to the 6...d6 7.Nc3 a5/7...Qe8 lines?
  2. After 7.Nc3, can Black give the game an independent twist with an alternative to 7...d6 (or are these moves just move-order tricks avoiding the 6...d6 7.b4 option?)?
I don't have the answers yet but I will keep you informed about my investigations.