A Question of Style

As a comment to my entry 'A Grossly Unfair Test', a reader has sent the following question:

"In your book, 'The Ruy Lopez: A Guide for Black', you recommend the move 8...d5 after the moves 1 e4 e5 2 Nf3 Nc6 3 Bb5 a6 4 Ba4 Nf6 5 Qe2 b5 6 Bb3 Be7 7 0-0 0-0 8 c3 (Dia) and you say that 'the more conservative 8...d6 9 Rd1 Na5 10 Bc2 c5 11 d4 Qc7 leads to standard Chigorin positions where White's chances should be slightly preferable.' The 8...d6 line is recommended by Nigel Davies in 'Play 1 e4 e5!' and by Mihail Marin in 'A Spanish Repertoire for Black' which came out after your book. You have 'Play 1 e4 e5!' in your bibliography so you must have thought that White has improvements over Davies' lines. What do you think about Marin's analysis of this line? How does both books analysis compare to Greet's Play the Ruy Lopez?"

Actually the decision to recommend 8...d5 was not mainly a result of any dissatisfaction with 8...d6 or Davies' analysis of the move. The two moves have a roughly equal theoretical status so our choice was more a matter of taste and of finding a move that fitted into our general repertoire. Space considerations also was an issue as the Worrall clearly had to be considered a minor line - even more so before the arrival of Greet's book.

It must be taken into consideration that Davies and Marin both are recommending a Chigorin based repertoire where 8...d6 followed by ...Na5 and ...c5 fits very nicely in - the main strategies are the same whether White's rook is on e1 or on d1. We, however, offer a Zaitsev based repertoire. That doesn't totally exclude Chigorin like lines but it would require extra space for strategical explanations. Another factor is more subjective: we chose to recommend the Zaitsev variation because it involves rapid and natural development. Correspondingly we avoided the Chigorin because we were not really happy with the knight excursion to a5. It is a fact that Black often finds it quite hard to activate this knight in the Chigorin. These considerations apply in the diagram position too.

So, why didn't we recommend a Zaitsev development scheme with ...Bb7 and ...Re8 then? That is indeed a good question and this should have been stated clearly in our book: Against the Worrall attack, 8...d6 9.Rd1, 9...Bb7 doesn't seem to be working very well, as 10.d4 creates threats to e5, thanks to the pin in the d-file.

Fortunately this isn't a great problem as 8...d5 is an active and strong move which fits well with the general philosophy behind the Zaitsev. It takes a bit more theoretical preparation than 8...d6 but once Black masters a few sharp lines he can expect quite a pleasant life against the Worrall. This claim has to be backed up by analysis (and in our book we supply some). But to some extent it can also be supported by visual evidence.












If you compare these two diagrams which shows the positions after 11 moves in the two mainlines there can be no doubt that Black appears more active in the second:

• In the first diagram (arising from 8...d6 9.Rd1 Na5 10.Bc2 c5 11.d4 Qc7) White has achieved an central advantage (e4&d4 vs. e5&d6) while Black's knight on a5 appears somewhat misplaced.
  
• In contrast White in the second diagram (arising from 8... d5 9. d3 Bb7 10. Nbd2 Re8 11. a3 Bf8) has spent a move on the modest a3 while Black is almost fully developed (Zaitsev style!) and has even taken the active stance in the centre (e5&d5 vs. e4&d3).

Obviously this doesn't prove that 8...d5 is better than 8...d6 but I think it shows that 8...d5 is a more ambitious approach (and consequently better if it actually works).

A Grossly Unfair Test

A couple of reviewers have compared our "The Ruy Lopez: A Guide for Black" (RLGB) with Greet's "Play the Ruy Lopez" (PRL) by checking the line where our black repertoire meets Greet's white repertoire. This of course is a test we are doomed to lose, or in Watson's words: "Taking this as a 'test' would be grossly unfair to Johnsen & Johannessen, who must spend the bulk of their efforts to justifying Black's position against the many mainstream attacks against the Ruy Lopez; hence they are hardly about to invest a lot of time and space into addressing the Worrall System."

But as two reviewers have already had problems getting this comparison right, let's go through the exercise anyway. It's the PRL-guy with the white pieces against the RLGB-guy, and there is no doubt about the first moves:

1.e4 e5 2.Nf3 Nc6 3.Bb5 a6 4.Ba4 Nf6 5.Qe2 b5 6.Bb3 Be7 7.0–0 d5 8.d3 0–0 9.c3 Bb7 10.Nbd2 Re8 11.a3 Bf8
(Dia.)
12.exd5!?
Here RLGB runs out of theory, as only 12.Ba2 and 12.Rd1 are given. Giving up White’s central foot-hold doesn’t seem very desirable and there is only one game in MegaBase 2006 with this move: Summerscale-Mannion, Dublin 1993. But let’s assume that the game doesn’t end there, and Black plays the natural recapture:
12...Nxd5
The PRL-guy is still not out of book, and flashes out:
13.Ne4
This activates White's queen-side, but isn’t really forcing. Black seems to have several playable alternatives. However, one move that completes development stands out as particularly natural:
13...Qd7!
This is the main-line in PRL, so White’s reply again comes instantly:
14.Re1
(Dia.)
With some luck ‘our’ man will now discover that he’s back in book. This position can be found in RLGB under the move-order 10.Re1 Re8 11.Nbd2 Bf8 12.a3 Qd7 13.exd5 Nxd5 14.Ne4. That is not a trivial transposition but the RLGB-guy’s odds are considerably improved by the fact that there’s a diagram with the exact position at page 160.

Now Black has three alternatives that are listed in the same order in both books:
a) 14...Na5?!, which both books give as leading to an advantage for White.

b) 14...Rad8, which in RLGB gets 10 lines of analysis and seems to be fine for Black. This move is however shown to be dubious by some quite impressive analysis by Greet where he improves on Short-Almasi, Wijk aan Zee 1995.

c) After 14...f5!?, both books offers the drawing line 15.Ba2 Kh8 16.Nfg5 h6 17.Qh5 fxe4 18.dxe4 Nf6 19.Nf7+ Kh7 20.Ng5+ 1/2–1/2 of Tiviakov-Grischuk, Linares 1999, and RLGB stops there. However, this obviously isn’t theoretically satisfactory for White (although it may have great practical value), so PRL offers two pages of analysis on 15.Ng3!?. The move in itself isn’t very impressive, but Greet’s analysis beginning with 15...g6 16.Bg5 quickly gets sharp, and it cannot be denied that the RLGB-guy will have a tough task for the next dozen moves if he stumbles into Greet’s mainline. Whether White objectively has any advantage is another question. Let me suggest the natural 16...h6 17.Bf6 Bg7!?, which Greet doesn’t mention. Whether this will be sufficient for equality I cannot really tell. Most analysis engines indicate that White has a very small plus but for a further evaluation I would have to consult my GM co-author.

Conclusion: A RLGB reader will have more of a challenge than a PRL reader reaching the position after 14.Re1. If he does reach it, he could get into trouble if he choses option b), the natural looking 14...Rad8. If he instead plays option c) the risky-looking 14...f5, he may have to accept a quick draw or face the unknown, but not too dangerous looking move 15.Ng3.

Watson on the Ruy Lopez Guide for Black

I am embarassed that I have not been able to keep up my updates for some time. I have simply been too busy on too many fronts. But now it's starting to clear up, so maybe I will be able to fill an almost normal quota for April anyway.

I was pleased to see a review by Watson at the Week in Chess news site. It was fairly short (JW: "Unfortunately, this is one of the titles that I'm going to simply recommend without giving it the close consideration it deserves") but in general very positive. Watson, like Silman, draws attention to the alternatives we offer to the Zaitsev main line and warns amateur players against the dangers and efforts associated with the Zaitsev. It's becoming obvious that I will have to comment upon this in an entry quite soon.

I was more surprised to note that not even Watson manages to get the comparision with the Greet book on the Worrall quite right, as he too seems to miss that our lines merge quite quickly after the first departure. This actually is a quite straightforward subject, and another good theme for a blog entry.

Play the Ruy Lopez

When I some 10 days ago received Andrew Greet’s “Play the Ruy Lopez” - an impressive tome of 376 pages - my instinctive reaction was to check how his analysis of the Worrall compared to our coverage in “Ruy Lopez: A Guide for Black”. It came as no shock that his 131 pages (!) offered quite a lot more than our 8.5 pages did. The Worrall is after all a rather minor option for White. (But our 4 pages on 7.d3 and 7.a4 may be of some interest even to Greet’s readers wishing to expand their Worrall repertoire a bit). Whether he really has made 5.Qe2 a promising line for White is another question, to which I may return later if I reach any conclusion.

The next thing I did was to check what Greet had to offer on the Norwegian variation. As a good Norwegian I have always wanted to play this risky line, but I have never dared to. Instead I have from time to time checked its theoretical status and recent games by Norwegian players. In order to play it successfully you need good defensive technique, good nerves and a deep understanding of chess - three qualities I have never claimed to possess.
The variation’s theoretical standing has always been shaky, and without offering a lot of new analysis, I believe Greet’s book has made Black’s task even harder; mainly by pointing out White’s most promising course but also by offering some small improvements for White where needed. Below is my summary of one of the mainlines with some additions from Greet's book :

1.e4 e5 2.Nf3 Nc6 3.Bb5 a6 4.Ba4 b5 5.Bb3 Na5
This is the Norwegian Variation. In Greet's words it is "...arguably Black's most direct method of fighting against the Ruy Lopez".
6.0–0 d6 7.d4 Nxb3 8.axb3 f6 9.Nc3 Bb7 10.Nh4 Qd7
Up to here, Greet deals systematically with all of Black's possible deviations. But every Norwegian knows that this is the first real junction for Black (with a possible exception for Zwaig's 7...f6). Now, however, 10...Ne7 is a major alternative which deserves a separate entry.
11.Nd5!
(Dia)
Greet recommends this over 11.f4, which is also quite challenging.
11...g6!?
This is not a perfect solution but the alternatives are even less tempting:
a) 11...Ne7?! 12.Qh5+ Kd8 13.c4 when Black’s king soon felt very vulnerable in the centre in Short-Sulskis, Bled 2002.
b) 11...0–0–0 12.c4 Ne7 13.Be3 Nxd5 14.cxd5 Kb8 15.Qe2 Re8 16.dxe5 dxe5 17.Rfc1 when Black’s king again was under heavy fire, this time on the queenside, in Leventic-Krstic, Zadar 2004.
c) 11...Qf7 has been considered the main continuation but after 12.c4 c6 there are problems ahead:
c1) 13.Ne3?! Ne7 was not too bad for Black in Anand-Timman, Linares 1993.
c2) Anand later suggested 13.Nc3! as an improvement.
c3) Greet suggests 13.Nb6! which he modestly attributes to his computers. My Fritz 9.0 agrees that White has a very clear advantage after e.g. 13...Rb8 14.d5. It is not inconceivable that the knight may turn out to be trapped, or at least misplaced at b6, but I honestly cannot see how. I would not feel comfortable on the dark side here.
12.c4 Bg7

(Dia)
In Gutsche-Boog, corr 2000, 12...Rd8 was made to look quite silly after 13.Bd2 exd4 14.Ba5. 13.dxe5
Here Greet concludes with Anand's recommended 13.f4, presumably agreeing that White is clearly better. I happen to know that some Norwegian players disagree with that evaluation. I will not try to reproduce the analysis I saw a couple of months ago, but I can promise that things are not at all clear. However, judging from available games, this central exchange may be more critical.
13...fxe5
Could 13...dxe5!? be the right move?
14.Bg5
(Dia)
This could be the critical position. Black has a difficult task ahead, e.g.:
14...Rb8
14...h6? 15.Nxg6 hxg5 16.Nxh8 Bxh8 17.Qh5+ obviously is not an option.
15.Qd3 bxc4 16.Qxc4 c6!?
Giving up the light-squared bishop with 16...Bxd5 17.exd5 of Raidna-Boog, corr 2000 is a desperate measure.
17.Nb6!? Qf7 18.Rfd1 Bf6
18...h6? seems to lose after the slightly surprising 19.Qb4!
a) 19...hxg5? is out of the question because of 20.Qxd6.
b) 19...Qc7 20.Nc4 c5 21.Qa4+ and Black’s fortress is cracking again.
19.Nf3 Qe7 20.Be3
This position is not necessarily lost for Black, but it cannot be what Black hoped for when entering the Norwegian variation.