Rybka 3 commercial opening book by J Noomen : a few comments

[Marc Lacrosse]

In Jeroen's presentation of his Rybka 3 opening book at http://www.rybkachess.com/index.php?aus ... bka+3+book we see some very ambitious claims :

For whom is the Rybka 3 opening book interesting?
Although the Rybka 3 book is especially designed for the Rybka 3 engine, it can be used as a theoretical guide by a wide variety of chess players:

Club players
Correspondence chess players
Tournament players
International title holders (FM, IM, GM)
Chess enthousiasts interested in playing engine-engine games
ChessBase users
Fritz, Junior, Hiarcs or other ChessBase engine users, who want to update their opening books
All other chess players interested in the latest theory



So this should be a theoretical guide for... ...almost everybody ?!

Before buying this selfclaimed masterpiece I asked someone who had got his copy to have a look at some precise points who should be of interest for myself.

Here is the first line I checked:

1.e4 c5 2.Nf3 Nc6 3.c3 Nf6 4.e5 Nd5 5.Bc4 Nb6 6.Bb3
[D]r1bqkb1r/pp1ppppp/1nn5/2p1P3/8/1BP2N2/PP1P1PPP/RNBQK2R b KQkq - 0 6

I am told that Jeroen's book recommends 6..d5 with a probability of being played of 100%.

Quite reasonable would you think, knowing that 6..d5 has been played by Kasparov, Topalov, Leko and Anand (not a bad team indeed!).

But there is a problem. Another move seems to be clearly better : 6..c4 (played by Ivanchuk, Kramnik, ... : not a bad team too).

Not only is c4 more active than d5 but :
- it has better statistics in recent GM play (47% vs 43%)
- it is played twice as frequently as 6..d5 in top 2008 GM games (8 vs 4 in my own database)
- it is clearly preferred by Rybka 3 (6..d5 is +0.10 while 6..c4 is +0.00 at depth 20 )

So if you are Black you get a mediocre suggestion with Jeroen's book at move 6.

Should you have played the most promising continuation (6..c4) then after 7.Bc2 Qc7 8.Qe2 g5! you would reach a much interesting and dynamic position of the kind that this book is supposed to go for :

[D]r1b1kb1r/ppqppp1p/1nn5/4P1p1/2p5/2P2N2/PPBPQPPP/RNB1K2R w KQkq g6 0 9

I know this position quite well as I have four correspondence games running in this line in a match against ICCF GM Balabaev (by the way my opponent chose to play 6..c4 - the move that Jeroen does not consider - in all four games!).

The second line I asked to have a look at in Jeroen's book is one of my favorite defences as Black, the Basman-Sale variation that runs 1.e4 c5 2.Nf3 e6 3.d4 cxd 4. Nxd4 Bc5 !?

Clearly this is a more offbeat line so I would agree with light coverage leading to a clearly advantageous position for white..

What is the coverage of this one in Jeroen's book ?

Here Jeroen's only choice is 5.Nb3, a move that leads to quiet positional manoeuvering without much hope for real chances of winning IMHO. This is a bit surprising as there are several possibilities for white to go for more unbalanced positions (the main way being 5.Be3).

After Jeroen's chosen 5.Nb3 Bb6 6. Nc3 Ne7

[D]rnbqk2r/pp1pnppp/1b2p3/8/4P3/1NN5/PPP2PPP/R1BQKB1R w KQkq - 0 7

Jeroen's book continues with 7.Bd3 which is a somewhat timid move.

Theory clearly prefers 7.Bf4 (the threat of occupying the d6 square leaves Black with no other possibility than 7..d5 after which 8.exd5 leaves Black with an isolated d pawn around which the play will be centered afterwards).

After Jeroen's recommended 7.Bd3 0-0 8.0-0 the book coverage is already at its end.

But here Black has the typical 8..f5! counterblow after which Black's chances are fairly good in a very complicated position.

What I can say for sure is that the white player should not be left here without further analysis.

Against a well-prepared opponent this would be extremely dangerous.
(At the opposite, had Jeroen chosen the more common 7.Bf4 line, then it could be reasonable to leave the white player on its own afterwards, be it a human or computer player).

A little comment

In summary line one is an example of slightly outdated theory with suboptimal main line choice.
Example two chooses a sideline within a sideline and leaves the player in a minefield without further analysis instead of going into a quieter well-known favorable line.

OK these are only two lines but they illustrate that this book is far from achieving all his promises and I must say I am really disappointed with the coverage of these lines.

In summary I am ready to accept the claim according which this book is at the edge of latest dicoveries in the most topical mainstream openings (Najdorf, SemiSlav, ...) and that there are even interesting novelties at move 25-30 in some ultrasharp variations.

But at the other end this book is clearly lacking regarding optimised lines for playing against less fashionable but still correct openings.

Who will be satisfied to get a good guide for novelties at move 25 in the Naj while being left with a mediocre recommendation at move six against a line that has been played recently by Shirov, Morozevich, Adams, Nisipeanu, Tiviakov and quite a few other topguns with positive results ?


Here I come back to Jeroen's presentation of his work : it can be used as a theoretical guide by a wide variety of chess players

I do not agree with these advertisement words.

My two little examples clearly show that this kind of book is not the new ECO as he seems to claim.

For what regards myself I won't buy this one.

Marc

[Patatov]

Hey Marc,

I couldn't check both lines but I asked a friend of mine who bought the book from Chessbase and for the first variation, 6...c4 is the main move on Jeroen's book (71.5% with 6...d5 being 64%) and up to 8...g5 we are still on Rybka main's line on this variation.

I'll see if I can get info regarding the second line as well.

[Nimzovik]

Hmmmmmmmm. As said book may be configured to RYBKA's style. Often a move is deemed "better" because it fits with a sense of the authors style when a book is written for humans. However your point is "objectively better " moves. I wonder.....such outstanding names supporting two different moves in same variation is indeed a plus for both moves. Sooooo I am wondering if 'statistics' always generates the objectively correct move? Please do continue this conversation if posssible as I am very interested in this topic and in his book.

[bob]

Hmmmmmmmm. As said book may be configured to RYBKA's style. Often a move is deemed "better" because it fits with a sense of the authors style when a book is written for humans. However your point is "objectively better " moves. I wonder.....such outstanding names supporting two different moves in same variation is indeed a plus for both moves. Sooooo I am wondering if 'statistics' always generates the objectively correct move? Please do continue this conversation if posssible as I am very interested in this topic and in his book.

There have been many cases where move X was played by almost everyone, until someone springs a better move when they play an important game. And suddenly the new move is played by everyone, until yet another novelty is discovered to either refute that move or improve on it further.

This is a book nightmare of the worst kind, as if you make a book for a program and you use some sort of "popularity" to choose among multiple alternatives, you will _never_ choose the most modern moves because of all the "history" behind the other alternatives. And you get whacked over and over as a result.

[Oscar L]

I am not an expert at all on openings theory, so I do not know if you are right or not. But note that any or both of these options can be true too:

1 Perhaps your friend is using chessbase 10 for browsing the book. It has already been posted that the statistical information provided is buggy, look at the screenshot in this thread

http://f23.parsimony.net/forum50826/messages/185684.htm

So can you ask him to check these variations under a fritz gui? Or CB9?


2 The book has already been modified by engine testing

[Oscar L]

Or perhaps the problem is this?

http://rybkaforum.net/cgi-bin/rybkaforu ... 2#pid87617