I’ve been looking at my library, and it got me to wondering, with all of these opening books entitled How to win with White, How to win with Black, Crushing the Caro-Kann, Winning with the Caro-Kann, etc., how much should we rely on these books, esp. the ones that are written like Cliff-Notes?
I’ve noticed that many omit serious lines.
For instance, looking at Surprise your opponents with the Stonewall!, from Valeri Lilov. When I put some of those lines in an engine, it seems that White is doing mighty fine in lines that Lilov considers winning for Black. Just tonight, I played a training game with a friend who played a line that wasn’t included in Lilov’s analysis, nor in the engine, yet White still held over 1/2-pawn advantage in some lines, and a full pawn in others.
I guess it boils down to how much can we rely on, or trust the analysis from GMs, IM, etc.? It seems like they intentionally omit possible lines that refute their ideas.
Just to make sure that I following you correctly, are you referring to the software thingy, The Dutch Stonewall by Valeri Lilov? uscfsales.com/fritz-trainer- … lilov.html
I once saw a 2007 John Nunn comment that some authors irresponsibly leave out troublesome possibilities, but, if I remember correctly, he was writing about authors of books that recommend very unusual lines. I do not know if the Stonewall Dutch would count as such an opening or if Lilov would count as such an author, but, in general, with regard to lines that are left out by an author, it seems to me that these sorts of factors might be involved.
(1) If one is reading about a black reaction to a standard white opening (such as 1 d4), it is not likely that there is prescription for equality.
(2) An author can be compelled by space considerations to be selective in what is covered. I think this is more true for software than it is for books.
If the author is making reasonable choices as to what to select, I think that he could be doing a real service for a student, who, after all, is not going to remember everything anyway and might benefit from help with an identification of the most important stuff to look at.
I know that I am always pleasantly surprised when someone plays the Stonewall.
There are so many books on the Stonewall, besides Lilov’s. The book, “Dutch Stonewall,” by Jacob Aagaard has one of the longest introductory chapters, 49(!) pages, of any opening book. The book itself is only 160 pages, including the index, table of contents, and cover page.
I am not saying the Stonewall Dutch is a bust. At one time I proofread a book on this opening and made some suggestions and improvements for the Black side. On a practical level, I have always been happy to play against such a structure as I like openings where color complexes play a role. Several Petrosian games show methods of how to break down the Stonewall. There are also a couple of model games by S. Tartakower on how to play the opening dynamically for Black. I don’t know what my exact scoring percentage has been, but I expect it is fairly high against this opening ever since reading Euwe’s books two volume set in hardback on the middlegame several decades ago. Kramnik seems to think well of it, however. There is a section in “Positional Chess” by Mark Dvoretsky on this opening with analysis by Vladimir Kramnik. Note though that there are no example games in the article against Karpov or Petrosian in this opening.
I don’t have enough knowledge of playing as Black, (although, as Black, I do welcome it when White plays it) to make an opinion on it, as I’ve just started looking at it as a defense vs Q-pawn openings.
But the main jest of my question is that even with the videos, of which Louis mentioned, a lot of lines are simply not mentioned.
Another is the Smith-Morra Gambit, where Black plays the Sicilian Kan set-up, and I suppose it can get into what is called the Chicago Defense, with the Rook-lift to d7, but very little is written in the specialty books about this plan.
In 2007, a GM John Nunn book (Secrets of Practical Chess) had some possibly relevant comments.
So, it seems that one can encounter a “less honest author”, but it also seems to me that your software does not fit the profile.
It doesn’t sound as though the Dutch Stonewall would be the sort of opening that Nunn would put in the same category as, for example, the Latvian Gambit. At uscfsales.com/, I see more than twenty Lilov products, so he seems to have a trustworthy reputation. On the other hand,
My guess would be that the primary reason for omissions was an intention for the software to focus on “main lines … the general structure of the opening … illustrative games … general plans and ideas”, leaving the unusual lines to be filled in later. For that unusual stuff, you might want to try Win with the Stonewall Dutch by Sverre Johnsen and Ivar Bern.
Somewhere or other, I once saw a site that suggested the “labor saving opening repertoire … Colle-Zukertort, 150 and Barry Attacks as White … Rubinstein French and Stonewall Dutch as Black” as presented in the four books, Zuke 'Em-The Colle Zukertort Revolutionized, Starting Out: d-Pawn Attacks, How to Play Against 1 e4, and Win With the Stonewall Dutch. The writer added,
Perhaps I should also mention some comments in connection with computers:
I might take some of the commentary on chess engines from 2007 or earlier with a grain of salt. That was several generations of Fritz ago, a virtual eon in computer history. That review by John Nunn was done way before the emergence of Houdini, Stockfish, and Komodo, too. 2007 compared to today is like looking at the early dinosaur period. Deep Blue is no more; the program that beat Kasparov could probably not hold a candle to the newest programs and hardware. Can they play “quiet” positions? Yes. Have their openings improved? Absolutely. Can they access endgame table bases? Yes they can. They are not perfect, but they are frustrating to play on their highest levels, with their deep search algorithms and seeming ability to learn from mistakes. If these programs were not good, there would not be so much effort to stop computer cheating in tournament play.
That said, I really don’t like to use the programs that much. I am trying to train my brain to think, not just passively look at the little monster doing cartwheels over the game we are playing. Great for double checking, or looking up games, or testing an idea. Bad for relying on its judgment, substituting the monster for your own efforts. Plus when they get bad positions, they sometimes freeze and won’t move, the computer version of throwing a tantrum. An older version of Fritz trash talks you, which I think is funny. The voice sounds like Robin Williams. When it is losing it offers a draw. No, I am too “old school” and in love with my old friends, books with model games, interesting puzzles, and timeless advice. I also don’t want to be caught in the net, like the buyers of iphones and android devices, of buying the newest iteration of the programs as that gets pretty expensive.
I should probably have made it clear that the John Nunn comment was a book quote (Secrets of Practical Chess) rather than a review. I believe that you are correct about Houdini, Stockfish, and Komodo not being considered by Nunn in 2007, but, on the other hand, he was about a decade past the Deep Thought days and had had a lot of personal experience with subsequent programs. He also was aware of endgame table bases. In 2007, JN certainly was not trying to tell anyone that programs were not good.
It seemed to me that, in 2007, JN was trying to advise some caution about excepting computer judgments of specific positions. I have not seen any data on the degree to which such caution is still appropriate, but my guess would be in agreement with your “not perfect” comment. newinchess.com/Shop/Images/Pdfs/955.pdf
So, are you saying that the Stonewall is a bust?