Pessoa wrote on 12/23/19 at 09:54:17:

Coming back to my first posting ...

The blurb on the back of the book says: "Using carefully selected examples, the authors want to make you familiar with the strategic ideas behind the famous Maroczy bind."

I would say the authors succeed in doing so. Plans for White and Black are explained clearly, using much prose. I learned quite a bit from just reading the text and looking at the diagrams. (So yes, on that the book "delivered".)

However, as soon as I got into the details, i.e. to the games and variations, this first, positive impression turned upside down.

The book contains 127 complete games, annotated and analysed in more or less detail. I looked up these games in MegaBase 2019, started to play them over and compared the authors' annotations with what Stockfish 10 has to say …

Having looked more carefully at the first 20 games, I come to the conclusion that you cannot trust a single line given by the authors (M&M) in their annotations to the games. I also learned that you even cannot trust the game notations they give!

Here goes:

L. Portisch – H. Pfleger (Manila 1974), p. 35 ff.
1.Nf3 Nf6 2.c4 c5 3.Nc3 Nc6 4.d4 cxd4 5.Nxd4 g6 6.e4 Bg7 7.Be3 Ng4 8.Qxg4 Nxd4 9.Qd1 Ne6 10.Rc1 d6 11.b4 0–0 12.Be2 a5 13.a3 axb4 14.axb4 Bd7 15.0–0 Bc6 16.Qd2 Ra3.

Now M&M comment: "It is not possible to capture the pawn because of the forced line 16...Bxc3?! 17.Rxc3 Bxe4 18.Bh6 Re8 19.Bg4 f5 20.Re1! and there is no defence against the move Rxe4."

This suggests that White is winning, but Stockfish 10 comes up with the continuation 20...fxg4 21.Rxe4 Ra1+ 22.Rc1 Rxc1+ 23.Qxc1, with just a slight advantage to White. By contrast, in the above line 19.Re3! (instead of 19.Bg4) wins immediately.

But the story does not end here. The game is annotated by GM Ljubojevic in Chess Informant (volume 18, game 354). After 16…Bxc3 he gives 17.Qxc3 (not 17.Rxc3) Bxe4 18.Bh6 Re8, and now 19.Rce1 with the idea Bg4. But here the immediate 19.Bg4! is much better indeed, and Stockfish 10 comes up with 19…f5 20.Rfe1 with the idea Rxe4, winning. This is "almost" the line given by M&M in their book, where it is messed up, with 17.Rxc3 given instead of 17.Qxc3.


L. Portisch – L. Ljubojevic (Moscow 1977), p. 37 ff.
1.Nf3 c5 2.c4 g6 3.e4 Bg7 4.d4 cxd4 5.Nxd4 Nf6 6.Nc3 d6 7.Be2 0–0 8.0–0 Nbd7 9.Be3 Nc5 10.f3 Bd7 11.b4 Ne6 12.Qd2 a5 13.a3 Nh5 14.Rfd1 Nhf4

M&M: "Yet another typical method – Black makes tactical use of the undefended knight on d4: 15.Bxf4? Bxd4+!"

A very short note, just two half-moves, but even this the authors get wrong, because both 15...Nxd4 or 15...Nxf4 are better than 15…Bxd4, according to the engine. A "patzer with a Pentium comment" perhaps, and maybe just a minor detail, but still annoying enough.


L. Van Wely – D. Reinderman (Wijk aan Zee 1999), p. 40 ff.   (This game is annotated by GM Alterman in MegeBase.)
1.d4 Nf6 2.c4 g6 3.Nc3 Bg7 4.e4 d6 5.Nf3 0–0 6.Be2 c5 7.0–0 cxd4 8.Nxd4 Nc6 9.Be3 Bd7 10.Qd2 Nxd4 11.Bxd4 Bc6 12.f3 a5 13.Rab1 Nd7 14.Be3 Nc5 15.Rfc1 Qb6 16.b3 Rfc8 17.Bd1

Here M&M quote Alterman, without checking his line with a modern engine: 17.Qc2 Qd8 (instead, the not mentioned 17...Bh6! leads to a black advantage, Ambrozova-Motuz, Vsetin 2015) 18.Bf1 etc., Fogarasi-Martin, Festuge 1991.

17...Qd8 18.a3 b6 19.b4 axb4 20.axb4 Nd7

M&M: "Definitely not 20...Na4? 21.Bxa4! Bxa4 22.b5 and White is close to winning."

This is not given by Alterman but is original (?) analysis by M&M. However, it appears to be completely wrong, as after 22.b5 Black still has 22…Rxc4 with unclear play. Instead, the much stronger 22.Nd5! almost wins.

21.Be2

M&M: "21.Nd5 e6 only favours Black, e.g. 22.Nf4 Qf8 23.Ne2 Be5 with play on the dark squares."

But according to the engine, White was much better after the subsequent 24.Nd4 in Farkas-Paschall, Szeged 1998, a game quoted by Alterman.

21...Ra3! 22.Bd4 Bxd4+ 23.Qxd4 Rca8 24.f4 Qf8 25.Nd5 e6 26.Nxb6

M&M: "If White retreated with 26.Nc3, Black would start realizing his dark-squared strategy – 26…e5 27.Qd2 exf4 28.Qxf4 Ne5=."

This is simply copied from Alterman, but it has not been checked: instead of 28…Ne5, the engine gives 28...Qe7! as clearly better for Black.


S. Furman – E. Gufeld (Vilnius 1972), p. 46 ff.
1.Nf3 c5 2.c4 g6 3.e4 Nc6 4.d4 cxd4 5.Nxd4 Nf6 6.Nc3 d6 7.f3 Nxd4 8.Qxd4 Bg7 9.Be3 0–0 10.Qd2 Qa5 11.a3 Be6 12.b4 Qc7 13.Nb5 Qd7

M&M: "Mikhail Tal handled this position more carefully: 13...Qd8 14.Nd4 Bd7 (Savon-Tal, Sukhumi 1972)."

But in that game (Chess Informant volume 14, game 363; MegaBase 2019) Tal retreated his queen already one move earlier: 12...Qd8 13.Rc1 Rc8 14.Nb5 a6 15.Nd4 Bd7 …


A. Beliavsky – M. Pavlovic (Linares 2003), p. 47 ff.
1.d4 Nf6 2.c4 g6 3.Nc3 Bg7 4.e4 d6 5.Nf3 0–0 6.Be2 c5 7.0–0 cxd4 8.Nxd4 Nc6 9.Be3 Nxd4 10.Bxd4 Bd7 11.f3 a5 12.a3.

At this point M&M mention the game Schroeder-Shvayger, Gibralter 2017, which continued with 12.c5 instead of 12.a3. In the note M&M give this game completely and analyse it very lightly. The strange thing is: the same game is also given as one of the main games in the book on page 221 ff., but there the analysis deviates from the analysis on page 48. (By the way, in MegaBase you will find this game under Schroeder-Naiditsch, as the black player was Yuliya Naiditsch, born Shvayger.)

12...a4 13.c5 dxc5   

At this point M&M quote the game Lenic-Mohr, Ljubljana 2008, continuing with 13…Qa5. The game is followed (and lightly analysed) until the moves 26.Qf4 Be5 27.Bxe5 Nxe5, when M&M comment "with unclear play". On page 221 the same game is quoted in a note to the game Schroeder-Shvayger (mentioned above), but this time with no analysis at all and with the notation ending after 26.Qf4 Be5, when M&M comment "with equality" …


A. Suetin – L. Shamkovich (Leningrad 1967), p. 52 ff.

According to M&M, this game was played in 1961. However, according to MegaBase 2019 it was played in 1967. The game is also given in volume 3 of Chess Informant (game 64), which contains only games played in the first half of 1967. – Again perhaps just a minor detail, but symptomatic for the "accuracy" with which the games are treated by the authors in this book.

1.c4 c5 2.Nf3 Nf6 3.Nc3 Nc6 4.g3 g6 5.Bg2 Bg7 6.0–0 0–0 7.d4 cxd4 8.Nxd4 Nxd4 9.Qxd4 d6 10.Qd3 Bf5 11.e4 Be6 12.b3 Nd7 13.Be3 Qa5 14.Rac1 Rac8 15.Qd2 Rfd8 16.f4 Nf6 17.h3 Bd7 18.g4 Bc6 19.f5 Nd7 20.g5 b6 21.h4 Qe5 22.Qf2 Rf8 23.f6 exf6 24.Bd4 Qe6 25.gxf6 Bh6 26.Rcd1

According to Chess Informant and MegaBase 2019, the game now continued 26…Nc5 27.Qg3 Rce8 28.Rfe1 Qc8 29.Nd5 Bxd5 30.cxd5 …

Instead of this, M&M give the (apparently wrong) move order 26...Ne5 (?) 27.Qg3 Rce8 28.Rfe1 (here 28.Bh3! would win immediately) 28...Qc8 (and here 28...Qxf6 would have been possible, with advantage to Black) 29.Nd5 Bxd5 30.exd5 (?). This position even gets a diagram in the book …

I have no idea how such things can be possible in our time and day.


Thus, so far we have seen:

Old analysis, quoted wrongly.
Old analysis, copied and pasted without checking it with a modern engine.
Wrong analysis by the authors.
Wrong game quotations.
Wrong game notations.
Inconsistency between analyses of the same game in two different parts of the book.

The conclusion:  One cannot trust the notation of the games, and one cannot trust a single line of the analysis given by the authors.

Quite disappointing. And a wasted chance in my view.

But the prose is fine. (However, one has to keep in mind that the prose ultimately depends on the analysis of the games …)

Pessoa wrote on 12/23/19 at 09:54:17:


Mikhalchishin doesn't seem to use engines very much when writing. His rook and pawn endings books from Chess Evolution have serious errors, as did his Mastering Complex Endgames from Thinkers.

More evidence is his claim that "real trainers" don't use tablebases. Check the replys to Nielsen's comment here:

https://www.facebook.com/jacob.aagaard.75/posts/1533263773399072

I have the book too and have read it with mixed feelings.

The strategical explanations are certainly quite useful. However, the annotations to the games have to be checked very carefully by the reader himself. Apparently, very often some old analysis has been copied simply without checking for blunders with a modern engine. And in at least one case the authors even got the notation wrong, which has lead to some strange errors.

I shall elaborate on this later.

Does anybody have this book by Mikhalchishin & Mohr and what do you think of it?