Welcome to Easychess
Play 1.b3! The Nimzo-Larsen Attack: A Friend for Life (New in Chess) by Ilya Odessky
Recently grandmaster Jonathan Rowson wrote in his column at ChessPublishing.com that at the Dresden Olympiad, he felt he was suffering “from a distinct sense of opening theory fatigue in general, and Slav fatigue in particular”. He wanted to play “something non-theoretical instead and ‘just play chess’ as they say”. He thought that 1 b3 was a move that allowed you to create and make your opponent think from an early stage of the game ...
I should make a confession here. I came to this conclusion around 1985 and, as a consequence, took up 1 b3 as my main White weapon. For around 10 years, I played Larsen’s opening and achieved some of my best results with this line. Rowson thinks:
“The only problem is that 1.b3 is not a great move. It is not even a good move in fact, because Black has numerous ways to develop his pieces and castle without losing either space or structure. However, it is not a bad move, and if you can play it with one or two supporting ideas, an open mind, and some freshness, there are chances to cause problems for the opponent, perhaps no less so than with main lines.”
Correct. The main advantage of playing 1 b3 in the pre-database age days was that you would know the little theory there was on this opening better than even your strongest opponent. I often got my opponent thinking by move five or six and reached congenial positions at a very early stage. Although 1 b3 is not a primarily tactical opening, the middle game is reached with a positionally complex struggle in prospect. It is certainly not an opening like the French Exchange or Petrov where a draw is just around the corner. Eventually, because I grew a little stale (after ten years, this happens) or because in the age of databases even 1 b3 became the object of more analysis, my results started to tail off. I still occasionally roll out 1 b3 for old times sake, but it is no longer my main weapon with White.
I still have fond memories of 1 b3 and purchased Ilya Odessky’s book with some anticipation. Books on 1 b3 don’t appear that often. When I began playing the opening, my bible was Ray Keene’s Nimzowitsch/Larsen Attack – if only because it was the only book on the subject.
The good news is that Odessky plays 1 b3 himself and knows a great deal about the opening. From my point of view it was also interesting that Odessky and I came independently to similar conclusions about certain variations of this opening. I will share some of these conclusions with you later.
First, a word of warning. This is not a repertoire book, nor is it an attempt to cover 1 b3 in depth. For that you must turn to the Nimzo-Larsen Attack by Jacobs and Tait (2001). I must also register a caveat about Odessky’s prose which is very flowery, even to the point of being annoying on occasion. No matter. This book is more than saved by its analysis and thoughts on 1 b3 which you will not find anywhere else. Not all Odessky’s verdicts are favourable to 1 b3, but his judgements are invariably interesting and stimulating.
The chapters are divided as follows:
To begin with, Odessky makes the important point that 1 b3 and 1 … b6 are often similar. Around 1975, Miles, Speelman and other strong English players began to play 1 … b6 with success and the opening therefore became known as the English Defence. Around the same time, 1 b3 fell away from its brief period of popularity when it had been championed by Larsen, Ljubojevic, and even Fischer. But it did not take me long to realise that many of the ideas of 1 … b6 could also be used with 1 b3 – and there was much more theory available on 1 … b6.
Odessky made the same discovery. In the chapter, ‘Wanderer, There is No Path Through…’ (typical Odessky prose) he notes that, except for the placing of the c-pawn, the following two variations are mirror images.
That is how Odessky (and I) came upon the move 6 Nh3 – which is not even mentioned in the book by Jacobs and Tait. Odessky can be strange. He discusses the respective merits of 6 Nh3 and 6 Nf3 at the beginning of the chapter without mentioning that 6 Nf3? would simply be met by 6 … ef whilst 6 Nh3 would answer 6 … ef with 7 Nxf4.
The position after 6 Nh3 has many possibilities, but Odessky only covers two, 6 …Bxh3 and 6 … Nge7, taking the latter as his main line. To be honest, 6 … Bxh3 is met by 7 Qh5+ and 8 Qxh3 and cannot be regarded as Black’s most challenging continuation, but there are at least three other moves worth a mention.
In the last round at the Charlton Congress in 1986, Murray Chandler played 6 … Nh6 (after 1.b3 e5 2.Bb2 Nc6 3.e3 d5 4.Bb5 Bd6 5.f4 f6 6.Nh3). I replied 7.fxe5 Bxe5 (if 7...fxe5 8.Qh5+ Nf7 9.0–0 Be6 10.Nc3 Qe7 11.e4! Berry-Plaskett, Leeds Quickplay 1986) 8.Bxe5 fxe5 9.Qh5+ Nf7 10.0–0 Be6 11.Nc3 with a challenging position for both sides. In the London League in 1994, D. Agnos even tried 6 … h5!? I replied 7.Nc3 Bg4! (7...Bxh3 8.gxh3 exf4 9.Qf3 was my idea) 8.Be2 Bxh3 9.Bxh5+?! though 9.gxh3 exf4 10.Nxd5 would have been better for White.
Simon LeBlancq (Islington Congress, 1985) tried a third possibility 6...Be6 after 6 Nh3. I played 7.0–0 Qd7 8.fxe5 Bxe5? (8 … Bg4 was necessary) 9.Bxe5 fxe5 10.Bxc6 Qxc6 11.Qh5+ Kd7 12.Qxe5 and White had won a valuable pawn.
This trick of winning the pawn on e5 is worth remembering and it appears in many guises if you play 1 b3. One particularly attractive line is 1.b3 e5 2.Bb2 Nc6 3.e3 d5 4.Bb5 Bd6 5.f4 Qh4+ 6.g3 Qe7 7.Nf3 f6 8.Nc3 Be6 9.0–0 a6? 10.Bxc6+ bxc6 11.fxe5 fxe5 12.Nxe5 Bxe5 13.Qh5+ Bf7 14.Rxf7! Qxf7 15.Qxe5+ Qe7 [15...Kd7 16.Na4] 16.Nxd5!
I never got the chance to play this trap, the closest being a game against Keith Arkell in a Kings Head Quickplay. He must have smelled a rat, because on move 12 he diverged with … Nf6 giving up a pawn. Naturally, Black is worse in that position.
Let’s return to Odessky’s 6 … Nge7. He spends time analysing some of his games with7 fxe fxe 8 O-O Bxh3?! (always a positionally dubious move) 9 Qh5+ g6 10 Qxh3 Qd7 11 Qxd7 Kxd7 12 Nc3 and decides that White stands better. He then mentions the possibility of 8 … Bf5! which he regards as the full answer to 6 Nh3.
In fact after 8 … Bf5, the line 9.Qh5+ Bg6 10.Qf3 Qd7 11.Nc3 (better than Odessky’s 11 Ng5) 11 … a6 12.Be2 Nb4 (a computer suggestion) 13.Rac1 gives White the sort of game you can expect with 1 b3. Neither side is obviously better, but there is all to play for. This analysis of 6 Nh3 should give you a flavour of Odessky’s book, not comprehensive, but full of interesting suggestions for both sides.
The next chapter covers the strange and original line 1.b3 e5 2.Bb2 Nc6 3.e3 Nf6 4.Bb5 Bd6 5.Na3. In my old copy of Keene’s book on 1 b3 (1977), 4 … Bd6 is given as a suggestion of Hartston’s whilst 5 Na3 (intending 6 Nc4) is Keene’s own idea. Originally regarded as ‘insane’ (Keene), in the last ten years or so 4 … Bd6 has become one of the main antidotes to 1 b3. Odessky devotes one of his most detailed chapters to this variation and it gets the thumbs up for Black.
I first met the line playing against John Pigott in the London League (1993).
1.b3 e5 2.Bb2 Nc6 3.e3 Nf6 4.Bb5 Bd6 5.Na3 Pigott played what was to me at that time the amazing 5 … Na5 (to stop 6 Nc4). After the further 6.Nf3 a6 7.Be2 Qe7 8.Qc1?! [ 8.Nb1 is better] 8...0–0 9.0–0 b5 10.c4 b4 11.Nc2 c5 12.d3 Bb7 13.e4 Nc6 14.Nh4 we had reached a typical 1 b3 position. Balanced but with chances for both sides.
I should first point out that after 5 Na3 White should look carefully at the move 5 … e4, if only because this is the move the computer plays against me with annoying success. Odessky regards it as “A rare move, but extremely poisonous and dangerous.” Indeed, the whole idea of playing … e4 by Black against 1 b3 should be considered more carefully. Remember that a strategic aim of 1 b3 is to pressure the black pawn on e5. Moving the black pawn one square forwards can be a rather simple solution to this. For instance, after 1 b3 e5 2 Bb2 even 2 … e4 is not totally stupid. Again, in the game Berry – Briscoe, Surrey Championship, 2007, after 1.b3 e5 2.Bb2 Nc6 3.e3 Nf6 4.Bb5 my opponent tried 4 … e4!? which is not given in any books I know of. I have to admit that after 5.Ne2 a6 6.Bxc6 dxc6 7.Ng3 Bd6 8.Bxf6 Qxf6 9.Nc3 Bxg3 10.hxg3 Bf5 White could hardly claim an advantage.
5 … Na5, comparatively unknown in 1993 (Odessky says that 5 …Na5 was first played in Sakaev v Sveshnikov, Gausdal 1992), is today the main line against 5 Na3. Odessky continues with 6 Be2 a6 7 c4 O-O 8 Nf3 and analyses some lines to suggest that the line is playable for Black. But it’s worth remembering that White too is fine and you again have a fresh and interesting position to play. A warning for Black: I would say that you should play 6 … a6, rather than 6 … c6 in this variation. In the game Berry - P. Sowray (Surrey Championship, 2002) Black tried (after 1.b3 e5 2.Bb2 Nc6 3.e3 Nf6 4.Bb5 Bd6 5.Na3 Na5 6.Nf3) the move 6...c6 but this makes the return of the knight from a5 more difficult. The game continued 7.Be2 Qe7 8.Nb1(I had learned from the game against Pigott) 8 … b5 9.0–0 Bc7 10.a4 e4 11.Nd4 d5 12.f4 Bd7 13.axb5 and White stood better.
Move order is a problem with 1 b3. Odessky spends one long chapter ‘The Student’s Problems’ discussing what one should play after the sequence 1 b3 e5 2 Bb2 d6. He considers 3 c4, 3 e3 and 3 g3 and eventually he plumps for 3 e3 because against 3 c4 and 3 g3, 3 … c5 is considered a sufficient answer. He then decides that after 1 b3 c5 2 Bb2 d6 3 e3 Black can play 3 … Nc6 and 4 Nf3 (intending 5 d4) would be met by 4 …e5. Finally, the chapter ends with Odessky triumphantly solving this move order problem with 1 b3 c5 2 Bb2 d6 3 e3 Nc6 4 Ne2! again intending d4. If Black goes 4 … e5 White will reply 5 d4 and 5 … e4 will not hit a white knight on f3.
All well and good. But now we turn to the chapter ‘There is Happiness in Life’ where Odessky considers the move order 1 b3 e5 2 Bb2 Nc6 3 e3 d6. The reader will immediately note that we could have got here via 1 b3 e5 2 Bb2 d6 3 e3 (Odessky’s recommendation in the above paragraph) and now 3 … Nc6. What does he think of this position? Well, after 4 c4 g6 5 Nf3 Bg7 Odessky analyses both 6 d4 e4 7 Nfd2 f5 and 6 Be2 f5! and comes to the conclusion that Black has the initiative. He opines, “All White’s problems, I am sure, start with the move 4 c4. It loses a tempo, it loses the thread of the game, it loses the independence of b2-b3. One must play 4 Bb5. But I really don’t want to. Maybe 4 Nf3!? I don’t know what else to say.”
This is Odessky hedging his bets, especially as he does not analyse either 4 Bb5 or 4 Nf3 in this book. When Black plays either 1 b3 e5 2 Bb2 Nc6 3 e3 d6 or 1 b3 e5 2 Bb2 Nc6 3 e3 Nf6 White has the choice of going with 4 c4 or 4 Bb5. Contra Odessky, I preferred the latter move and achieved decent enough results before running into a the huge problem of the Minasian v Adams game from the European Championship (1992).
After 1 b3 e5 2 Bb2 Nc6 3 e3 d6 4 Bb5 (Adams questions this move) 4 … Bd7 5 Ne2 a6 6 Bxc6 Bxc6 7 O-O Qg5! 8 f3 Nf6 9 c4 d5! 10 Nbc3 O-O-O Black had a tremendously active position and went on to win impressively. I think this line busts 5 Ne2 and was one of the reasons I turned away from 1 b3. If White wants to play 4 Bb5 (remember that Odessky is unhappy with 4 c4) he should try 5 Nf3 when a critical line is that of Berry – Tom Farrand, Kings Head v Wood Green, 1993. The game went 1.b3 d6 2.Bb2 e5 3.e3 Nc6 4.Bb5 Bd7 5.Nf3 and now 5 … e4 6.Nd4 Qg5 7.0–0 Bh3 8.g3 !?
This is an exchange sacrifice which is worth a look. I got the idea from a line in the Sicilian which goes:
1.e4 c5 2.Nf3 e6 3.d4 cxd4 4.Nxd4 Nf6 5.Nc3 Bb4 6.e5 Nd5 7.Qg4 0–0 8.Bh6 g6 9.Bxf8 Qxf8
After 8.g3!? one line might go 8 … Bxf1 9.Qxf1 Nge7 10.Nc3 d5 11.Bxc6+ bxc6 12.Qa6 Qg6 13.Qb7 Rc8 with Black rather tied up. White could open up the game with 14.d3 here and Fritz 11, for one, thinks this is good.
Julian Hodgson tried a more conservative course in Berry – Hodgson, London League, 1994. After 1.b3 e5 2.Bb2 Nc6 3.e3 d6 4.Bb5 Bd7 5.Nf3 he played 5...a6 6.Bxc6 Bxc6 7.0–0 Nf6 8.c4 Be7 9.Nc3 0–0 10.d4 exd4 (10...e4 11.Nd2 d5 12.Qc2) 11.Nxd4 Bd7 12.Qf3 Rb8 and now, if I had played 13.Nf5, things would have been fine for White.
This is all food for thought. But if you don’t like 4 Bb5, I can only suggest 4 Nf3 (played by Sadler) as the move for you.
There are a number of chapters which I don’t intend to cover in any depth in this article. They include two on playable gambits, for those who like that sort of thing (and many do):
In the 10 years that I almost exclusively played Larsen’s Opening with White, the sequence1 b3 d5 2 Bb2 Bg4 became increasing popular, and for two reasons. First, as a result of Julian Hodgson, the Trompovsky 1 d4 Nf6 2 Bg5 was played increasingly frequently in the 1980s. Both 2 Bg5 and 2 … Bg4 have the same aim – to capture the kings knight and double pawns. There is the further point to 2 … Bg4 that White has already prevented his normal counterplay against the early moving of Black’s queen bishop (c4 and Qb3) by playing 1 b3.
Odessky’s final verdict on the Litus Gambit is typically, “Once it was young, then, sadly, it became decrepit rather quickly. But such is life.” In my experience you do not have to play this way with 1 b3. A perfectly acceptable line for White is 1.b3 d5 2.Bb2 Bg4 3.f3 Bh5 and now 4.e3 with the idea of either g4 or the eventual Ne2-g3. Berry – S. Joachim, Dresden Open, 2003 continued 4 … Nd7 5.c4 dxc4 6.Bxc4 e5 7.Ne2 Bd6 8.Ng3 Bg6 9.0–0 Ngf6 10.Nc3 Qe7 and now 11.f4 would have secured a slight edge for White. Equally, the Dutch Defence is not such a fearsome opening that you have play a gambit against it. Just the normal moves are sufficient.
Odessky’s has three chapters on how, by transposition, Petrosian played 1 b3. They will be instructive for those people who intend to play 1 b3 with c4, intending to get into a reverse Sicilian. Odessky then moves on to the chapter ‘Dutch Motifs’ which centres on lines 1 b3 e5 2 Bb2 Nc6 3 e3 d6 4 c4 f5 and 1 b3 e5 2 Bb2 d6 3 e3 f5. I circumvented the first line by 4 Bb5, but the second line was rather popular with strong players. Odessky gives two ways of continuing for White after 1 b3 e5 2 Bb2 d6 3 e3 f5 4 d4 (4 f4 is completely fresh and worth a thought) 4 … e4.
The first is simple development by 5 c4 Nf6 6 Nc3 when one line would be 6 … Be7 7 Nge2 O-O 8 Nf4 c6 9 h4 Na6 10 Be2 Nc7 11 d5 c5 with White castling queenside and eventually breaking with f3 and g4. I played this way myself in Berry v Neil Carr, Middlesex - Essex, 1988.
That game went 1.b3 e5 2.Bb2 d6 3.e3 f5 4.d4 e4 5.Ne2 Nf6 6.Nf4 Be7 7.c4 0–0 8.Nc3 c6 9.h4 Na6 10.Be2 Nc7 11.d5 Nd7 12.h5 Ne5 13.Qd2 c5 14.0–0–0 Bf6 15.f3! exf3 16.gxf3 a6 17.Rdg1 b5 18.h6 g6 19.Nh5 Bh8 20.f4 Nd7 and after 21.e4! b4 22.Nd1 fxe4 23.Ne3 Nf6 24.f5! White had an attractive attacking position.
The second way to play after 1.b3 e5 2.Bb2 d6 3.e3 f5 4.d4 e4 is with the immediate 5 d5!? This is also promising. In the game Minasian-Nevednichy, (Azov 1991) the game went 5 … Nf6 6 Nh3 Nbd7 7 Nf4 Ne5 and now Odessky recommends 8 Bb5+. After 8 … c6 9 dc Qa5+ 10 Nc3 bc 11 Ba4 Be7 12 Qd2, threatening Nxe4, gives White the advantage according to Odessky.
An important chapter is ‘The Anonymous Ending’. This comes after the sequence:
The only difference is that White has a pawn on g3 in variation A, on g2 in variation B. For the record, I found that my opponents almost invariably checked on move five, forcing what they thought was a weakness on the white squares. Odessky takes the main line as variation B, but, for the purposes of this ‘anonymous ending’, it does not matter over much. Odessky (and I) realised that White has a pleasant ending here. White has two pawn islands, Black has three. White can double rooks on the f-file and maybe swing a rook to attack the black a-pawn. The black pawns on the c-file may also become weak. In contrast, Black’s only effective plan is organising d5-d4. Odessky considers five moves here, 13 …Qxf4, 13 …Rfe8, 13 … Rae8, 13 … Qe7 and 13 … Qd6 and regards White’s position as better in all cases. I can add a little to what he says.
In my game against Andrew Muir, the Scottish IM, in the Surrey League 1988 the game went:
1.b3 e5 2.Bb2 Nc6 3.e3 d5 4.Bb5 Bd6 5.f4 Qh4+ 6.g3 Qe7 7.Nf3 Bg4 8.fxe5 Bxe5 9.Bxe5 Bxf3 10.Qxf3 Qxe5 11.0–0 Nf6 12.Nc3 0–0 13.Bxc6 bxc6 14.Qf4
Note that Black played 5 … Qh4+ here, but it really does not make a difference to the ending we are looking at. After 14 …Rae8, which Odessky considers to be one of Black’s best replies, we played 15.Qxe5 Rxe5 16.Rf4 Ne4 17.Nxe4 dxe4 18.Raf1 f5, this move again regarded by Odessky as Black’s best chance. Odessky then recommends 18 g4! g6 19 gf gf 20 Kf2 (it’s move 18 because Black has not checked on move five in Odessky’s example) “ and the further transfer of the rooks to the g- and h-files.”
On the same wavelength as Odessky, my game against Muir continued 19.g4 g6 20.gxf5 gxf5 21.R1f2 (I believe this is just as good as Odessky’s Kf2) 21 …Rf6 (If 21...Rc5 22.c3 Ra5 23.a4 Rb8 24.b4! would be good for White) 22.Kf1 Kf7 23.Ke2 Rc5 24.Kd1 Rd5 25.Kc1 Ke6 (White has marched his king into safety and protected the queenside pawns at the same time) 26.Rh4 Rd7 27.Rh5 Rg7 28.Rf4 h6 29.Rfh4 Rgg6.
(A rather unpalatable situation for Black. The game finished.)
30.Kb2 c5 31.Kc3 Kd5 32.d3! exd3 33.cxd3 Kd6 34.Ra4 Rg5 35.Rh3 Rg1 36.Ra6+ c6 37.Rxa7 Rfg6 38.Rh7 R6g2 (Black threatens mate!) 39.R3xh6+ Kd5 40.Rd7+ Ke5 41.Re7+ Kd5 42.e4+ but it’s White, in fact, who checkmates on the next move.
I had a number of similarly pleasant experiences with this ending and came to the conclusion that Black should keeps the queens on. One idea not mentioned by Odessky is the paradoxical (after 1.b3 e5 2.Bb2 Nc6 3.e3 d5 4.Bb5 Bd6 5.f4 Qh4+ 6.g3 Qe7 7.Nf3 Bg4 8.fxe5 Bxe5 9.Bxe5 Bxf3 10.Qxf3 Qxe5 11.0–0 Nf6 12.Nc3 0–0 13.Bxc6 bxc6 14.Qf4) 14 … Qe7 15 Qh4 (Odessky) and now 15 … Qe5!? The black queen returns to e5, prevents 16 Rf4 because of 16 … g5 and prepares … Ne4. John Benjamin played this way in the Surrey Championship of 1991 and I have to confess that I failed to get any advantage.
Odessky has two chapters which consider 1 b3 d5 2 Bb2 Nf6 3 Nf3 Bf5 4 g3 and 2 … Bg4 3 g3. I personally don’t believe that the g3 setup is the best against … Bf5 (New York System) and, as mentioned earlier, 2 … Bg4 should be dealt with by 3 f3. Nevertheless, if you intend combining 1 b3 with Reti-type systems, these chapters will interest you.
I want to end with Odessky’s coverage of what he calls “The Nimzowitsch Attack”. In my opinion, this is the best thing in the book. Odessky covers two main positions:
a) 1.b3 d5 2.Bb2 c5 3.e3 Nc6 4. Bb5 Nf6 5.Nf3 e6 6.0–0 Be7 7.Bxc6+ bc 8.Ne5 Qc7 9.f4 0–0 10.Rf3
b) 1.b3 d5 2.Bb2 c5 3.e3 Nc6 4. Bb5 Bd7 5.Nf3 Nf6 6.0–0 e6 7.d3 Be7 8.Nbd2 0–0 9 Bxc6 Bxc6 10.Ne5
I always looked forward to playing this way with White, but Odessky has uncovered many defensive resources for Black. I no longer believe that White’s position is anything special in these lines.
In variation a) everything hangs on a nice trap. After 10 … Nd7 11 Rh3 g6? White plays 12 Qh5! Nf6 (12 … gh 13 Rg3+ Kh8 14 Nxf7 mate) and now 13 Ng4!!
Odessky gives the following analysis 13 … gh (13 … Nxh5 14 Nh6 mate) 14 Nxf6+ and now:
All winning prettily for White.
But if Black avoids the mistake 11 … g6? and plays 11 … f6! then White has absolutely nothing. Indeed, after 12 Qh5 fe 13 Qxh7+ Kf7 14 Rg3 Bf6 Odessky believes that White stands worse. What’s more, if Black plays 10 … Ne8!? (instead of 10 …Nd7), then 11 Rh3 g6! is OK because 12 Qh5? fails to 12 … gh 13 Rg3+ Ng7! 14 Nxc6 f6. As Odessky shows, White also cannot get the advantage by with 12 Qg4 because of 12…Bf6.
Odessky further demonstrates that Black can play 6…Bd6 (instead of 6 …Be7) and after 7.Bxc6+ bc 8.Ne5 Qc7 9.f4 0–0 10.Rf3 either 10 … Nd7 or 10 … Ne8 are again fine for Black. The verdict? If Black plays 10 … Nd7, he should follow with …f6. If he plays 10… Ne8, then …g6 is necessary. On the more general level, Odessky points out that White is only attacking with four pieces, sometimes fewer. Should he really be expected to mate Black? Or as Odessky puts it, “In the Nimzowitsch Attack, White makes all the running. And Black determines the result of the game.”
The white setup in variation b) above is more restrained. It was a favourite with Nimzowitsch, but again Odessky is on the side of Black. I met this line when playing Harriet Hunt in the Sutton Congress of 1993. After:
1.b3 d5 2.Bb2 Nf6 3.e3 c5 4.Nf3 Nc6 5.Bb5 Bd7 6.0–0 e6 7.Bxc6 Bxc6 8.Ne5 Rc8 9.d3 Be7 10.Nd2 0–0 11.f4 Nd7 12.Qg4 Nxe5 13.Bxe5
we had reached (by transposition) the famous games Nimzowitsch v Spielmann (New York 1927) and Fischer v Mecking (Palma de Mallorca, 1970) both of which seemed to suggest that this line is much better for White. Odessky shows otherwise and brings many improvements to the table.
.13…Bf6?! was the choice of the black players in all the games. First, Odessky shows that Black should play 13 … g6!, defend the e6 pawn and then push …f6. He is then probably better.
After 13…Bf6?! 14.Rf3 Fischer v Mecking continued 14…Qe7?! 15.Raf1 a5 16.Rg3 Bxe5 17.fxe5 f5 18.exf6 Rxf6 19.Qxg7+! Qxg7 20.Rxf6 Qxg3 21.hxg3 Bd7 22.g4! and White won the ending. In my game against Hunt, everything was the same except that Harriet played 15 … b5 (rather than 15 …a5). That game too went 16.Rg3 Bxe5 17.fxe5 f5 18.exf6 Rxf6 19.Qxg7+! Qxg7 20.Rxf6 Qxg3 21.hxg3 Bd7 22.g4 Rc6 and now 23 e4! proved decisive (if 22 …Ra6? 23 ed wins).
But, in these games, Black could have played 15 … g6 or even 16 …g6 and stood OK. It is only the persistent refusal of Black players to push … g6 which has unjustly resulted in this line being judged as good for White.
Odessky has a further point to make. Instead of 14 … Qe7?! Black can play 14 … Bxe5 15 fe Qc7 16 Qh5 as in the Nimzowitsch v Spielmann game. Here Black played 16 … h6? but after 17 Raf1 g6 18 Qxh6 Qxe5 19 Rf6! stood worse. His best is 16 … Be8 when I wanted to play what I believed to be the powerful move 17 Rf6! stopping …f5 by Black.
As Odessky points out, the rook cannot be taken. 17 … gf? 18 ef Kh8 19 Nf3 Rg8 20 Ng5 Rxg5 21 Qxg5 and I had been happy with the knowledge of this. Unfortunately, Odessky also points out that 17 … c4! defends for Black and he seems to be right. For instance, 18 Nf3 cd 19 cd (19 Ng5 h6 20 Rxh6 gh 21 Qxh6 f6) 19 … Bb5 20 Rd1 Bxd3! works for Black after 21 Rxd3 Qc1+ 22 Kf2 gf 23 ef Qb2+ 24 Rd2 Qxf6. All in all, Black is very much OK in this line.
After the sequence 1.b3 d5 2.Bb2 c5 3.e3 Nc6 4. Bb5 Bd7 5.Nf3 Nf6 6.0–0 e6 7.d3 Be7 8.Nbd2 0–0 9 Bxc6 Bxc6 10. Ne5, apart from 10 … Rc8, Odessky also considers the moves 10 … Be8, 10 … Qc7 and 10 … Nd7. His opinion on the Black prospects is generally favourable, especially with the last move.
The final paragraph of Odessky’s book reads:
“Dear readers! Play 1 b3. Probably it’s not so good. But it’s so much fun.”
I can agree that the move is fun, but I think Odessky rather downplays the strength of 1 b3. If you are not worried when your opponent’s pawns occupy the centre, perfectly happy to play the Modern Defence or Gruenfeld with Black, then this could be the opening for you. I doubt if, in the modern age, you should make 1 b3 your main weapon with Black. But if you do decide to use it as a surprise weapon, you will be pleasantly surprised to find that some of your opponents have used over 30 minutes on the clock before they have reach move 10. You may find too, as I did, that your results improve as you play 1 b3. Primarily a 1 e4 player, I found that using Larsen’s Opening improved my positional play. You rarely achieve the crushing positions you can sometimes get with 1 e4, but you do have to learn how to nurture a small advantage, one of the most important prerequisites of being a decent chess player.
As I have stressed, 1 b3 is primarily a positional opening, but attacking play is by no means excluded. I will end this review with two games which, I hope, will encourage the reader to try the unjustly neglected 1 b3.
(1) Berry,S - J. Klinger (G.M.) [E14]
Cappelle la Grande, 1994
(2) Berry,S - Britton,R [A01] Peter de Beauvoir, 1995