1.b3 I didn't know Wei Kai played this. I couldn't remember any details from previous attempts to study this move. I did remember reading somewhere that White's first move announces a fight for the e5 square and that 1...d5 concedes the fight prematurely. So I decided to go with the (amazingly subtle) plan of playing 1...e5 instead, developing some pieces, and seeing what happens. Actually now I come to write this I remember that the unknown author who supplied the nugget of wisdom in question was writing about 1.f4, so I wasn't actually starting the game on a firm foundation. As one starts so one continues might be a subsidiary theme of the game. ( Our main game was played in the C Grade of the Wellington Chess Club's Autumn Cup. I am just going to insert another (slightly higher level :-) game at this very early stage only due to the technical issue that it doesn't have even a single move in common with the main game. I'll come back to it, but for now you'll have to trust me that it is relevant to the themes we are going to cover. Aronian-Kramnik, World Championship Candidates, Berlin 2018 1.e4 e5 2.Nf3 Nc6 3.Bb5 Nf6 4.d3 Bc5 5.Bxc6 dxc6 6.O-O Qe7 7.h3 Rg8!! 8.Kh1 Nh5 9.c3 g5 10.Nxe5 g4 11.d4 Bd6 12.g3 Bxe5 13.dxe5 Qxe5 14.Qd4 Qe7 15.h4 c5 16.Qc4 Be6 17.Qb5+ c6 18.Qa4 f5 19.Bg5 Rxg5 20.hxg5 f4 21.Qd1 Rd8 22.Qc1 fxg3 23.Na3 Rd3 24.Rd1 Bd5 25.f3 gxf3 26.exd5 Qe2 27.Re1 g2+ 0-1 ) 1...e5 2.Bb2 Nc6 3.e4 After the game Wei Kai told me he usually plays the more thematic and usual e3 instead. That move in conjunction with Bb5 adding pressure to e5 (see our main theme later) is very logical. But here I think Black is already comfortable, as the fianchetto doesn't usually fit in with the kind of classical 1.e4 e5 position that develops now  3...Bc5 4.Nf3 d6 5.Bc4 Nf6 6.Nc3 Bg4?!

Someone should write a book about all the ideas wrapped up in playing Bg4/g5/b4/b5 and the possible responses to h2-h3 (or h7-h6 and the rest) including the exchange on f3 or the retreat and the possible further retreat after g2-g4, Bh5-g6.

Incidentally as an old person I was brought up on descriptive notation, and I still maintain it has real advantages, including discussions like this. In old money we'd simply say B-N5 P-R3; B-R4 P-N4; B-N3. This would cover all four possible arrangements of the concept, without worrying whether it was necessary to clumsily enumerate them all. Cecil Purdy would use this property of descriptive notation extensively in his wonderful books and articles, and it is really painful to see his elegant prose mangled by the necessity to translate to algebraic in modern re-publications.

In the absence of a Bg4 book, I am going to briefly outline some of the ideas, no doubt a strong player will let me know if I am getting any of this wrong.

Sometimes the idea is simply to exchange the bishop, either because the bishop is a poor piece or to relieve a cramp when your pawn structure doesn't really provide a large enough home for all your minor pieces. That's not what I was going for here.

Sometimes you are trying to pressure the d4 square, so much the better if there's a pawn on it. That doesn't really apply here either.

Sometimes you are just hoping for a persistent (even permanent) annoying pin. If Be2 or Nbd2 is either unavailable or unattractive (because the piece in question is already developed), that adds weight to this idea. Now we're talking.

Annoyingly of course unless there's no pawn on g2 the opponent can always just unpin by moving their queen away, letting you capture and spoil their pawn structure. Often it's not that big of a deal in the end, even if they have committed their king to the damaged side. But let's ignore that aspect for now and leave it to the book I'm waiting for.

If YOU HAVEN'T castled short that adds weight, because the response h3 and g4 isn't going to create a pawn storm attack on your king with tempo. BUT really, you only get the best of that aspect if THE OPPONENT HAS castled short. Because then challenging the pin with h3 and maybe g4 creates weaknesses in front of their king that you can exploit by castling the other way.

Even great players can forget about this, see Aronian-Kramnik at the start of the article. Kramnik never actually played Bg4, but no matter Aronian made the characteristic h3 weakness anyway anticipating the key move. I wonder whether Aronian realised what he'd done immediately, or whether he only broke into a cold sweat after 7...Rg8 appeared on the board? It's a measure of Kramnik's awesomeness that he can play an immortal B-N5 game without even playing B-N5.

I was hoping to spring the same positional trap on my opponent, but crucially I forgot I needed to wait until he had played O-O before I ventured Bg4. Of course although I'm old, I'm a millennial at heart so I'm going to blame someone else for my mistake. Some lazy bugger hasn't got around to writing the book I needed to get my thinking straight in advance.

Actually if you really think about it, it's not just the chess publishing industry that should take the blame, algebraic notation is also at fault. If we still used descriptive, you could title the book B-N5! If a great title like that was available I'm sure the book would have been written by now. Unfortunately the publishers realise that "Bg5!/Bb5! or ...Bg4!/...Bb4!" is not a title that's going to zoom to the top of any bestseller lists.

My brief sojourn has barely touched the surface of the material a B-N5! book could cover. For example, in the most fundamental classical openings, White plays B-QN5 (Spanish) and B-KN5 (Queens Gambit), for a variety of subtle reasons that I won't discuss because I don't really understand them. I'm waiting for the book.  (clearly I needed to play a waiting move, perhaps 6...a6 first and wait for him to castle ) 7.h3 Bh5 8.g4 Bg6 Of course by now I had realised my mistake, White is never going to castle kingside now.  9.Qe2 Qd7 10.O-O-O O-O-O I felt obliged to follow else I am castling into a pawn storm that's already well underway.  11.d3 Kb8 This way I can kind of pretend I've castled normally on the kingside like a sane person. I almost always do this after castling long. I'm not very rational when it comes to chess. ( After the game Michael Sole pointed out that with e4 now firmly protected White can play Na4 effectively, so I should definitely get in 11...a6 now to provide a retreat for the bishop ) 12.Nd5 (12.Na4! Stockfish agrees with Michael Sole, rating this move as equal and other moves as advantage to Black. ) 12...Nxd5 13.Bxd5 Ne7 I am actually threatening c6 and b5, so Wei Kai played  14.d4?! which feels wrong because it tends to create an open position where I have the bishop pair  14...exd4 ( I didn't play the superior 14...Nxd5 because I didn't like  15.dxc5 Nf4 and White has a pin on the d-file. But the computer simply plays  16.Qd2 Qe7 17.cxd6 Rxd6 and Black is better ) 15.Nxd4 (15.Bc4 White could have kept his bishop ) 15...Nxd5 16.exd5 Rhe8

17.Qd2 ( I admit it, I was hoping for 17.Qb5?? Bxd4! and wins. I hope this doesn't mean I'm a bad person. ) 17...Be4 18.Rhg1! ( I initially thought this was a mistake, I expected the more natural 18.Rhe1 and I was struggling to find a good move after  18...Bxd5 19.Nf5 Bf3 20.Nxg7 but when the same moves were played in the game I belatedly realised that Wei Kai's move was in fact correct because  20...Rg8! would have been available if he'd played my way at move 18  ) 18...Bxd5 19.Nf5 Bf3 20.Nxg7 Re2 This looks good too, Black's initiative is growing and the bishops are doing good work  21.Qd3 Qc6 22.Rg3 I was expecting this logical looking move, but the computer hates it. As I type this I still don't know why, I guess I (we) will find out in a few moves ( of course not 22.Qxh7?? Be4 ) ( the computer suggests 22.Rgf1 and grovelling ) 22...Rxf2 23.Re1 Qd5 I spent time here looking for a combo finish, but couldn't find anything. So I decided to try and just convert my extra pawn and two bishops. But my pieces get slightly uncoordinated (   23...Bg2! The computer calmly moves the Bishop to a protected square, puts the Rg3 in a box of sorts and, keeps the queens, and the pressure, on. It rates this a -2 pawns advantage ) 24.Qxd5 Bxd5 25.Nf5 Rf4

The computer is unimpressed, but dammit I like this move - I need to get my other rook in the game to win, and I can do it by disputing the e4 square. I also threaten Bf2  26.Rd1 Be4 27.Nd4 Re8 28.Re3 d5! I've straightened out my pieces, they're all doing God's work now and I should win  29.Rde1 a6 Just in case of later back rank issues  30.c4

Lashing out hopefully  30...dxc4 31.bxc4 Rd8?? I still can barely believe I played this. I was experimenting with a technique involving playing some moves instantaneously if play was heading down a road I'd anticipated and I'd thought about them enough in advance. But really, to make that work I need to be able to properly 'see' a position two half moves in the future and clearly I can't!  32.Rxe4 Oh well. I tried to keep a straight face. May as well keep as many pieces on as possible, try to swap pawns only, play on for a while, miracles can happen ( Somehow I envisaged a position where white had to play 32.Rd1 and be all tied up ) 32...Rf2 33.Re8 Kc8 34.Nb3 Bb4 35.Rxd8+ Kxd8

36.Bd4?? Miracles do happen. My opponent picked up his bishop with the intention of playing Bf6+ and noticed just in time that my rook covered that square. It might have been better to just play it - in this form maybe I wouldn't have noticed! Sadly for him, touch move means that he has to choose from the possible Bishop moves and they're all blunders  36...Bxe1 37.Bxf2 Bxf2 (=) Why the draw offer in this winning position? My opponent was visibly distraught at his mistake, and I knew exactly how he felt having been there myself only five minutes earlier. He had only made half a blunder, it was very unfortunate that he had no decent Bishop move. So my sin of one massive blunder exceeded his of only half a massive blunder, why should I win the game? I also felt some residual guilt from the previous week when I had inadvertently (tried to) violate the touch move rule. On that occasion I too had picked up a bishop, realised it was a mistake, and put it down. I then decided Kg5 was the right move, realised I couldn't play it (touch move), but then unaccountably tried to play pawn f2-f3 instead! At least my opponent had the decency to obey touch move without being asked! Also I kind of felt the two horrible incomprehensible blunders had surely ruined the game and rendered any result meaningless. Finally the same game a week earlier had also proved that I'm not up to converting winning endings anyway (it was a good bishop against a knight with an extra pawn that time too). Still, I probably should try and develop a ruthless streak. Maybe try to play chess in the manner of a flinty Australian cricketer who likes nothing better than being dropped because they then get to enjoy the extra pain their runs inflict!  1/2-1/2