Time To Improve

I PLAYED in a nine-round swiss in Malta from Nov 19-25.
The tournament was organised by Juergen Wempe, who also runs one of my all-time favourite events, Bad Woerishofen in Bavaria.
The Maltese open was blessed by excellent weather and very good playing conditions at the Topaz Hotel, Bugibba.

The well-lit main playing hall

My final score of +3=4-2 was sufficient to gain 21.2 Fide elo, which in itself is good enough reason to go back next year!
They say you learn most from your losses, but I think I picked up quite a bit of knowledge from the following draw.
Per Green (1973) - Spanton (1859), Malta 2017, Round 8
I had noticed in preparing for this game that PG nearly always played 1.Nf3. I intended to offer a Sicilian by replying 1...c5, to which I expected, judging by his repertoire, that he would play 2.c4. We were chatting idly before the round started, and PG made reference to the fact that I enter a lot of tournaments. I took this to be evidence that he had looked me up. So when he opened with ...
1.c4
... I decided he wanted a Symmetrical English - these days I normally reply 1...c5 to 1.c4 - but without being committed to bringing the king's knight to f3. Accordingly, I switched to another old favourite of mine, viz:
1...e6
The game proceeded:
2.g3 d5 3.Bg2 Nf6 4.Nf3 c5 5.e3
I expected 5.cxd5 or 5.d4. In fact 5.0-0 is the commonest option. The text, although looking strange to me, is also popular.
5...Nc6 6.b3

Can Black take serious advantage of White's last move?

I thought White's last was a straightforward mistake, but the position has been reached by some strong players, including Botvinnik in a consultation game against Keres.
6...dxc4 7.bxc4 Qd3 8.Qb3!
Since this loses the bishop-pair, I had thought White was practically obliged to play 8.Bf1, after which 8...Qd7 9.Bg2 would have been a clear gain-of-tempo for Black.
8...Nb4 9.Na3N
This seems to be a new move. Two games in ChessBase's 2018 Mega database feature an immediate queen swop.
9...Qxb3
It was better to get on with development, but I did not want to miss the chance of exchanging knight for bishop.
10.axb3 Nd3+ 11.Ke2 Nxc1+ 12.Rhxc1
It was around here I realised that, far from having an advantage, I could be in serious trouble. Black has the bishop-pair, but White has a massive lead in development. Even his king is well-placed for the late middlegame, with nothing to fear from Black's inactive pieces.
12...a6
Played to prevent Nb5, but it does nothing of the sort!
13.Nb5 Kd8?
For some reason I rejected the superior 13...Rb8 with hardly a second glance.
14.Ne5 Ke7 15.Nc7 Ra7
Definitely not 15...Rb8? 16.Nxa6!
16.Nb5 Ra8 17.Rd1

Black faces a tricky decision after 17.Rd1

17...Ne8
I rejected 17...Nd7 because of 18.Nxd7 Kxd7 19.d4. I rejected 17...Rb8 because of 18.Na7 Nd7 (I think I also saw 18...Bd7 19.Rxa6!) 19.Nxc8+ Rxc8 20.Bxb7.
18.d4 f6 19.Nd3 cxd4 20.exd4 Kf7 21.Re1 Rb8 22.Nc3 h5 23.h4 Nd6 24.Kd2 Nf5 25.Ne2 Be7 26.Kc3 g6
Somehow I have managed to avoid material loss while slowly getting my pieces developed. 27.Nb4?!
Nevertheless White could have maintained a clear advantage with 27.Be4 according to Stockfish8, or 27.c5 according to Komodo9. The text is presumably aimed at trying to create a passed d pawn, but instead the knight becomes the means by which Black untangles his queenside.
27...Rd8 28.Rad1 a5 29.Nc2 b6 30.Bh3 Ng7
I tried but failed to make 30...Nxd4?? work. Instead, S8 points out 30...e5!? 31.dxe5 Rxd1 32.Rxd1 fxe5, when Black's bishops are starting to get some freedom.
31.Nf4 Bd6 32.Nd3 Bb7 33.Re3 Nf5!? 34.Bxf5 exf5
34...gxf5 looked too loosening to me, but is preferred by the analysis engines.
35.d5 Re8
White wins the e6 square, but it turns out he cannot really afford this as his king is vulnerable. S8 reckons Black is clearly better after 35...Rbc8 eg 36.Kb2 b5.
Instead, I neutralised White's pressure along the e file, but after ...
36.Rde1 Rxe3 37.Rxe3 Re8 38.Rxe8 Kxe8 39.f4
... the game ended in a draw following ...
39...Bc8 40.Kd4 Ba6 41.Kc3 Bc8 42.Kb2 Ke7 43.Nd4 Bd7 44.Kc3 Kd8 45.Nc2 Bc8 46.b4 axb4+ 47.Ncxb4 Bb7 48.Nc2 Bc8 49.Nd4 Bd7 50.Kb3 Kc7 51.Nc2 Bc8 1/2-1/2.