Crete Round Five

FACED a German.

Armin Thelen (1972) - Spanton (2013)
Amateur Chess Organization 50+ World Championship Group C

Vienna Gambit

1.e4 e5 2.Nc3 Nf6 3.f4 d5 4.fxe5 Nxe4 5.d3!?

How should Black proceed?

*****

*****

*****

*****

*****

5...Nxc3

This is the most popular move, whereas the tempting 5...Qh4!? has a bad reputation, despite being played by grandmasters. The mainline then continues 6.g3 Nxg3 7.Nf3 Qh5 8.Nxd5 Nxh1 9.Nxc7+ Kd8 10.Nxa8 Bg4 11.Bg2 (a key move, both defending f3 and attacking h1) Nc6 12.d4 and now, instead of 12...Bxf3?, Stockfish17 and Dragon1 recommend 12...f6!?, claiming complete equality.

6.bxc3 d4!? 7.Nf3 Nc6 8.Be2 Bc5 9.0-0!? dxc3+ 10.Kh1 0-0

Now both sides have castled, how would you assess the position (it occurs 107 times in ChessBase's 2024 Mega database)?

*****

*****

*****

*****

*****

Black is a somewhat-precarious pawn up, but has a kingside looking short of pieces. The engines reckon Black is slightly better, but I suspect White's position is easier to play, at least at club level.

11.Qe1 Nd4 12.Bd1

This is the commonest continuation, but the engines also like 12.Qxc3!?, when they reckon best-play runs 12...Bb6 13.Qd2 Nxf3 14.Bxf3 Bd4 15.Ba3 Bxa1 16.Bxf8 Bxe5 17.Bc5, claiming White's lead in development is almost enough compensation for a pawn.

12...Nxf3 13.Bxf3 Qd4

The engines like Glenn Flear's 13...a5!?, the idea being the black queen's rook can quickly switch to the kingside via a6.

14.Qg3 Re8 15.Bh6 Bf8 16.Rae1 Be6!

This apparent-novelty - known moves are 16...Qa4 and 16...Qb6 - is liked by the engines.

17.Bxb7 Rab8

But they reckon a better follow-up is to sac the exchange with 17...Qa4!?, claiming 18.Bxa8 Rxa8 gives Black full compensation.

18.Bc6 Red8 19.Bg5 Rdc8 20.Rf4 Qc5 21.Be4 Rb4 22.Rh4

The engines suggest 22.d4!?, one point being 22...Rxd4 can be met by 23.Bxh7+! Kxh7 24.Qh4+ Kg8 25.Rxd4, winning the exchange.

22...g6

The engines suggest sacrificing the exchange with 22...Rxe4!?, when 23.Rhxe4 (best, according to the engines) Qa3 gives more-or-less enough compensation as White seemingly no longer has mating chances.

23.Bf6?!

White has the upper hand after 23.a3, according to the engines, eg 23...Rb2 24.d4!? Qb5 (24...Qxd4?? 25.Bxg6) 25.h3, averting any chance of back-rack mate and preparing to continue attacking on the kingside.

23...Bg7?

Black is slightly better after 23...Rcb8!, according to the engines, one point being that 24.Rxh7? Kxh7 25.Qh4+ fails to 25...Bh6 26.Bg5 Rb1 27.Qxh6+ Kg8 and 28...Qxe5.

24.Qg5?

Winning seems to be the engines' 24.Bxg7 Kxg7 25.a3!, eg 25...Rb5 (threatening 26...Qxe5) 26.d4 Qe7 27.Bc6! Rb6 28.d5 Bd7 29.Bxd7 Qxd7 30.Qxc3, after which White is a pawn up and with the better-coordinated pieces.

24...Qf2!

The engines agree this is best, and certainly not 24...Qf8? 25.Be7.

25.Rd1?

Necessary is the passive 25.Rg1, but Black is better.

25...Bxf6 26.exf6 Qd2?

'Clever', but White has no good answer to 26...Qxc2 or 26...Rcb8 (27.Qh6 Qxf6).

27.Qxd2 cxd2 28.h3 Rcb8 29.Rxd2 Rb1+?!

This leads nowhere. Instead 29...Bxa2 equalises.

30.Kh2 Rf1?

Also bad is 30...Bxa2? as White can reply, as in the game, 31.Bf3. Instead 30...R1b2 makes sense.

31.Bf3 Rb2!? 32.a3 Rb6 33.Ra4 a6 34.Rb4 Ra1?

Much better is 34...Rxb4 35.axb4 Rb1, although the engines reckon White is still winning.

35.c4?

Much better is 35.Rxb6 cxb6 36.Re2 (or 36.d4), when 36...Rxa3 37.d4 h5 38.d5 Bf5 39.d6 wins easily.

35...Rxa3 36.Rxb6 cxb6 37.Rb2?!

Almost certainly better is activating the king with 37.Kg3.

37...Rxd3 38.Rxb6 h5 39.c5

Black to play and draw

*****

*****

*****

*****

*****

39...Rd8?

Black holds with 39...Rc3 and 39...Kh7, according to the engines, eg 39...Rc3 40.c6 Kh7! 41.Rxa6 Kh6 42.h4=. White cannot make progress on the queenside as after, say 42.Ra7 Kg5 43.c7, the c pawn will not be able to further advance.

The game finished:

40.Rxa6 Rc8 41.c6 Kf8 42.Kg3 Ke8 43.Kf4 Kd8 44.g4 hxg4 45.Bxg4 Bxg4 46.hxg4 Kc7 47.Ra7+ Kxc6 48.Rxf7 Kd6 49.Kg5 Rc4 50.Ra7 Ke6 51.f7 1-0