Lessons From Bregenz VII

TWO things I have a tendency to bash on about in this blog is the advantage of having the bishop-pair and the danger of giving up a good bishop when you have a bad one.

But look at the following position, which was reached in round seven, where I was Black against Bernard Logie (1822).

Black has just played 6...Qd8-b6

*****

*****

*****

*****

*****

The position occurs 40 times in ChessBase's 2026 Mega database, and in 22 of those games White continued 7.Bxc6+!?, which was also my opponent's choice, and is the top choice of Stockfish18 and Dragon1. The engines reckon the move gives White a slight edge (Stockfish18) or even the upper hand (Dragon1).

I recaptured with the pawn, which is the main line in Mega26, but the engines reckon recapturing with the queen is better.

Position after 7...Qxc6!?

*****

*****

*****

*****

*****

There is no doubt White has given up the bishop-pair, and not just on a short-term basis in the confident expectation of quickly swopping off one of Black's bishops.

There is also no doubt White has given up his better bishop, as the dark-square remainder is restricted by the pawn-chain b2-e5.

White's one advantage in the position, as it seems to me, is having extra space in the centre, which must be why the engines reckon White is better.

A good way to try to understand such a position, I believe, is to let the engines play it out against each other, so here goes.

8.dxc5!?

All seven games to reach the position in Mega26 saw 8.0-0, but the text is Stokcfish18's top choice and Dragon1's second choice (behind 8.h3).

The text makes White's dark-square bishop less bad, but also slightly opens the position, which normally favours the bishop-pair, and weakens e5.

8...Bg4!? 9.Nbd2!? e6

Not 9...Bxe5? 10.Nxe5! Bxd1 11.Nxc6, when White has won a piece.

10.b4

The engines fluctuate between 10.b4, 10.0-0 and 10.Qb3, but come to more-or-less settle on the text.

10...b6!? 11.h3 Bxf3 12.Nxf3 bxc5

The bishop-pair has gone, but Black has recaptured the pawn lost at move eight

*****

*****

*****

*****

*****

13.0-0

The engines cannot agree on which continuation is best, but this is the only move that makes both engines' top two.

13...Ne7 14.Bg5 Qc7 15.Qa4+ Nc6!?

The only move that makes both engines' top two.

16.Be3!? Qd7

Again the engines are unsure what to play, and this is the only move that makes both engines' top two.

18.Bxc5 Nxe5 19.Qd1 Nc4 20.Rc1 Rc8

I am going to leave it here as we have come a long way from the initial decision to swop bishop for knight

*****

*****

*****

*****

*****

The final position looks unclear to me, with lots of imbalances that would make for a tricky middlegame, but the engines give White a slight edge (Stockfish18) or the upper hand (Dragon1).

LESSON: there is no doubt having the bishop-pair is usually advantageous, and giving up a good bishop when you have a bad one is usually disadvantageous, but what really counts is what happens appen next, and it seems most people who have reached the position in the first diagram as White have understood what was going on rather better than I managed.

Bregenz Seniors Round Seven

Overnight hail and snow

FACED a Belgian.

Bernard Logie (1822) - Spanton (1911)

Sicilian Hyper-Accelerated Dragon

1.e4 c5 2.Nf3 g6 3.c3 d5 4.e5!?

Stockfish18 and Dragon1 prefer this over the more popular 4.exd5, and it scores five percentage points better in ChessBase's 2026 Mega database.

4...Nc6 5.d4 Bg7

This natural-looking move may be a little slow; the engines prefer 5...Bg4 or 5...cxd4.

6.Bb5!? Ob6!?

How should White proceed?

*****

*****

*****

*****

*****

7.Bxc6+!?

Giving up White's good bishop is the top choice of the engines.

7...bxc6

The engines reckon the less popular 7...Qxc6!? is better.

8.0-0 e6?!

The engines suggest an apparent-novelty in 8...f6!?

9.Na3 Ne7 10.Nc2 0-0

Now both sides have castled, how would you assess the position?

*****

*****

*****

*****

*****

White's central space advantage is much more important than Black's bishop-pair, according to the engines, and they reckon White has a positionally won game.

11.b4

The engines suggest 11.b3.

11...c4?!

Reducing the light-square bishop to a sorry looking piece,

12.a4

BL pressed the clock and immediately offered a draw.

12...Nf5 13.a5 Qc7 14.g4!?

This is the top choice of Stockfish18, and is also liked by Dragon1, although the latter marginally prefers 14.Bf4.

14...Ne7 15.g5!?

Again Stockfish18 approves; Dragon1 suggests 15.h4 or 15.Qe2.

15...Nf5 16.Ne3!? Nxe3

Otherwise White has 17.Ng4.

17.Bxe3 Ba6 18.Re1 Rfd8 19.Kg2 Rab8 20.Qc2 Bb5 21.h4

White I was thinking about my next move, BL offered a draw, commenting that he was tired

*****

*****

*****

*****

*****

White's advantage is worth about two pawns, according to the engines, but in human terms the advantage is surely worth much more as White can build up at leisure on the kingside, safe in the knowledge that any possible counterplay by Black will involve giving up a pawn for nebulous compensation.

½–½

This game ended a sequence for me of 11 without a draw, and I have had just two draws in my last 21 games.

Bad Wörishofen Senioren Round Four

Ulrich Fernschild (1671) - Spanton (1919)

Colle-Zukertort

1.Nf3 d5 2.d4 c5!? 3.e3

This narrowly edges out 3.c4 for popularity in ChessBase's 2026 Mega database, but the latter move is preferred by Stockfish17.1 and Dragon1.

3...Nf6 4.b3 e6 5.Bb2 cxd4!?

How should White recapture?

*****

*****

*****

*****

*****

6.exd4!?

This is the top choice of the engines, even though the move blocks the long dark-square diagonal. The point is that the white queen's bishop will play an important role in supporting the king's knight, which at some point is likely to land on e5.

6...b6!?

A rare continuation.

How should White respond?

*****

*****

*****

*****

*****

7.Bd3

Yahli Sokolovsky (2369) - José Eduardo Martínez Alcántara (2591), Chess.com Blitz 2022, continued 7.Bb5+ Bd7 8.Bd3, with a slight edge for White, according to the engines (but 0-1, 55 moves).

7...Bb7 8.Nbd2 Nc6!?

This interferes with the light-square bishop's action on the long light diagonal, but means White will probably be obliged to spend a tempo on preventing ...Nb4.

9.a3

This is not immediately necessary, eg 9.Ne5!? prevents 9...Nb4?? as the latter loses to 10.Bb5+ Ke7 11.Ba3, with c3 to come.

9...Bd6

Carlsen played 9...g6!? in a rapid game, but the engines are unimpressed. 

10.Ne5 Qc7 11.f4

The engines agree this is better than 11.Ndf3, but they also like 11.Qe2!?

11...Ne7!?

The engines suggest an apparent-novelty in 11...Ne4!?, the point being 12.Nxe4 dxe4 13.Bxe4 can be met by 13...Nxe5, when 14.Bxb7 Qxb7 15.fxe5 Qxg2 16.Rf1 Rc8!? seems roughly equal.

12.Rc1 0-0 13.0-0

Now both sides have castled, how would you assess the position?

*****

*****

*****

*****

*****

White has a well-placed knight at e5, and more space on the central-kingside, but Black has no weaknesses, and enjoys some pressure against the white queenside. Equal chances, according to the engines.

13...Rac8 14.Qe2 Ng6 15.c4

The engines suggest another apparent-novelty, this time 15.c3!?, a possible continuation being 15...a6!? 16.a4 Ne7!? 17.Bb1!? (17.Bxa6 Bxa6 18.Qxa6 Nf5 gives Black reasonable compensation for a pawn, according to the engines) Ng6!? 18.Bd3!? Ne7!? 19.Rf3, with a roughly equal position, but there is a lot about these moves I do not understand.

Popular in Mega26 is 15.g4, when the engines suggest 15...b5!?, the point being 16.Bxb5 can be met by 16...Ne4, eg 17.Qe3 Nh4 18.Be2 f5, with what they reckon is more than enough for a pawn. And if White instead continues on the kingside with 16.g5, then 16...Ne4 17.Nxe4 dxe4 18.Bxe4 Bxe4 19.Qxe4 f5!?, and again Black has good compensation for a pawn, according to the engines.

15...Qe7 16.b4 dxc4 17.Ndxc4 Bb8 18.Nxg6

Vacating e5 so the square can be occupied by White's other knight, but the engines disapprove, suggesting 18.b5!?, 18.f5 or 18.g3.

18...hxg6 19.Ne5 Nd5

Both knights are well-placed

*****

*****

*****

*****

*****

20.Ba6?!

Seeking to swop off White's good light-square bishop, but this is probably inadvisable at a time when White is saddled with a dark-square bishop that could be a liability in an endgame as it is hemmed in by its own pawns.

20...Rxc1 21.Bxc1

Forced, as the king's rook is needed to defend f4.

21...Rc8 22.Bxb7

It is difficult to safely withdraw the bishop, eg 22.Bd3 can be met by 22...Rxc1 23.Rxc1 Nxf4 24.Qe3 Nxg2. However, possible is 22.Nxg6!? fxg6 23.Bxb7 Qxb7 24.Qxe6+, but it seems the engines' 24...Kf8 25.f5 g5 wins for Black, although the position remains sharp.

22...Qxb7 23.Bb2

23.Nxg6!? fxg6 24.Qxe6+ transposes to the previous note.

23...Rc7 24.b5!?

How should Black respond to White's threat to play Nc6, blocking the open file? 

*****

*****

*****

*****

*****

24...Ne7

This may be best. Black could sacrifice the exchange with 24...Qc8 25.Nc6 Rxc6!? (25...a6 26.a4) 26.bxc6 Qxc6, but 27.Rc1 lets White fully mobilise, giving a completely equal position, according to the engines. The alternative pawn captures, 26...Bxf4 and 26...Nxf4, do not seem to be improvements. However the engines point out the possibility of temporarily sacrificing the knight with 25...Nxf4!?, when 26.Rxf4 Rxc6 27.bxc6 Bxf4 nets a pawn, although the position is not completely clear after the continuation 28.g3!? Bd6 29.Qc4. One point about the text is that, as well as preventing Nc6, it rules out Nxg6 ideas.

25.Qd2

The engines dislike this, suggesting 25.h4, or the pawn sacrifice 25.d5!? to liberate White's bishop. Either way, they reckon Black has at least the upper hand.

25...Qd5 26.Qb4?

This allows liquidation into an ending that White has little chance of holding, so the engines suggest 26.Qd3 or 26.Qe2!?, albeit agreeing White is anyway losing.

26...Rc2 27.Rf2

Both 27.Nf3? Bd6, and 27.Rf3? Bxe5 28.fxe5 (28.dxe5? Qd1+ 29.Rf1 Qe2) Nf5, are even worse for White.

27...Rxf2 28.Kxf2 Bxe5 29.dxe5

White's king is even more exposed after 29.fxe5?!

How would you assess this queen-and-minor-piece ending?

*****

*****

*****

*****

*****

White's bad bishop and exposed king, combined with Black's good knight, make this a very good ending for Black, provided a) perpetual check is prevented, and b) queens stay on the board (at least until any resulting minor-piece ending is a trivial black win).

29...Nf5 30.g3?

UF was presumably worried about ...Nh4, but the engines give 30.Bc3!? as 30...Nh4 can be met by 31.Kg3! However, 30...Nh6 is strong, eg 31.Bd4 Qe4 32.Qd2 Qb1 33.Bb2 Qa2 34.h3 Qb3 35.Qe2 Nf5, when White is still hanging on, but the practical difficulties of defending such a position are huge.

30...Qd3

Probably even stronger is 30...Qh1.

31.Qc3!?

This is best, according to the engines, but now b5 falls.

31...Qxb5 32.Qd2 Qc5+ 33.Kg2 Ne3+ 34.Kf3 Nd5 35.Qc1!? Qb5

35...Qxc1?! may win, but it is a lot less clear than keeping queens on.

36.Kg2?!

This speeds the losing process, but White was anyway the equivalent of almost a rook down, according to the engines.

36...Qe2+ 37.Kh3 Ne3 0-1

Entrance to Villa Diogenes at Hotel Alpenhof

Bad Wörishofen Senioren Round Two

FACED an Austrian.

Robert Hafner (2031) - Spanton (1919)

Colle-Zukertort

1.d4 d5 2.Nf3 e6 3.e3 Nf6 4.b3 Bd6

Throwing in the moves 4...Bb4+ 5.c3, and then 5...Bd6, probably does not make much difference as White will likely want to play c4 at some point anyway.

5.Bb2 0-0 6.Be2 b6 7.Nbd2 Bb7 8.0-0

Now both sides have castled, how would you assess the position?

*****

*****

*****

*****

*****

The pawn-structure is symmetrical, and the piece-setup is not far off it. Stockfish17.1 and Dragon1 reckon the game is equal.

8...Nbd7 9.c4 Rc8 10.Rc1 Qe7 11.Qc2 c5 12.Rfd1 Rfd8 13.Qb1 h6!?

Possibly a novelty. The engines suggest 13...dxc4, 13...cxd4 or 13...Bb8.

14.h3 Ne4!? 15.cxd5!?

This is the engines' top choice. They reckon 15.Nxe4 dxe4 gives Black at least a slight edge.

15...exd5 16.dxc5!?

Again the pawn capture gets engine approval.

16...Bxc5

The engines are not keen on this, reckoning 16...Ndxc5 keeps White's advantage to a slight edge.

What is Black's threat, and how should White respond?

*****

*****

*****

*****

*****

17.Nd4!?

This is the engines' top choice, although they also quite like 17.Nxe4. White had to be careful, though, as Black threatened 17...Nxf2!

17...Qh4?

Best seems to be the engines' 17...Ne5, but also better than the text are 17...Nd6 and 17...Bxd4!?

18.Nxe4 Qxe4

Worse is 18...dxe4? 19.Nf5.

19.Bd3

Even stronger, according to the engines, is 19.Qxe4!? dxe4 20.Nf5, eg 20...Bf8 21.Rxc8 Bxc8 22.Bb5 a6 23.Bc6 h5 24.g4 hxg4 25.hxg4 g6 26.Bf6 Re8 27.Bxd7 Bxd7 28.Rxd7 gxf5 29.gxf5, when White is a pawn up and with much the better pieces. If 20...f6, then  21.Bc4+ Kf8 22.Be6 Bc6 23.Nd4 Ke7 24.Bg4 is very good for White.

19...Qe8 20.Nf5 Ne5

Not 20...Bf8? 21.Nxh6+!

21.Be2!?

The engines fluctuate between the two bishop retreats.

21...Bf8 22.Qa1!?

The engines like this. It induces a weakening of the black king's position.

22...f6 23.Qb1!? Rxc1!?

Played on the principle that a weak IQP is easier to defend with fewer heavy pieces on the board.

24.Rxc1 Rc8 25.Nd4!? Rxc1+ 26.Qxc1 Qc8 27.Qxc8 Bxc8

How would you assess this minor-piece ending?

*****

*****

*****

*****

*****

The isolated queen's pawn constitutes a major weakness, giving White the upper hand, according to the engines.

28.f4!? Ng6 29.Bf3 Bb7 30.Kf2

In the postmortem RH suggested 30.Nb5, when the engines continue 30...a6 31.Nc7 Bc5 32.Kf1!?, the point being 32...Bxe3 33.Nxd5 Bxd5 34.Bxd5+ Kf8 35.g3 leaves White with the bishop-pair on a relatively open board. The engines reckon Black should prefer 32...Ne7, but agree 33.Bd4!? Bxd4 34.exd4 gives White at least a slight edge.

30...Bc5 31.Ne6 Bd6 32.g3 Ne7 33.Nd4!?

This is better than 33.Nc7 Bc8!?

33...Kf7 34.Nb5 Bb8 35.Ba3

How should Black proceed?

*****

*****

*****

*****

*****

35...Ke6?

Several moves seem to hold, including 35...Bc6 and 35...Bc8.

36.Bg4+

There is no satisfactory reply to this.

36...Kf7

Certainly not 36...Nf5?? 37.Nd4+, while after 36...f5?! 37.Nd4+ Kf6 38.Bxe7+ Kxe7 39.Nxf5+ the bishop-pair is nowhere near enough compensation for being a pawn down and with a continuing weakness at d5.

White to play and gain what is probably a winning advantage

*****

*****

*****

*****

*****

37.Nd6+?

RH rejected 37.Bd6!? because of 37...a6 38.Bxb8 axb5 39.Bc7 Nc8, when 40.Bxc8? Bxc8 41.h4 and 42.Bxb6 wins a pawn, but opposite-colour bishops should save Black. However, White has several improvements at move 40, including 40.Bd7!?, when 40...Bc8 41.Bxb5 Bxh3 leaves the number of pawns equal, but 42.a4!? sets in motion White's newly acquired farside majority, and that wins, according to the engines. Note that 37...Ba6? 38.Bxb8 Bxb5 39.Bxa7 gives White a huge advantage.

37...Bxd6 38.Bxd6

Black to play and draw

*****

*****

*****

*****

*****

38...f5?!

The simple 38...Bc8 gets the bishop-pair off the board, and so draws easily. Other moves also probably hold, according to the engines, including 38...g6, 38...Nc6 and 38...Ke8.

39.Bf3 Ke6 40.Be5

This is a good outpost for the bishop, and meanwhile Black's remaining bishop looks a sorry piece.

40...g6 41.Bg7

White has the upper hand after 41.g4!?, according to the engines. They reckon the reply 41...fxg4!? is almost mandatory, but after 42.hxg4 they are unsure how Black can put up the best resistance, one line running 42...Ba6 43.Ke1 Bb5 44.Bg7 Ng8 45.Kd2 Bd7 46.Kd3, with continuing white pressure.

41...h5!?

This puts another black pawn on the same colour complex as Black's bishop, but now White will find creating a breakthrough on the kingside difficult.

42.Be2 Nc8 43.Be5 Nd6!?

Correctly judging that the good-v-bad bishop ending is drawn.

44.Bxd6

White can hardly hope for winning prospects if the knight is allowed to settle on e4.

44...Kxd6 45.b4 Bc6 46.Ke1 a5 47.a3 axb4 48.axb4 b5!?

Blockading the queenside, but Black's task is made trickier by putting yet another black pawn on a light square

*****

*****

*****

*****

*****

49.Kd2 Bd7 50.Kc3 Kc6 51.Kd4 Kd6
Clearly White's king cannot be allowed access to e5 or c5.

52.Bf3 Be6

52...Bc6!? allows 53.g4, when Black may still be OK, but the defence gets more complicated.

53.Bg2 Bf7 54.Bf1 Be8 55.Be2 Bd7 56.Bd1 Be6 57.Bb3 Bf7 58.Kd3 Be6 59.Bc2 Kc6 60.Kd4 Kd6 61.Bd3 Bd7 62.Kc3 Kc6 63.Kd2 Kd6 64.Ke2 Kc6 65.Kf2 Kd6 ½–½

Tournament hall

Weimar Round Four

Ralf Schober (2070) - Spanton (1939)

Bird

1.f4 c5 2.Nf3 d5 3.e3 g6 4.Bb5+!?

This is the top choice of Stockfish17.1 and Dragon1.

4...Bd7

The engines like 4...Nd7!?, although Dragon1 comes to marginally prefer the text.

5.Qe2 Nc6 6.Bxc6!?

This may be a novelty. Only castling is in ChessBase's 2026 Mega database.

6...Bxc6 7.Ne5 Qc7 8.b3 Bg7 9.Bb2 Nh6 10.0-0 0-0

With both sides having castled, how would you assess the position?

*****

*****

*****

*****

*****

White has more kingside space; Black enjoys more space on the central/queenside. White has a well-posted knight at e5, but Black's bishop-pair is a genuine asset as the knight is pinned, and so cannot simply chop on c6. The engines give Black a slight edge.

11.d3 f6?!

But this breaks the pin, and so gives up the bishop-pair - probably too high a price to pay for evicting the knight. The engines suggest 11...Nf5, 11...b5 or 11...Rad8, in each case preserving the bishops.

12.Nxc6 Qxc6 13.Nd2 Rae8

White has a slight edge after 13...e5 14.fxe5 fxe5 15.e4, according to the engines.

14.e4 e5

The engines reckon it is better to advance the e pawn only one square as, after the text, Black gets an isolated king's pawn.

15.fxe5 fxe5 16.Rxf8+ Rxf8 17.exd5 Qxd5

Black has a central isolani, which also means an extra pawn-island, and a bad bishop

*****

*****

*****

*****

*****

18.Re1 Re8 19.Qe4 Qxe4 20.Nxe4 b6 21.Nd6 Re7 22.Nc4 Nf7

All black pieces, arguably including the king, are less active than their white counterparts, but the e pawn is not in imminent danger of being captured, which is why the engines give White no more than a slight edge 

*****

*****

*****

*****

*****

23.a4 Re6 24.Kf1 Kf8 25.Ba3 Ke8 26.Re2 Kd7 27.Bb2 Kc6 28.h3 Kd5

The black king has been fully activated, helping to overprotect e5, while the white king has hardly moved, leading Stockfish17.1 to call the game equal, although Dragon1 still gives White a slight edge

*****

*****

*****

*****

*****

29.Rf2!?

Positions with a central isolani are usually easier to defend without heavy pieces on the board, but RS could presumably find no other way to try to make progress.

29...Rf6 30.Ne3+ Ke6 31.c4!?

The engines are happy with this. It makes d3 backward, but creates a potential outpost at d5.

31...Rxf2+ 32.Kxf2 h5 33.Nd5 Kd6 34.Nc3 Kc6

Not 34...Ke6? 35.Nb5 a6 36.Nc7+, or 35...a5 36.Nc7+ Kd7 37.Kc7 Ne7+, with White in each case winning at least a pawn.

35.Ne4 Kd7 36.Kf3 Nd8!?

This and 36...a5!? are the engines' top choices, but 36...Nd6!? may also be playable, although Dragon1 is doubtful.

37.Bc3 Ne6 38.g3 Kc6 39.h4 Kd7 40.Ke3 Bh6+ 41.Kf2 Bg7 42.Kf3

The white king's triangulation has brought about the same position as after 39...Kd7, but this time with Black, rather than White, to move. However, that makes no significant difference.

42...Kc6 43.Nf2

How should Black proceed?

*****

*****

*****

*****

*****

43...Kd6

Not 43...e4+? 44.Nxe4 Bxc3 45.Nxc3 Nd4+ as 46.Kf4 Nxb3 47.Kg5 wins easily for White.

44.Ke4 Nf8 45.b4!? Nd7!?

Dragon1 strongly dislikes this. Stockfish17.1 is not so sure, but the engines agree that better is 45...cxb4!? 46.Bxb4 Kc6! 47.Bxf8 Bxf8 as 48.Kxe5 Bd6+ 49.Kf6 Bxg3 50.Ne4 Bxh4 51.Kxg6 is completely equal. But note that the natural-looking 46...Ke6? loses to 47.Nh3!, eg 47...Nh7 (47...Nd7? 48.Ng5+ Kf6 49.Kd5) 48.Ng5+ Nxg5 49.hxg5 Bh8 50.Bf8 Kd7 51.a5! bxa5 52.Kd5 a4 53.c5! Kc7!? 54.c6!? Kc8 55.Ba3 Kd8 56.Ke6 Kc7 57.Kf7 Kxc6 58.Kxg6 Kd5 59.Bb2 Kc5 60.Kh7 e4 61.Bxh8 exd3 62.Bc3 a3 63.g6 a2 64.g7 d2 65.Bxd2 a1=Q, after which, despite queening first, Black has no perpetual, and cannot pick up the loose bishop, and so is lost - a long line, with several hard-to-spot moves, but the engines show much of it is more-or-less forcing.

46.b5 Nf6+ 47.Ke3 Bh6+ 48.Kf3 Ke6 49.a5

How should Black continue?

*****

*****

*****

*****

*****

49...bxa5!?

Possibly best is 49...Bf8, despite this allowing 50.a6, after which Black has to watch out for White sacrificing on c5 or b6 to get the a pawn home.

50.Bxa5

How big is White's advantage?

*****

*****

*****

*****

*****

Dragon1 gives White the upper hand, but it reckons both of Black's most plausible moves, 50...Bf8 and 50...Kd7, leave White exactly 0.75 ahead, which in endings can be a sign that the position should be drawn with best play. Stockfish17.1 gives White a slight edge, which tends to confirm that the game is holdable.

50...Bf8 51.Ne4!?

How should Black respond?

*****

*****

*****

*****

*****

51...Nxe4!

This is best, according to Stockfish17.1, despite giving White a good-v-bad bishop ending. Dragon1 reckons 51...Bd6 is equally good, but Stockfish17.1 disagrees.

52.Kxe4

The key to the ending is whether Black can avoid falling into a zugzwang that either loses a pawn or lets the white king in at d5 

*****

*****

*****

*****

*****

52...Bd6 53.Bd8 Bb8 54.Bg5 Bd6 55.Bh6 Be7 56.Bd2 Bd8 57.Be3 Be7

An only-move as 57...Bb6? loses to 58.Bf2 Kd6 59.g4! hxg4 60.Bg3 Ke6 61.Bxe5, eg 61...Bd8 62.Bg3 Bf6 63.Kf4 be7 64.Kxg4, after which Black is a pawn down and still has weaknesses.

58.Bh6 Bd6 59.Bg7 Bc7 60.Bf8 Bd6

Another only-move as 60...Bb6? loses to triangulation, eg  61.Ke3 Kf5 62.Bd6 Ke6 63.Bb8 Kf5 64.Kf3 Ke6 65.Ke4 and 66.Bxe5(+), and if 61...Kf7, then 62.Bd6 Ke6 63.Bb8 wins in the same way.

61.Bh6

Or 61.Bxd6 Kxd6, when the pawn ending is also drawn.

61...Bc7 62.Be3 Bd6 63.Bd2 Bc7 64.Bc3 Bd6 65.Ba5 Bb8 66.Bd8 Bd6 67.Ba5 ½–½

Benidorm Bali Round Six

FACED a Singaporean junior (born 2007).

Rahul Lakshminarasimhan (1626) - Spanton (1896)

London System

1.d4 d5 2.Nf3 e6 3.Bf4 Bd6 4.Bg3 Nf6 5.e3 Ne4!?

Stockfish17.1 and Dragon1 prefer the mainline 5...0-0, or 5...c5.

How should White respond?

*****

*****

*****

*****

*****

6.Bxd6!?

White usually plays 6.Nbd2 or 6.Bd3, eg 6.Nbd2 Nxg3 7.hxg3, when the half-open h file and 'promotion' of the h pawn to a more-central g pawn are roughly balanced by Black having the bishop-pair and more flexible kingside pawn-structure. The engines give White the better part of equality.

White has also tried 6.Bh4, when Black has 6...g5!?, as well as the more conventional 6...Be7.

6...cxd6!? 7.c4 Qa5+ 8.Nbd2

The engines come to marginally prefer 8.Nfd2!?

8...dxc4 9.Bxc4 d5 10.Bd3!?

The engines agree this apparent-novelty is a slight improvement on the known 10.Be2.

10...Nc6 11.0-0 Nxd2 12.Nxd2 0-0

Now both sides have castled, how would you assess the position?

*****

*****

*****

*****

*****

White has much the better bishop, which the engines agree gives White a slight edge, although Stockfish17.1 is less sure than Dragon1.

13.f4

This looks dangerous, but the engines prefer immediate queenside play with 13.Nb3.

13...Bd7 14.Nb3!?

A move such as 14.Rf3, 14.Nf3, or even 14.Qh5!?, intending to meet 14...g6 with 15.Qe2, would seem more consistent, but the engines reckon White 's kingside play is not dangerous, and they prefer the text.

14...Qc7!?

This invites a pin, which is why the engines suggest 14...Qd8.

15.Rc1 Rac8

This is Dragon1's top choice; Stockfish17.1 prefers 15...Rfc8!?

16.Nc5 b6?!

This is Dragon1's top choice, at least for a while, but Stockfish17.1 much prefers 16...g6 or 16...Ne7, and Dragon1 comes to agree, presumably because the text is too loosening.

17.Nxd7 Qxd7 18.Qa4

How should Black proceed?

*****

*****

*****

*****

*****

18...Rc7?
Black had to play 18...Qd6 or 18...Nb8, although neither is pleasant, and it is unclear which is better, eg 18...Qd6 19.Ba6 Rc7 20.Qa3! Qd8! (20...Qd7 21.Qc3 is much worse, according to the engines) 21.Qc3 Re8 22.Bb5 Qb8! 23.Rc2 (23.Bxc6? Rec8 equalises) Rec8 24.Rfc1 Qb7 25.f5!? exf5 26.e4!, with the upper hand for White, according to the engines, or 18...Nb8 19.Qxd7 Nxd7 20.Ba6 Ra8 21.Rc7, with pressure, but only a slight edge for White, according to Stockfish17.1, although Dragon1 again awards White the upper hand. Both lines are sharp and there are alternatives along the way.

19.Bb5 Rfc8 20.Ba6 Rcd8 21.Bb5 Rdc8

White to play and win

*****

*****

*****

*****

*****

22.Qa6?

Doubling on the c file, starting with 22.Rc3, gives a simple win, eg 22...Qd6 23.Ba6 Rd8 24.Rfc1 etc. The engines suggest 23...b5!?, but 24.Qxb5 Rb8 25.Qc5 Qd8 26.Rfc1 wins, eg 26...Rxb2 27.Qxc6! (27.Bb5 is also good enough) Rxc6 28.Rxc6, after which the black queen falls.

22...Qd8

Also drawing is 22...Qe7!?, as long as after 23.Bxc6 Black plays 23...Qd8! and not 23...Rxc6? 24.Qxc8+! etc.

23.Bxc6 Rxc6 24.Rxc6 Rxc6 25.Qxa7

White has won a pawn, but Black's activity on the open file forces a draw

*****

*****

*****

*****

*****

25...Rc2 26.Qa3 g6 27.Qb3 Qc7 28.Rf2

There is nothing better.

The game finished:

28...Rc1+ 29.Rf1 Rc2 30.Rf2 Rc1+ 31.Rf1 Rc2 ½–½

Bodensee Senioren Round Six

FACED an Austrian Fide master.

Frank Belke (2200) - Spanton (1902)

QGD Semi-Tarrasch

1.d4 d5 2.c4 e6 3.Nc3 Nf6 4.Nf3 c5 5.cxd5 cxd4!? 6.Qxd4 Nxd5!?

The mainline in ChessBase's 2024 Mega database runs 6...exd5 7.e4!? dxe4!? 8.Qxd8+ Kxd8 9.Ng5 Be6 10.Nxe6 fxe6 11.Bg5, when Stockfish16 reckons the position is equal, but Komodo14.1 fluctuates between equality and a slight edge for White.

7.Nxd5

This scores an excellent 63% in Mega24, but the more popular 7.e4 scores 67%.

7...Qxd5 8.Qxd5 exd5

How would you assess this IQP position?

*****

*****

*****

*****

*****

With queens and a pair of knights exchanged, Black is highly unlikely to generate attacking chances, which, as I understand it, means the IQP is more of a weakness than a strength. The engines give White a slight edge.
9.e3!? Nc6 10.Bb5!? a6 11.Ba4 Bb4+!?

The engines are OK with this, but as a general rule the player with an IQP should not be seeking to swop off his better bishop.

12.Bd2 Bxd2+ 13.Kxd2 b5 14.Bb3 Be6 15.Nd4!? Nxd4 16.exd4 Kd7

Now both players have isolated queen's pawns, the major (effectively only) difference between the sides is White has a good bishop and Black a bad one

*****

*****

*****

*****

*****

17.a4 Kd6 18.Rhc1 Rhc8 19.Rxc8 Rxc8 20.axb5 axb5 21.Ra6+ Kc7!

As will become soon become clear, this is much better than the 'obvious' 21...Rc6?, allowing an exchange of rooks.

22.f4 Kb7 23.Ra1 Kc6?

The game is equal after, for example, 23...g6, according to the engines.

24.Rc1+ Kd6 25.Rxc8 Bxc8 26.Kc3 Bf5 27.Kb4 Kc6

The rook-and-bishop ending was drawn, but this bishop ending is lost, according to Stockfish16, although Komodo14.1 at first gives White 'only' the upper hand

*****

*****

*****

*****

*****

28.Ka5 Bd3 29.Bd1 g6

It seems there is no defence. Black can hold the b pawn for now, as long as the black bishop stays on the c4-f1 diagonal, but White can force concessions on the kingside, after which it is a matter of time before the black position collapses.

30.g3 Bf1 31.g4 Bd3 32.g5 Bc4 33.h4 Bd3 34.Bg4 Kc7

If 34...Bc4 then 35.f5 is strong.

35.h5 Kc6 36.h6 Kc7 37.Bf3 Kc6 38.Bg2 Be2 39.Bh3 Bd3 40.Bg4 Kc7 41.Kb4 Kc6 42.Bh3 Kd6 43.Kc3 Bc4

If 43...Bb1 then 44.Bf1.

44.f5 Be2 45.fxg6 fxg6 46.Bc8 Bf3 47.Bf5!? Be4

If 47...gxf5 then 48.g6 etc.

48.Bg4 Bg2 49.Kb4 1-0

The position is hopeless: 49...Bf1 loses to 50.Bf5!, and 49...Kc6 loses to 50.Be6 followed by Bg8 etc.

Mariánské Lázně Round Four

PLAYED a Russian this morning as part of the double-round day.

Alexander A Yakimenko (2055) - Spanton (1825)

Mariánské Lázně Seniors 50+

Spanish Berlin

1.e4 e5 2.Nf3 Nc6 3.Bb5 Nf6 4.0-0 Nxe4 5.Re1

This is less than half as popular as 5.d4 but scores the same, 54%, in ChessBase's 2022 Mega database.

5...Nd6 6.Nxe5 Nxe5 7.Rxe5+ Be7 8.Bf1!?

This is the main continuation, probably because other retreats of the bishop, although playable, are problematic, eg 8.Ba4 0-0 9.d4 Nc4 or 8.Bd3!? 0-0 9.Nc3 c6!? 10.b3 Ne8 11.Bb2 d5, while 8.Be2, which cuts off the retreat of the rook, does not appear in Mega22.

8...0-0 9.d4 Bf6 10.Re1 Nf5

Zukertort played 10...Re8 and went on to beat Steinitz in game four of their 1886 world championship match.

11.c3

Carlsen has played the interesting unbalancing move 11.d5!?

11...d5 12.Bf4 c6 13.Nd2 Nd6!?

The main move in Mega22 is 13...Nh4.

14.Qf3

Possible is 14.Bd3 as, unlike after 13...Nh4, the reply 14...Bf5?? loses material to 15.Bxd6.

14...Bg5 15.a4 Qf6 16.Bxg5 Qxg5 17.Qe3 Qxe3 18.fxe3!?

An interesting way to keep life in the game, and one that is considered by Stockfish14.1 and Komodo12.1.1 to be at least as strong as 18.Rxe3, not that they reckon either move gives White an advantage.

18...Bf5 19.c4 Nxc4 20.Nxc4 dxc4 21.Bxc4 Rfe8

How would you assess this position?

*****

*****

*****

*****

*****

Black has fewer pawn-islands, pressure against the white central pawns and a lovely square at e4 for a piece; White has more space. Komodo12.1.1 reckons the position is dead-equal but Stockfish14.1 gives White an edge worth about a fifth of a pawn.

22.b4 a6 23.Kf2 Kf8 24.Rac1 Be4 25.Rc3 Red8 26.Rec1 Rd6 27.Bf1!? Rf6+ 28.Kg1 Ke7 29.Rc5 Rd8 30.Ra1 Rd5 31.Ra2 Rdf5 32.Rxf5 Rxf5 33.Rf2 Rxf2 34.Kxf2

How would you assess this bishop ending?

*****

*****

*****

*****

*****

White has an extra pawn-island but has the better bishop (the b7 pawn is weak, even though it may not appear so at present, because it is a target for the white bishop). The engines reckon the position is dead-equal.

34...Bc2

The engines like 34...b5!?, which solves the problem of the weak b7 pawn but creates new weaknesses at c6 and a6 after 35.a5.

35.a5 Be4 36.h4 h5?!

Ceding space is almost certainly a lesser evil than putting another pawn on a light square.

37.Be2 g6

Another pawn has to go on a light square as 37...Bg6? leaves Black very inflexible.

38.Bf3

How should Black proceed?

*****

*****

*****

*****

*****

38...Bd5?

This turns a difficult position into what is almost certainly a lost one.

Correct is 38...Bxf3, when I was worried about 39.gxf3. The engines reckon only 39...f6! equalises, eg 40.e4 (40.f4 f5) g5! (another only-move, according to the engines), after which Black will get a protected passed pawn if White ever plays f4, while 41.d5 can be met by 41...Ke6. The alternative recapture, 39.Kxf3, again needs an only-move, this time 39...f5!, eg 40.Kf4 Kf6 41.g3 Ke6 42.Kg5 Kf7.

39.e4 Be6 40.Ke3 Kd6 41.Kf4 Bc4 42.Kg5 Be6?!

42...Ke7 probably puts up tougher resistance, but White wins after something like 43.Kh6 Kf6 44.g4 hxg4 45.Bxg4 Be6 46.Bf3 Bb3 47.Kh7 Be6 48.Kg8 Ke7 49.d5 cxd5 50.exd5 Bc8 51.d6+! Kf6 52.Bd5 Ke5 53.Bxb7! Be6 54.Kg7 Kxd6 55.Bxa6 etc.

43.Kf6 Bc4 44.g3 Bb3 45.g4 hxg4 46.Bxg4 Ba2

Or 46...Kc7 47.Ke7 Bc4 48.Bd7 etc.

47.e5+ Kd5 48.e6 1-0

One of Mariánské Lázně's finer hotels

Chess Evolution: The French Defence

HERE I want to give an overview of an opening - namely the French Defence - by looking at how it developed over the centuries.

The French, as is quite well known, is named after a correspondence game between chess clubs in London and Paris.

That was in 1834, when national postal systems were in their infancy, and international mail even more so.

But by then 1...e6 in reply to 1.e4 had been known for more than 200 years, first being published, it seems, in manuscripts of the Italian master Gioachino Greco.

Here is the earliest game in ChessBase's 2020 Mega database

Greco - ??*

Rome (?) 1620

French Advance

1.e4 e6 2.d4 d5 3.e5

I find it interesting that in the first known French Defence, whether a genuine game or not, White should 'play' the Advance Variation (my guess would have been for 3.exd5). The move e5, either here or later, is thematic in that White seeks to take advantage of Black ceding space in the centre and thus on the kingside. It is also quite committal: White is more-or-less obliged to follow up with a kingside attack as otherwise the move e5 will likely prove a tempo expended for no good reason.

3...c5

Attacking the base of White's pawn-chain is second-nature to modern players, but Joseph Blackburne experimented in 1885 with attacking the head of the pawn-chain with 3...f6!?, drawing with Max Weiss and George Mackenzie.

4.c3 cxd4!?

This is generally regarded today as premature, not least because it allows White to clear the c3 square for the white queen's knight. However it has appeared in computer games and was played in 1981 by Bulgarian international master, and future grandmaster, Krum Georgiev (but the game was agreed drawn in 10 moves, so perhaps the players were peacefully inclined from the start).

5.cxd4 Bb4+

Patzer sees check, patzer plays check (or, probably in this case, is made to play a check by the game's composer). Strong players who have reached the position after 5.cxd4 by transposition, eg Francisco Vallejo Pons and Ruifeng Li, have generally preferred 5...Nc6. In their cases the games - one by Pons, two by Li - went 1.e4 c5 2.c3 e6 3.d4 cxd4 4.cxd4 d5 5.e5. All three games were won by Black.

6.Nc3 Bxc3+!?

Black gives up his good bishop to saddle White with a backward pawn on a half-open file. But the pawn on c3 will support White's centre and should not be difficult to defend.

7.bxc3 Nc6 8.Bd3

8.Qg4 is at least as strong.

8...Nge7 9.f4

9.Ne2 was played by a 2265 in a Hungarian game in 1994. The text is often played by White in this type of pawn-structure.

9...Nf5 10.Nf3

How should Black proceed?

*****

*****

*****

*****

*****

10...0-0?

Black's play has been reasonable up to this point. The analysis engines Stockfish11 and Komodo11.01 reckon White only has a slight edge after 10...Na5, targeting the c4 square. The text, although plausible at first glance, is an example of "castling into it."

11.g4 Nh4 12.0-0 Nxf3+

The best alternative the engines can come up with is 12...f5 13.exf6 Nxf3+ 14.Qxf3 gxf6, but reckon White has a winning attack after 15.Qh3.

13.Qxf3 Bd7?!

Black had to gain some breathing room on the kingside, eg 13...f5 14.exf6 Qxf6, but White keeps a large advantage.

14.Qh3 g6 15.f5 exf5 16.gxf5 gxf5 17.Rxf5!?

Objectively not best, according to the engines, but it leads to an instructive finish.

17...Bxf5

The engines give best play as 17...Kh8 18.Kh1 Ne7 19.Bg5 f6 20.Bxf6+ Rxf6 21.exf6 Bxf5 22.Bxf5 Nxf5 23.Qxf5, when White is only a pawn up but is completely winning.

18.Bxf5 1-0

*Many of Greco's recorded 'games' are thought to have been composed by him for instructional purposes rather than being genuine over-the-board encounters, and that is probably the case with this example.

(To be continued)

Morphing The French III

HAD a third chance to try playing against the French Defence à la Paul Morphy in this morning's top-board game in the fourth round of the 23rd 4NCL U2000/175.
Spanton (1927/168) - Michael Dunkley (1732/152)
French Exchange
1.e4 e6 2.d4 d5 3.exd5 exd5 4.Nf3 Bd6 5.Bd3 Nc6
This has been tried by aggressive players such as Morozevich and So.
6.0-0 Nge7 7.Re1 0-0 8.c3
Not 8.Bxh7+?? Kxh7 9.Ng5 Kg8 10.Qh5 Bf5.
8...Bf5 9.Nh4!?
This may be a novelty. The idea is to oblige Black, if he wants to exchange off White's 'good' bishop (based on the central pawn-structure), to do so in a way that develops the white queen to a good central square. Another way of trying to accomplish this is 9.Bg5, which is the most popular move in the position, but I rejected it because of 9...f6, when it is not clear White has achieved much.
9...Bxd3 10.Qxd3 Qd7 11.Nd2 Qg4?!
This looks aggressive, but the queen will be driven back with loss of time.
12.Ndf3 h6 13.h3 Qd7
Black loses his queen after 13...Qh5?? g4.
14.Ne5

Black to decide what to do about the knight on e5

*****
*****
*****
*****
*****
14...Bxe5!
The problem with 14...Nxe5 15.dxe5 Bc5 16.Be3 is that White's 4-3 kingside majority could form the basis of a dangerous attack.
15.dxe5 g5!?
Clearly this weakens Black's king, but MD has a specific idea in mind.
16.Nf3 Qf5 17.Qxf5?!
I thought 17.Qb5 left my kingside weak, but Black has no real attack.
17...Nxf5 18.g4 Ng7 19.Nd4?! Nxd4 20.cxd4 Ne6 21.Be3
Now White has two central pawns on the same-coloured squares as the bishop, which is severely restricted, and Black has a fine blockading square for his remaining knight. However, the bishop is not completely toothless as it has pressure against Black's g and h pawns.
21...a5
I felt 21...f6 was more promising for Black, but the text is fine.
22.Kg2 Kg7 23.h4 b6
Preparing ...c5.
24.Rac1?
Playing to stop ...c5, but it does not! Better was 24.Rad1, which prevents ...c5 because of the latent threat to Black's d pawn.
24...a4
Black could have played 24...c5 as 25.dxc5 is met by 25...d4 26.Bd2 bxc5, with a good position for Black.
25.Rh1 Rg8 26.hxg5 hxg5 27.Kg3 Ra5 28.Rhd1 Rb5 29.Rd2 Rh8 30.f4!?
Fighting for the initiative as my position has become very passive.
30...gxf4 31.Bxf4 Rb4 32.Be3 Rb5
This was another good time to play ...c5.
33.Rf1 Rh7
Black has a small edge after 34...Kf8, according to Stockfish10 and Komodo10.
34.Rf6 Rb4?
Again ...Kf8 was best, according to the engines.
35.g5!?
Putting another pawn on a dark square, but menacing Black's kingside. However, possibly even stronger was 35.Rfd2. I missed that 35...Kf8 can be well-met by 36.Bh6, which threatens to win with 37.Rxe6! fxe6 38.Rf8#.
35...Nf8?!
35...Kg8 is probably better, although White has the advantage after 36.Kg4.
36.Rc2
Even stronger is 36.Rdf2 Kg8 37.g6! I missed that 37...Nxg6? fails to 38.e6. That leaves 37...Rg7 38.Rxf7 Rxg6+ 39.Kh4 Ne6 40.Re7 with much the better game for White, according to the engines. One likely line, in what is a sharp position, runs 40...Nxd4 41.Re8+ Kg7 42.Kh5 Re6 43.Rxe6 Nxe6 44.Rf6 Nc5 45.Bh6+ Kg8 46.Rf8+ Kh7 47.Rf7+ Kh8 48.Re7 Rxb2 (not the best, but illustrative of Black's difficulties) 49.Kg6 Rg2+ 50.Bg5 Rxg5+ 51.Kxg5, and White wins the rook-v-knight ending despite being two pawns down.
36...c5! 37.dxc5 d4?
The engines give 37...Re4 as equal.
38.Bd2 d3 39.Rc3?
Stronger is 39.Rc1, threatening to win the exchange, as 39...Rxb2 is met by 40.c6.
39...Rxb2?
Black should have tried 39...bxc5 40.Rxd3 Rxb2, when 41.e6! Nxe6 (not 41...fxe6? 42.Rd8) 42.Bc3 Rc2 43.Rf2+ Rxc3 44.Rxc3 is certainly good for White, but the game is not over.
40.c6 Rxd2
Not much better is 40...Rh8 41.Rxd3 (41.c7 is also strong) Rc2 42.Rdf3 etc.
41.c7 Rdh2 42.c8=Q Ne6 43.Rxf7+ Kxf7 44.g6+ Kxg6 45.Qxe6+ Kg5 1-0

Pim O'Clock

THE annual Pimlico Summer congress, held over five consecutive Thursdays, began last night with a full complement of 64 entries.
It is split into three sections - open, U150 and U120 - with a time limit of 75 minutes plus a 15-second increment.
Combined with a 6.30pm start, this means the vast majority of games are over before 9pm.
Paul Carlucci (154) - Spanton (171)
Colle-Zukertort
1.d4 d5 2.Nf3 Nc6
The fourth-most popular move in ChessBase's 2019 Mega database, and the worst-scoring percentage-wise among the top 10 replies. Nevertheless it has been tried by strong players including Carlsen and Rapport.
3.e3
The Colle route. More popular these days is the London System with 3.Bf4, while 3.c4 transposes into a mainline of the Chigorin.
3...Bg4 4.Be2 Nf6
Rapport and Morozevich have preferred 4...e6, which was played all the way back in 1882 by Winawer in a win over Blackburne. I guess the two moves will normally transpose.
5.0-0 e6 6.b3 Bd6
Morozevich chose 6...Ne7!? in a 1995 rapidplay win against Malaniuk (both players had ratings just above 2600). There are no other examples of the move in Mega19, and analysis engines Stockfish10 and Komodo9 disapprove, but it is dangerous to rely on engines' positional judgments in the opening.
7.Bb2 0-0 8.Ne5 Bf5!?
There is a lot to be said for 8...Bxe2. It exchanges White's good bishop, but also develops White's queen to a decent square.
 9.c4
An obvious alternative is 9.Nxc6 bxc6, but Black will be able to dissolve his doubled pawns. The engines continue 10.Nd2 (Stockfish10) or 10.c4 (Komodo9), with Black in each case replying 10...c5.
9...Nd7?!
The engines are not keen on this, preferring 9...Ne7 (Stockfish10) or 9...dxc4 (Komodo9). My idea, after a move such as 10.f4, was to evict White's knight with 10...f6, but then 11.Nxc6 bxc6 leaves Black looking vulnerable to a kingside pawn-storm.
10.Nxd7!?
This surprised me. The engines reckon it is enough for a slight edge, but they prefer 10.Nxc6 or 10.f4. Note that after 10.Nxc6 bxc6 White can prevent undoubling by 11.c5!?
10...Qxd7 11.Bd3?!
An old rule-of-thumb has it that when you have a bad bishop - in this case White's dark-square bishop, which is shut in by White's central pawns - you should not trade your good bishop. True, the dark-square bishop is doing a useful job in helping control the e5 square, and the coming exchange develops White's queen to a good square, but I do not believe these factors compensate for White's vulnerability on the light squares.
11...Bxd3 12.Qxd3 f5 13.a3

Black to make his 13th move

The stage is set for a battle of rival attacks - White on the queenside, Black on the kingside. Regardless of the merits of the two attacks, there will always be the realisation in both players' minds that White is playing for a long-term positional advantage while Black is playing for mate.
13...Rf6
I rejected 13...f4 because 14.Nd2 prevents an effective ...f3 as Black is not in a position to sac the exchange with 14...f3? 15.Nxf3 Rxf3?
14.c5!?
This comes to be Stockfish10's choice, at least for a while, but Komodo9 is unimpressed. As usual, pushing on like this is double-edged - it is certainly consistent with White's plan of queenside expansion, and it drives Black's bishop to an inferior square, but it means Black's kingside attack is less likely to be inconvenienced by central counterplay.
14...Be7 15.b4 Raf8
The engines do not like this, but I felt committed to playing on the kingside, and I wanted to be able to meet b5 with ...Nd8 without obstructing the a8 rook's route to joining the attack.
16.f4 h6?!
This may be too slow. I wanted to play 16...g5 but thought White was doing OK after 17.fxg5 Rg6 18.h4? h6 19.gxh6 Rxh6? 20.g3, which may be true. However, 19...Bxh4 is a major improvement.
17.Bc3?!
Komodo9 is happy with this move - for a while it is the engine's second choice - but if the idea is to stop b5 being met with ...Na5, then the idea is misplaced as 17.b5 Na5? 18.Nd2 will leave Black's queenside in disarray after a later Bc3 (indeed Stockfish10 reckons Black's best after 18.Nd2 is to sac a pawn with 18...Nc4 etc). However, having said all that, 17.Bc3?! does make the bishop better-placed for defensive duties on the kingside.
17...g5 18.fxg5?!
It was probably better to leave Black to open lines with an eventual ...gxf4 as g2 will be easier to defend than h2. Black could also push on with ...g4, but then will find it very difficult to open any lines on the kingside.
18...hxg5 19.b5 Nd8 20.Nd2 Rh6 21.Nf3 Bf6

White faces a tricky defence

*****
*****
*****
*****
*****
22.Ra2?
Best seems to have been 22.Be1. The point is that if play proceeds as in the game with 22...g4 23.Ne5 Bxe5 24.dxe5 Qh7, White defends h2 with 25.Bg3. The text also envisages a defence of h2, but at great positional cost.
22...g4 23.Ne5?!
Komodo9 prefers 23.Ne1 (but not 23.Nd2? Qh7) although Stockfish10 is unsure; either way, Black is well on top.
23...Bxe5 24.dxe5 Qh7 25.g3 Nf7
White's light-square weaknesses (see note to White's 11th move) mean he is lost despite the material balance.
26.e4!?
Objectively this cannot be good, but trying for counterplay may be White's best practical choice.
26...dxe4 27.Qd7 Ng5
Also tempting was 27...Rd8!? but there is no need to complicate.
28.Qxh7+ Rxh7 29.Rd1 Nf3+ 30.Kh1 f4
Simpler is 30...Kf7 (or 30...Kg7), meeting 31.Rd7+ with 31...Kg6, but the text also wins.
31.Be1
This is Stockfish10's choice for a while, but the engines come to agree that best was 31.gxf4, when Black has a pleasant choice between 31...Rxf4, 31...Rh3, 31...g3 and several other moves.
31...Nxe1 32.Rxe1 f3 33.Kg1 Rd8 34.Rf2 Rd4 35.Kf1 Rh5 (0-1, 45 moves)

Guernsey Round Four

FOR the second-day running I got down to an ending of bishops of the same colour.

White to make his 42nd move in Spanton (1923) - John Cummins (1748), but is the position winning for White, better for White but not necessarily winning, roughly equal, better for Black but not necessarily winning, or winning for Black?

****
****
****
****
****
****
****
****
This is about as simple-a-win for White as is possible in a position with equal material.
The key points are: Black's bishop is bad, ie it is severely restricted by its own pawns while being unable to attack White's pawns; White threatens at the right moment to create a passed pawn near a queening square on the kingside; White's king is nearer the centre and, by extension, nearer Black's vulnerable queenside.
42.a4
It makes sense to keep Black's queenside pawns on dark squares, partly to stop the black pawns rolling forward but also so White's king can enter via the light squares.
42...Bf6
Other moves were possible but none was any good - Black is totally lost.
43.Kd3 Kd7 44.Kc4 Ke6
44...Kc6 keeps White's king in its own half of the board, but then comes 45.g6 hxg6 46.Bxh6 etc.
45.Kb5 Bd8 46.Kc6 Be7
Or 46...Kf6 47.Kd7 Be7 48.Kc7 (there are other wins, but this is the simplest).
47.Kxb6 Bd8+ 48.Kxc5 1-0
I knew from the diagram that the position was a very easy win for White. If you were not so sure, you would probably benefit from reviewing any literature you have that covers good and bad bishops.