Lessons From Cap Negret IX

MY round-nine game, where I had black against Alberto Pastor Alonso de Prada (2000), reached the following position on move seven of a Réti.

White has just castled

*****

*****

*****

*****

*****

Usually in such Réti positions, White plays an early d3. I decided to punish his omission of this move by playing ...d3 myself.

But 7...d3? turns a slightly advantageous position, according to Stockfish17.1 and Dragon1, into a lost one (Stockfish17.1), or at least one in which White has the upper hand (Dragon1).

LESSON: greed in grabbing a poisoned pawn in the opening is a well-known phenomenon, but it is also possible to be greedy in grabbing space.

Lessons From Cap Negret VIII

I HAVE been asked a couple of questions about the ending in my round-eight game, where I had white against fellow Battersea Chess Club member Paul Stokes (1807).

The first question arises in the following position, where I have just played 65.Rc7-c5.

Black replied 65...Bf3?, but in my blog notes I pointed out that Stockfish17.1 and Dragon1 reckon both 65...Kg7 and 65...Kh7 hold

*****

*****

*****

*****

*****

First, let us see what happened in the game.

I played 66.Bd7!, to which Black gave a check with 66...Rb1+, only to resign after I continued 67.Kf2.

The point is that after 67...Be4 68.Bd5+, Black is losing a second pawn, giving White two protected passers.

I was asked, why does 65...Kg7 make such a difference, particularly as White is poised to win a second pawn anyway, eg with 66.Bc6? The point is that, with the black bishop on a secure square, 66...Rb1+ now draws, one line running 67.Kf2 Rb2+ 68.Kg3 Rb3 69.Bxd5 Rxe3+ 70.Kf2 Rf3+ 71.Ke2 Bxd5 72.Rxd5 Rf4 etc.

The second question concerns this position, after 66.Bd7!

*****

*****

*****

*****

*****

In my notes I covered what happens if, instead of checking with the rook by playing 66...Rb1+, Black secures the bishop with 66...Be4. The engines reckon 66...Be4 is probably Black's best try, but 67.Bc6 wins, eg after 67...Rb1+ 68.Kf2 Rb2+ 69.Kg3 Rb3 the same position is reached as in the previous note, except the black king is on g8 rather than g7. That makes all the difference as capturing on d5 now comes with check, and after 70.Bxd5+ Bxd5 71.Rxd5 Rxe3+ 72.Kf2 Black does not have ...Rf3+, and the rook-and-pawn ending is an easy win for White, according to the engines.

I was asked, doesn't 66...Rb2, rather than the checking 66...Rb1+, draw? After 67.Be6+ Kf8 68.Bxd5, this seems correct, as both 68...Bxg4, and 68...Re2+ followed by 69...Bxg4, give excellent drawing chances. However, the engines point out the strength of 68.Rc7! (other moves also apparently win), when Black can no longer capture on g4. Black could try 68...Re2+ 69.Kf1 Rxe3, but 70.h6 (even stronger than the also-winning 70.Rf7+) is decisive, eg 70...Be4 71.Rc8+ Ke7 72.Bf5 Bxf5 73.gxf5 Rh3 74.h7 Rxh7 75.Rc7+ etc.

LESSON: engine analysis, especially when options are relatively few, is nearly always correct, but a lot can be learnt by delving into what lies behind the engines' verdicts.

Lessons From Cap Negret VII

MY round-seven game, where I had black against Luxembourg-based Geoffrey Stern (1970), reached the following position after White captured on e4.

White has a slight lead in development, and pressure against Black's queenside, but Stockfish17.1 and Dragon1 reckon the game is equal

*****

*****

*****

*****

*****

*****

I went for the 'safety-first' 11...Qxd1?!, which, as I pointed out in my blog notes, has been played by a 2400+ player (he later became a grandmaster), but is probably a mistake.

The point is that after 12.Rfxd1, White's lead in development has grown, and he still has pressure against the Black queenside. The resulting position is deeply unpleasant for Black, who has lost all seven games in ChessBase's 2026 Mega database that saw ...Qxd1.

Black should probably prefer 11...Qf6, simultaneously attacking f4 and b2, or 11...f6!?, preparing ...e5.

I rejected 11...Qf6 for fear that 12.Bg5!? is dangerous, forcing, as it does, 12...Qxb2. White certainly gets decent compensation for a pawn, which is why 11...f6!? might be a better practical choice for the nervous defender.

LESSON: exchanges that increase an opponent's lead in development are rarely really a 'safety-first' option.

Lessons From Cap Negret VI

MY round-six game, where I had White against Josep Fernández Pérez (1801), featured a common pseudo-sacrifice that is perhaps not as well-known as it should be.

The relevant position arose after Black played 56...Ka6-a7.

White is a pawn up, but the extra pawn is on a rook's file, and there is no obvious way to safely advance it 

*****

*****

*****

*****

*****

57.Nxc6+!

This takes advantage of the fact that the b5 knight, although apparently occupying a strong outpost, is only supported by a pawn, which in turn is only supported by a piece that cannot help defend b5.

After the forced 57...Nxc6 58.Kxb5, I had won a second pawn, and Black resigned.

LESSON: knights, and to a lesser extent rooks, are vulnerable to this pseudo-sacrifice.

Lessons From Cap Negret V

MY round-five game, where I had black against Belgian Fide master Johan Goormachtigh (2188), reached the following position after White played 12.Qb3-d1.

Black has a slight lead in development, but has an IQP that looks vulnerable

*****

*****

*****

*****

*****

*****

*****

What I failed to appreciate is that, although my position is quite active, there is something of a logjam on the queenside, where my bishop has no retreat squares, and my queen and knight could be vulnerable to a pawn-fork on b4.

The engines like 12...Bxc3!?, giving up the bishop-pair but inflicting a weak pawn on White, while at the same time relieving the queenside logjam.

However, I was loath to give up my dark-square bishop, and instead defended d5 with 12...Rfd8?!, which is a developing move, but could have been met by 13.Nb5.

The game saw 13.Nd4, which might not be so strong, or at least would not be if I had taken a second opportunity to play ...Bxc3.

Instead, oblivious to the danger, I played 13...Bg6?, after which White took his second chance, and played 14.Ncb5, after which I made another mistake with 14...Ne6?, when the simple 15.Nxe6 fxe6 16.Bc7 gave White a large advantage.

Even with the superior 14...Na6, Black is trouble, eg 15.a3 Be7 16.b4 Qb6 17.Rc1, with what the engines reckon is a positionally won game for White.

LESSON: pieces need breathing room - when one is short of squares, particularly retreat squares, alarm bells should ring.

Lessons From Cap Negret IV

MY round-four game, where I had white against Luis Vicente Martín Martín (2000), was a Scandinavian that reached the following fairly common position after five moves.

It occurs 484 times in ChessBase's 2026 Mega database, with the vast majority of games continuing 6.Nf3 or 6.Bc4

*****

*****

*****

*****

*****

*****

I tried 6.g4!?, which is obviously double-edged, but also much liked by Stockfish17.1 and Dragon1. I had not prepared the move, but had seen it the previous evening in passing, either in the diagram position or in one much like it.

My opponent replied 6...Be6!?, of which there are two examples in Mega26, compared with 17 for 6...Bg6 (and two for 6...Bc8?!).

The engines come to like the text best, and the game continued 7.Bg2 Qd8!?, which is also how the two games in Mega26 went.

How should White proceed?

*****

*****

*****

*****

*****

Jean-Baptiste Mullon (2445) - Martin Herndlbauer (2125), EU Team Cup (Rhodes) 2013, went 8.f4!? Qxd4 9.Nf3 Qb6 10.f5 Bd7 11.Qd2, with what the engines reckon is a game-winning initiative for White (1-0,  23 moves).

Stefan Voigt (1745) - Andreas Fischer (1917), Münster Team Tournament 2019, saw 8.g5!?, which the engines agree is the strongest move in the position. After 8...g6, White got cold feet and somewhat spoilt his position by defending d4 with 9.Nge2!?, whereas the engines reckon White can continue offering the pawn with 9.Qe2, or defend it with the more-active 9.Nf3. Nevertheless, White was considerably better for much of the game, until blundering a piece (and even then he should not have resigned, as Stockfish17.1 shows White still had plenty of resources).

I played 8.Be3, easily the least-dynamic of the three moves. After 8...Nf6 9.h3 Nd5 10.Nxd5 Bxd5 Black was close to equalising.

Going back to the diagram, the main feature of the position is the difference in development. White has developed three pieces, and is two moves away from being able to castle on either flank. Not only that, but if White castles long, the queen's rook will be on the half-open d file, currently occupied by the black queen. Black has developed one piece, and will require at least six moves to connect rooks.

In addition, it is White to move, so all in all the position is crying out for dynamic play, not stodgy consolidation. But playing Be3 moves an already developed piece, and puts it on a square where it can be attacked by a black knight on d5.

LESSON: when your advantages are dynamic, the position almost certainly calls for dynamic play - otherwise your advantages will wither away as the opponent catches up in development and/or coordination.

Lessons From Cap Negret III

PAWNS, provided they are not in danger of being captured, become more powerful the further they advance.

This increase in strength tends to be incremental, until the pawn safely reaches the seventh rank, when its power takes off thanks to the threat of promoting.

These factors were an important consideration in round three, where, as Black against Salvador Freijedo Álvarez (2123), I sacrificed a pawn to get the following late-middlegame position.

I have just made a recapture on b7

*****

*****

*****

*****

*****

White has an extra pawn, and the pawn is protected, but Black's passer has already crossed the halfway line, giving Black active play.

Stockfish17.1 and Dragon1 agree Black does not have enough compensation for a pawn, but they disagree as to White's best continuation.

Stockfish17.1 suggests 23.Qf3, but Dragon1 prefers the game's 23.Qc2, and I replied 23...c3, getting my passed pawn to the sixth rank.

My opponent continued with 24.Qe4, attacking my rook and centralising the queen, while also continuing to prevent my passer advancing further.

The engines prefer 24.Rc1, bringing the rook into play, although note that White is not threatening to capture the pawn, even if it were White's move, as 25.Qxc3?? Qxc3 26.Rxc3 would lose to a back-rank mate.

I met White's queen move with 24...Rb8, and we both took time out to ensure back-rank mate was no longer possible: 25.g3 g6.

Also possible was 25...Qxa2!? - the engines reckon both moves give equality.

White now played the blunder 26.Qe5??, and was obliged to resign after 26...Qxe5 27.dxe5 c2 - the black pawn, having reached the seventh rank, cannot be stopped from becoming a queen.

The strength of 27...c2 can be gauged from the fact that if I had followed the well-known axiom that rooks belong behind passed pawns, and instead played 27...Rc8, the engines reckon White would be winning, not Black.

LESSON: I like to think of a pawn on the seventh rank as having reached Seventh Heaven - certainly its presence often means Hell for the opponent.

Lessons From Cap Negret II

THE modern main line of the Exchange Variation of the Spanish starts 1.e4 e5 2.Nf3 Nc6 3.Bb5 a6 4.Bxc6 dxc6 5.0-0 Bg4.

Position after 5...Bg4

Castling by White on move five goes back to at least 1867, but only became popular after it was taken up by Bobby Fischer.

His early games with it elicited the reply 5...f6, but that has been overtaken in usage by the pinning bishop move.

My round-two game, where I had white against Alonso Infante Martín (1556), began 1.e4 e5 2.Nf3 Nc6 3.Bb5 a6 4.Bxc6 dxc6 5.d4.

Emanuel Lasker helped popularise 5.d4, although, at least according to ChessBase's 2026 Mega database, he played 5.Nc3 slightly more often.

The position after 5.d4 occurs 5,222 times in Mega26

In 4,822 of those games, Black continued with 5...exd4, but my game saw 5...Bg4!?

After the further moves 6.dxe5 Qxd1+ 7.Kxd1 0-0-0+ 8.Ke2 Bc5?! 9.Be3 Bxe3 10.Kxe3 White is a pawn up, and Black no longer has the bishop-pair as compensation for having a queenside pawn-majority that cannot make a passed pawn with normal play.

Stockfish17.1 and Dragon1 reckon White has a winning advantage.

I do not know why my opponent played 5...Bg4!?, but I suspect he was playing by analogy, knowing it was a sound move against 5.0-0 and thinking it must be right against 5.d4 too.

As a matter of fact, as I pointed out in my blog notes, 5...Bg4!? is playable, but Black must have one of two plans in mind.

Historically, the first plan was to make the game a definite gambit with 8...f6!? (instead of 8...Bc5?!). After 9.exf6!? Nxf6 10.Nc3 Bb4 11.h3 Bh5 12.g4 Bg6 13.e5 Nd7 Black had excellent compensation for a pawn in Kornél Havasi - Alexandru Tyroler, Hungarian Championship (Temesvár - now Timișoara, Romania) 1912, with Black going on to win in 28 moves.

The second plan, apparently introduced by Siegbert Tarrasch against Walter John at the 1914 German Chess Federation Congress at Mannheim, was to immediately go after White's extra pawn with 8...Re8.

That game continued 9.h3 Bxf3 10.Kxf3 Rxe5, with an equal game, according to Stockfish17.1 and Dragon1, although White went on to win in 36 moves.

However, the point is that Black more-or-less must adopt one of those two plans (make the game an out-and-out gambit, or fight to win back the pawn); merely getting on with development is unlikely to be enough.

LESSON: similar positions sometimes require similar plans, but you cannot just play by analogy as there will be occasions when, as in the two lines of the Spanish Exchange given above, an apparently small difference (5.d4 instead of 5.0-0) has a big effect.

Lessons From Cap Negret

IN my round-one game, where I had black against Manuel Lozano Marqués (1635), I sacrificed a pawn in the opening, and pretty soon blundered a second one.

There was no problem recapturing one of them, but for the rest of the game I was a pawn down.

Nevertheless, I can remember thinking I had plenty of compensation, and hoped, bearing in mind the 261-rating difference, to wrangle a win.

As my blog notes to the game show, I was fooling myself.

Ridiculously, even after reaching this opposite-coloured-bishop ending a pawn down, I thought I had winning chances

I would have done well to remember David Bronstein's advice in his famous book Zurich International Chess Tournament 1953: "To lose one's objectivity is almost invariably to lose the game as well."

Fortunately, I drew the game, but it was a rotten way to start a tournament.

LESSON: it is very difficult to play well if you evaluate positions hopelessly optimistically (or hopelessly pessimistically, for that matter).

Summing Up Cap Negret

MY score in the 65+ seniors of +4=2-3 gained 11.4 Fide elo.

Cap Negret Round Nine

Alberto Pastor Alonso de Prada (2000) - Spanton (1896)

Réti

1.Nf3 d5 2.c4 e6 3.b3 d4!?

3...d4!? is fourth in popularity in ChessBase's 2026 Mega database, behind 3..Be7, 3...c5 and especially 3...Nf6, but is the top choice of Stockfish17.1 and Dragon1

*****

*****

*****

*****

*****

*****

4.e3 c5 5.g3 Nc6 6.Bg2 Nf6 7.0-0 d3?

This turns a slightly advantageous position into a losing one (Stockfish17.1), or at least one in which White has the upper hand (Dragon1).

8.Bb2 Bd6 9.Nc3 e5 10.e4?!

But this apparent-novelty throws away a lot of White's advantage. The engines prefer the known 10.Qb1.

10...Bg4 11.h3 Bh5 12.Re1 Bxf3?

Better is 12...Nd4, although the engines agree White has at least a slight edge.

13.Qxf3 Nb4 14.Red1 Bc7 15.a3 Nc2 16.Rab1 Nd4 17.Qxd3 Nxb3 18.Nd5 Nxd5?

Better is 18...Na5.

19.exd5 Na5 20.Bc3 Bb6?!

Almost certainly better is 20...Bd6.

21.Re1

21.Bxe5 is perfectly safe, but there is no rush to grab material.

21...0-0!?

This is the engines' top choice, ahead of 21...f6 22.Bxa5 Bxa5 23.Rxb7.

How big is White's advantage?

*****

*****

*****

*****

*****

White is the equivalent of almost a rook ahead, according to the engines.

22.Rxe5 Nc6 23.Rh5 g6 24.Qe3!?

Stockfish17.1 fluctuates between the text and 24.Rh4; Dragon1 prefers the latter.

24...gxh5?!

Objectively bad, but Black is lost anyway.

25.dxc6 bxc6 26.Bxc6 Ba5!? 27.Qh6 Bxc3 28.Be4!? f5 29.Bd5+ Qxd5!?

Forced.

30.cxd5 Bg7

How would you assess this imbalanced middlegame?

*****

*****

*****

*****

*****

Materially, White is 'only' up queen and pawn for rook and bishop, but Black's weak pawns and exposed king mean White's advantage is almost the equivalent of being a queen up for nothing, according to Dragon1, although Stockfish17.1 is less effusive.

31.Qe6+ Kh8 32.Rb7 Rae8 33.Qd7 Rg8 34.Qxf5 Re5 35.Qf7 h4 36.g4 h5?! 37.f4 Re2 38.Qxh5+ 1-0

Cap Negret Round Eight

FACED a fellow Battersea Chess Club member.

Spanton (1896) - Paul Stokes (1807)

QGD Exchange

1.c4 e6 2.Nc3 d5 3.d4 Nf6 4.cxd5 exd5 5.Bg5 Be7 6.e3 c6 7.Bd3 Nbd7 8.Qc2 h6!?

Magnus Carlsen has helped make ...h6 a fashionable move in the Exchange Variation of the Queen's Gambit Declined.

9.Bh4 0-0

Levon Aronian (2777) - Carlsen (2685), Grenke Classic (Baden-Baden) 2015, went 9...Nh5 10.Bxe7 Qxe7 11.0-0-0!? Nb6 12.Nf3 Bg4 13.Kb1 Nf6 14.Rc1 Nfd7 15.Ka1!? 0-0-0, when White is slightly better, according to Stockfish17.1 and Dragon1 (½–½, 64 moves).

10.Nf3 Ne8 11.Bg3 Bd6 12.0-0 Bxg3 13.hxg3

With the opening over and the middlegame underway, how would you assess the position?

*****

*****

*****

*****

*****

It is largely typical of one arising from the Exchange Variation of the Queen's Gambit, with White ahead on development and ready to start a Minority Attack on the queenside. Perhaps the main difference from such typical positions is White's damaged kingside. Stockfish17.1 reckons the game is equal, but Dragon1 hovers between equality and giving White a slight edge.

13...Nd6 14.b4 Nf6 15.b5

The engines prefer preparing this with 15.a4, as in Jacobo Bolbochán - Carlos Enrique Guimard, Mar del Plata (Argentina) 1943, which continued 15...Be6 16.Rfc1 Rc8 17.Qb1 Qe7 18.b5 c5!? 19.dxc5 Rxc5 20.Nd4 Rfc8, with equal chances, according to the engines (½–½, 44 moves).

15...cxb5 16.Nxb5 Nxb5 17.Bxb5 Be6 18.Rfc1 Rc8 19.Qb3 Qb6 20.Ne5 Rc7 21.Rxc7 Qxc7 22.Bd3 Nd7!?

22...Nd7!? is Stockfish17.1's top choice, but Dragon1 is less keen for quite some time, until eventually agreeing the text is best (I thought it strange to offer to swop off the better of Black's two minor pieces, but the engines presumably take the view that the exchange is worth it to get rid of White's well-placed knight)

*****

*****

*****

*****

*****

23.Nxd7 Qxd7 24.Rb1 b6 25.Ba6!? Qc6 26.Qa3 Rd8 27.Rc1 Qd6!?

The engines prefer 27...Qd7.

28.Qa4

After 28.Qxd6 Rxd6 29.Bb5!? Black is slightly better, according to the engines.

28...Bd7 29.Qb3 Bf5 30.Qc3 Be6 31.a4 g6 32.Rb1 Rd7 33.Rc1 Re7 34.a5 bxa5 35.Qxa5 Kg7 36.Be2 Rb7 37.Ra1 Rc7 38.Bf3 Qb6!?

The engines are fine with this move - Stockfish17.1 reckons the game is completely equal, although Dragon1 is less sure - but after queens come off, White has new probing chances to explore.

39.Qxb6 axb6 40.Rb1

How should Black proceed?

*****

*****

*****

*****

*****

40...Rc6

Black does not need to immediately defend the b pawn, so the engines suggest 40...h5!? or 40...Rc8!? After, for example, 40...h5!? 41.Rxb6 Rc1+ 42.Kh2 Rc2, they reckon the game is completely equal.

41.g4!?

The engines at first rate this move quite highly, but given plenty of time they come to view it as not giving much of an edge at all, although agreeing it is the best try in the position.

41...g5!?

41...Kf6 is preferred by the engines.

42.Rb5 Rd6 43.Kh2!?

The king is headed for g2, which makes the text a complete waste of a tempo, but that is not serious in a position where the opponent has no counterplay.

43...Kf8 44.g3 Ke7 45.Kg2 Kd8 46.Be2 Kc7 47.f4!?

The point of White's manoeuvres - to open a second front.

47...f6 48.f5!? Bd7 49.Rb1

White now has more pawns than Black on light squares, but space has been gained, although the engines reckon the position is completely equal

*****

*****

*****

*****

*****

49...h5?

Several moves maintain equality, according to the engines, including 49...Rc6 and 49...Ba4. Their point is White cannot gain an advantage going after the h pawn as the b pawn is too fast.

50.Rc1+ Kb7 51.gxh5 Bxf5 52.g4

Even stronger, according to the engines, is 52.Rf1 or 52.Kf3.

52...Bh7 53.Rf1 Kc6 54.Rc1+ Kd7

If 54...Kb7, the engines reckon both 55.Kf2 and 55.Bf3 are winning.

55.Bb5+ Ke6 56.Rc7 Be4+ 57.Kg3 Rd8 58.Rc6+?!

White is still better after this, but 58.Rb7 wins a pawn as 58...Rd6? can be met by 59.h6 Rd8 60.h7! (60.Rxb6+ Ke7! holds, according to the engines) Rh8 61.Bd7+ Kd6 62.Bf5, winning.

58...Rd6?

58...Ke7 and 58...Kf7 seem to give fair drawing chances, although the engines disagree as to which move is better.

59.Rc8 Kf7 60.Rc7+ Kg8 61.Bd7!?

Black's rook is imprisoned, but it seems 61.Rb7 is even stronger

*****

*****

*****

*****

*****

61...Kh8

The engines much prefer 61...b5!, giving up the b pawn to get counterplay, eg 62.Rb7!? Ra6 63.Rxb5 Ra2 64.Be6+ Kg7 65.Bxd5 Rg2+ 66.Kh3 Re2 67.Rb7+ Kh8 68.Rb8+ Kg7 69.Rg8+ Kh7 70.Rd8 Rxe3+ 71.Kh2 Bc2, at the end of which White has a slight edge (Stockfish17.1) or at best the upper hand (Dragon1), although the line is very hard to visualise from the diagram.

62.Kf2?!

Almost certainly better is 62.Rb7 or 62.Bf5!?

62...Kg8

The engines reckon ...b5 is still the best try.

63.Ke1?!

Again the engines much prefer Rb7 or Bf5.

63...b5!

White has just a slight edge after this sacrifice, according to the engines

*****

*****

*****

*****

*****

64.Bxb5 Rb6 65.Rc5 Bf3?

The engines reckon Black holds with both 65...Kh7 and 65...Kg7.

White to play and win

*****

*****

*****

*****

*****

66.Bd7! Rb1+

Or 66...Be4 67.Bc6 Rb1+ 68.Kf2 Rb2+ 69.Kg3 Rg2+ 70.Kh3 Re2 71.Bxd5+ Bxd5 72.Rxd5 Rxe3+ 73.Kg2, after which White is 'only' a pawn up in a rook-and-pawn ending, but White has two passers, and the black pawns are held up by a solitary white one.

67.Kf2 1-0

After 67...Be4 68.Be6+ Kg7 69.Bxd5, White has an even better version of the line given in the previous note.

Cap Negret Round Seven

FACED a Luxembourg-registered player.

Geoffrey Stern (1970) - Spanton (1896)

Jobava-Prié

1.d4 d5 2.Nc3 c5!?

Not a common move, but it has been played by Magnus Carlsen, and it scores 52% in ChessBase's 2026 Mega database.

3.Bf4!?

Most popular in Mega26 is 3.e4, making the opening an Albin with colours reversed and White having the extra tempo Nc3. Stockfish17.1 and Dragon1 rate the two moves almost identically.

3...Nc6 4.e3 a6

Preventing 5.Nb5.

5.Nf3 e6 6.dxc5!?

Played so Bd3 cannot be met by ...c4. The engines prefer the more popular 6.Be2.

6...Bxc5 7.Bd3 Nf6 8.0-0 0-0

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

*****

*****

*****

*****

*****

White is slightly ahead on development, and Black has a problem-bishop at c8, but Black has more space in the centre and a 2-1 majority on the central two files. The engines reckon the game is equal.

9.e4 dxe4 10.Nxe4 Nxe4 11.Bxe4

How should Black proceed?

*****

*****

*****

*****

*****

11...Qxd1?!

This has been tried by at least one 2400+ player, but is probably a mistake. The engines like 11...Qf6 or 11...f6!?

12.Rfxd1 Re8!?

This apparent-novelty comes to be Stockfish17.1's top choice, although Dragon1 prefers another move not in Mega26, 12...Ba7!? However, both engines fluctuate between moves, with 12...f6 and 12...Nb4 among other suggestions.

13.Ne5!? Nxe5 14.Bxe5 f6 15.Bd6 Bxd6 16.Rxd6 f5 17.Bf3 e5?!

Probably better is 17...Kf7 or 17...Rb8, although the engines give White at least the upper hand.

18.Bd5+ Kf8 19.Rd1 Rb8

Perhaps 19...Ke7 is better, but 20.Rb6 comes anyway.

20.Rb6 e4

What should White play?

*****

*****

*****

*****

*****

21.c4!?

This is Dragon1's top choice for quite some time, but it eventually prefers Stockfish17.1's 21.h4, although the latter engine fluctuates between pushing the h pawn and playing 21.Bc6. It seems Black is so tied down that White can take time to play almost anything.

21...f4?

The engines much prefer giving up the b7 pawn with 21...Bd7!?, whereas the text loses material without freeing Black's queenside.

22.Re1 e3?

Better is 22...Bf5.

23.fxe3 Rxe3

Black gets mated, or loses a catastrophic amount of material, after 23...fxe3?? 24.Rf1+, eg 24...Ke7 25.Rf7+ Kd8 26.Rd6+ Bd7 27.Rdxd7+ Kc8 28.Rc7+ Kd8 29.Rfd7#'

24.Rf1 g5 25.h4 Bg4!? 26.hxg5 f3!? 27.Bxf3 Bxf3 28.Rxf3+ Rxf3 29.fxf3 Rc8 1-0

Cap Negret Round Six

Spanton (1896) - Josep Fernández Pérez (1801)

French Exchange

1.e4 e6 2.d4 d5 3.exd5 exd5 4.Nf3 Bd6 5.c4!?

Stockfish17.1 and Dragon1 slightly prefer this over the more popular 5.Bd3, and the text scores 14 percentage points higher in ChessBase's 2026 Mega database.

How should Black respond?

*****

*****

*****

*****

*****

5...c6

This is Dragon1's top choice, but Stockfish17.1 favours 5...Nf6.

6.Nc3 Ne7!?

The idea of this passive-looking placement of the king's knight is to meet Bd3 with ...Bf5, in order to swop light-square bishops, which generally speaking should benefit Black when pawns are fixed on d4 and d5, as in the diagram.

7.Bd3 0-0!?

The main move is 7...dxc4, giving White an IQP with tempo.

How should White proceed?

*****

*****

*****

*****

*****

8.0-0

Not 8.Bxh7+?? Kxh7 9.Ng5+ Kg8 10.Qh5 as Black has the winning 10...Bf5.

8...Bf5!?

Thematic, and it has been tried by players rated over 2400, but there is a problem with the move on this occasion.

9.Bxf5 Nxf5 10.Qd3
This is the engines' second choice, but it seems to be a novelty, probably because most players have preferred 10.cxd5 cxd5 11.Qb3.

10...g6

The engines suggest 10...Qd7 or 10...Ne7.

11.c5

Swoping on d5 is still the way to go, followed by 12.Qb5.

11...Be7 12.Bf4 Na6 13.h3 Nc7 14.b4 Ne6 15.Be5 Bf6 16.a4?

Better is 16.Rfe1.

16...Bxe5 17.Nxe5

If 17.dxe5, then 17...a5!

How should Black proceed?

*****

*****

*****

*****

*****

17...f6?

Black wins an important pawn with 17...Nfxd4, as attempting to trap the knight with 18.g4? only leaves matters worse after 18...f6.

18.Nf3 Qd7 19.b5 Rab8

The engines reckon Black, rather than defend the queenside, should go for counterplay on the kingside with 19...Nf4 or 19...Rae8.

20.Rfb1 Qc7 21.Rb4

The engines prefer 21.a5, and if 21...Qf4, then 22.Rd1 or 22.Kf1. The point about 22.Kf1 is that 22...Nfxd4?? fails to 23.Rb4 as ...Nxf3 does not come with check.

21...Qf4?!

This threatens 22...Nxc5, but that is easily countered.

It seems Black can more-or-less equalise by instead playing 21...a5!, coming at a time when White cannot gain advantage from opening the b file, eg 22.bxa6 bxa6 23.Rab1 Rxb4 24.Rxb4, and now 24...Re8!? or 24...Ng5!?, with serious kingside counterplay, one point being 24...Re8!? cannot be met by 25.Qxa6? as Black has 25...Nexd4!, when 26.Nxd4?? runs into 26...Re1+. After 24...Ng5!? 25.Nxg5 fxg5, the engines reckon White can play 26.Qxa6, but 26...Qf4 draws, eg 27.Qd3 Qc1+ 28.Kh2 Qf4+ etc, or 28.Nd1 Re8, with strong play for Black.

22.Rab1 a5

This comes to be Dragon1's top choice, at least for a while, but 22...Qc7 may be better.

23.bxa6 bxa6 24.Ne2!?

The key move, protecting d4 with tempo, leaving the white queen free to capture on a6.

24...Qc7 25.Qxa6 Ra8 26.Qd3 Rfb8 27.Rxb8+ Rxb8 28.Rab8+ Qxb8 29.a5 Nc7 30.Nd2

The engines prefer kingside play, eg 30.g4!? Ng7 31.g5!?, when Black has to worry about the king as well as about the a pawn queening.

30...Qb5 31.Qxb5

How should Black recapture?

*****

*****

*****

*****

*****

31...Nxb5?!

The engines strongly prefer creating a rival passed pawn with 31...cxb5.

32.Nb3 Nc7 33.Kf1 Ne7 34.Nc3 Nc8

The engines prefer activating the king.

35.Ke2 Kf7 36.Kd3 Ke6 37.Na2 Na6 38.Nd2 Na7 39.Nb1 Kd7 40.Na3 Kc7 41.Kc3 Kb7 42.Nb4 Nc7

The engines agree this is much better than swoping a pair of knights.

43.Nbc2?!

It seems this knight should have been left where it was to keep the black king out of a6.

43...Ka6

Now the black king is blockading the passed pawn, how would you assess the position? 

*****

*****

*****

*****

*****

Completely equal, according to Stockfish17.1, although Dragon1 gives White a token edge of about 0.1.

44.Kb4 Ncb5?!
Now White is back in business. Instead 44...Ne6 maintains equality, according to the engines.

45.Nb1 Kb7 46.Nd2 Nc8 47.Nf3 Ka6 48.Ne3

The engines reckon 48.g4!? gives at least a slight edge.

48...Ne7 49.g4

A move too late, apparently, as Black can now gain complete equality, according to the engines, although their assessment of other moves fluctuates. However it is possible 49.Ng4 or 49.Nc2!? is better.

49...h6

This may be enough for a draw, but the engines reckon 49...g5 virtually guarantees it.

50.Nc2 Nc7

Temporising with the king may be better.

51.Nfe1

51.Ka4 gives at least a slight edge, according to the engines.

51...Ne6 52.Nd3 g5!?

Dargon1 prefers temporising with the king, or pushing the h pawn, but Stockfish17.1, although at first sceptical, comes to be fine with the text..

53.Ka4 Nc7?

But this is a mistake. Instead 53...Kb7 seems to hold, although Dragon1 is less sure than Stockfish17.1.

54.f4

This seems enough for at least the upper hand, and is probably winning, but 54.Ndb4+ Ka7 55.Ne3 Kb7 56.f3!? Ka7 57.a6! definitely wins, according to the engines, eg 57...Nxa6 58.Nf5! Nxb4 (58...Nxf5 59.Nxc6+) 59.Nxe7 Nc2 60.Nxc6+ etc.

54...Nb5?

This makes it relatively easy. 54...Ng6 is trickier, but Stockfish17.1's 55.f5!? seems to win, although Dragon1 for quite some time disagrees, the main line running 55...Nf4 56.Nxf4 gxf4 57.Ne1 (Dragon1 now agrees White is winning) Nb5 58.Nf3 Nc3+ 59.Kb4 Ne4 60.h4 Ka7 61.Kb3 Ka6 62.Kc2! Kxa5 63.Nd2 Ng3 64.Kd3 Kb4 65.g5! hxg5 66.hxg5 Nxf5 67.gxf6 Nh6 68.Nf3 Nf7 69.Ne5 etc.

55.fxg5 fxg5 56.Ndb4+ Ka7

White to play and win

*****

*****

*****

*****

*****

57.Nxc6+! Nxc6 58.Kxb5 1-0

Streak Over

DEFEAT yesterday ended a 10-game unbeaten run, which is certainly one of the longest I have enjoyed over the board.

I scored +6=4-0, for a rating performance of 2039, which is 116 elo higher than my average rating during those games.

Cap Negret Round Five

FACED a Belgian Fide master on a live board.

Johan Goormachtigh (2188) - Spanton (1896)

QGD Ragozin

1.d4 d5 2.Nf3 e6 3.c4 Nf6 4.Nc3 Bb4 5.cxd5 exd5 6.Bf4!? 0-0 7.e3 Bf5

Our game from the same tournament last year went 7...c6 8.Bd3 Nbd7 9.0-0 Re8 10.Qc2, with a slight edge for White, according to Stockfish17.1 and Dragon1 (½–½, 59 moves).

8.Qb3 c5!? 9.dxc5 Qa5 10.Be2

The engines reckon White is at least slightly better after 10.a3 or 10.Nd4.

10...Nd7 11.0-0

The engines still prefer a3, or Bd6.

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

*****

*****

*****

*****

*****

Black has an IQP and is a pawn down, but that pawn-deficit is about to be erased with tempo, driving the white queen to the back rank. The engines prefer Black.

11...Nxc5 12.Qd1 Rfd8?!
Shant Sargsyan (2594) - Alan Pichot (2630), Abu Dhabi Online Blitz, went 12...Bxc3 13.bxc3 Nce4, when Black had a slight edge (Dragon1), although Stockfish17.1 is less enthusiastic about Black's position (0-1, 45 moves).

13.Nd4

Even stronger, according to the engines, is the immediate 13.Nb5.

13...Bg6?

Black should probably play 13...Bxc3, although the engines reckon 14.Nxf5 Bxb2 15.Rb1 Qc3 16.Bf3 gives White more than enough for a pawn.

14.Ncb5 Ne6?

14...Na6 defends the sensitive c7 square, but White is positionally winning according to the engines, eg 15.a3 Be7 16.b4 Qb6 17.Rc1 Rac8 18.Qb3, although the game is far from over.

15.Nxe6 fxe6 16.Bc7 Qa6 17.Nd4!? b5 18.Bxb5 Qb7 19.Bxd8 Rxd8 20.Qa4 Bd6 21.Nxe6

The flurry of tactics is complete, and White has emerged up the exchange and two pawns

*****

*****

*****

*****

*****

21...Rb8 22.Bc6!?

Simpler is 22.Nd4, but the text is also good, and may even be better (the engines are unsure).

22...Qxb2 23.Qd4 Qc2!?

This loses another pawn, but with queens off the board Black's position is hopeless.

24.Bxd5 Qf5 25.Nf4+ Kh8 26.Be6 Qe5 27.Qxe5 Bxe5 28.Nxg6+ hxg6 1-0

Cap Negret Round Four

I WAS downfloated.

Spanton (1896) - Luis Vicente Martín Martín (2000)

Scandinavian

1.e4 d5 2.exd5 Qxd5 3.Nc3 Qa5 4.d4 c6 5.Bd2!? Bf5

Probably not 5...Qb6 6.Nf3 Qxb2?! as 7.Rb1 Qa3 leaves White with a lot of compensation for a pawn, especially after the apparent-novelty 7.Ne5!?, which is suggested by Stockfish17.1 and Dragon1.

6.g4!?

This rare continuation is much-liked by the engines.

How should Black respond?

*****

*****

*****

*****

*****

6...Be6!?

The engines fluctuate between the text and the more common 6...Bg6, coming to prefer the former.

7.Bg2 Qd8!? 8.Be3

The engines prefer gambiting the d pawn with 8.g5!? Qxd4 9.Nf3.

8...Nf6 9.h3 Nd5 10.Nxd5 Bxd5 11.Nf3 Nd7 12.Qe2 Nb6 13.b3 e6 14.a3

The engines prefer 14.c4, meeting 14...Bb4+ with 15.Kf1!?

14...Be7 15.Qd3 0-0 16.0-0

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

*****

*****

*****

*****

*****

White has more space in the centre, but the white king's position is a little loose, and the engines reckon Black has a slight edge.

16...Bxf3!? 17.Bxf3 Bg5!?

Ensuring White will not be able to use the bishop-pair to gain an advantage.
18.a4 Nd5 19.Be4 Bxe3 20.fxe3

After 20.Bxh7+ Kh8 21.fxe3 Black can choose between three promising continuations: 21...Qh4; 21...f5; and 21...g6 22.Bxg6 fxg6 23.Rxf8+ Qxf8 24.Qxg6 Qg7!? 25.Qxe6 Nf4!

20...Qh4 21.Bg2 h5 22.e4 Nb4!?

Possibly better is 22...Nf6 or 22...Ne7.

23.Qd2 a5

The engines suggest offering a pawn with 23...c5!?, eg 24.dxc5 a5 25.c3 Nc6.

24.c3 Na6 25.Qf2!?

With the knight sidelined, it is probably the black king that is most in danger, in which case White should keep queens on.

25...Qxf2+ 26.Rxf2 hxg4 27.hxg4 e5!? 28.Rf5!? exd4 29.cxd4 Nb4 30.Rd1 Rfd8 31.d5!? cxd5 32.exd5 Rd6

Who, if anyone, stands better?

*****

*****

*****

*****

*****

The position is completely equal, according to the engines.

33.Re1 b6 34.Rd1!? Rc8 35.Be4

The engines prefer active defence of b3 by 35.Re5!?, eg 35...Rc3 36.Re8+ Kh7 37.Rf1 Rd7 38.Rf8 Rcc7, and not 38...f6? 39.Be4+ etc.

35...Rc3

The engines suggest 35...g6 36.Rf4! (36.Rf3? Re8 37.Bb1 Re5) Kg7, claiming a slight edge for Black.

36.Rf3 Rc7 37.Rf2 Kf8 38.Kg2 Rc3 39.Rf3 Rxf3 40.Kxf3

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

*****

*****

*****

*****

*****

The passed pawn is firmly blockaded, and under pressure from the well-placed knight in addition to the black rook. The engines agree the position is completely equal.

40...Ke7 41.Kf4 f6 42.Kf5 Kf7

Probably not 42...g5?! 43.Kg6!?

43.Kf4 Ke7 44.Kf5 Kf7 45.Bf3 Nc2 46.Ke4 Nb4 47.Kf5 ½–½

Bunker Mentality

ALICANTE was largely Republican during the Spanish Civil War.

The country's Mediterranean coast was blockaded by the Italian and German navies, supporting Franco's rebel Nationalists.

To guard against a possible invasion in the Altea area, which has long, flat beaches, eight machinegun pillboxes, called búnquers in Catalan (bunkers, or búnkeres, in Spanish) were built on the coast.

This pillbox near Cap Negret is one of only three survivors, and only two of those are now above sea level

Another view - none saw action, but they are believed to have had a deterrent effect

Cap Negret Round Three

Salvador Freijedo Álvarez (2123) - Spanton (1896)

QGD Semi-Tarrasch

1.d4 d5 2.Nf3 e6 3.c4 Nf6 4.Nc3 c5

This is fifth in popularity, behind 4...Nbd7, 4...Bb4, 4...c6 and especially 4...Be7.

5.e3

The main move is 5.cxd5, but there are 3,721 examples of the text in ChessBase's 2025 Mega database.

5...Nc6 6.cxd5 exd5

How should White proceed?

*****

*****

*****

*****

*****

7.Be2

This is the commonest continuation in Mega25, but Stockfish17.1 and Dragon1 marginally prefer 7.Bb5, when the main line in Mega25 runs 7...Bd6 8.dxc5 Bxc5 9.0-0 0-0 10.b3, with a slight edge for White (Dragon1), or at least the better part of equality (Stockfish17.1).

7...c4!?

This is Dragon1'a top choice by a small margin over the much more popular 7...cxd4, which is Stockfish17.1's recommendation.

8.0-0

This is the normal continuation, and the engines' preference. The problem with 8.b3!?, immediately attacking the advanced pawn, is that Black has the pinning 8...Bb4.

8...Bb4 9.Bd2 0-0

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

*****

*****

*****

*****

*****

White has a slight lead in development, but the game is probably going to hinge on how White deals with Black's space-gaining queenside pawn-majority. Stockfish17.1 gives White a slight edge; Dragon1 calls the game equal.

10.b3 Bxc3 11.Bxc3 Ne4

Possibly 11...b5 is a little better.

12.Rc1

This seems to be a novelty. The engines like 12.Be1!?, while three queen moves have been tried - to e1, c1 and c2.

12...Nxc3 13.Rxc3 b5 14.bxc4 Qa5 15.Rc1 bxc4

The engines reckon 15...dxc4?! strongly favours White after 16.a4!

16.Nd2?!

The engines give 16.e4!? dxe4 17.Rxc4, claiming equal chances.

16...Re8?!

Black has at least a slight edge, according to the engines, after 16...Rb8, occupying the open file.

17.Bf3

How should Black respond?

*****

*****

*****

*****

*****

17...Bb7!?

Preparing a pawn sacrifice for active play.

Note that in the diagram, White threatens 18.Nxc4, when 18...dxc4? makes matters worse as 19.Bxc6 wins back the piece and forks the black rooks.

The aggressive 17...c3? leaves Black in trouble after the simple 18.Nb3, and even worse is 17...Qxa2?? 18.Bxd5.

The engines recommend 17...Bd7, with what they reckon is equality.

18.Rb1 Rab8!?

Not 18...Nd8? 19.Nxc4!, but the engines suggest 18...Ba6, again with equal chances.

19.Nxc4!

The engines agree White has to accept the sacrifice to gain an advantage.

19...dxc4 20.Rxb7! Rxb7

The engines' 20...Nb4!? may be slightly better.

21.Bxc6 Reb8 22.Bxb7 Rxb7

How would you assess this heavy-piece late-middlegame?

*****

*****

*****

*****

*****

White is a pawn up, but Black's more-advanced passer is dangerous. Dragon1 reckons White is winning; Stockfish17.1 gives White only a slight edge. If the engines are given extra time to study the position, Stockfish17.1 comes for a while to give White the upper hand, but reverts to awarding only a slight edge. Dragon1, meanwhile, downgrades White's advantage to enjoying the upper hand.

23.Qc2
At first both engines reckon 23.Qf3 is best, but Dragon1 comes to prefer the text.

23...c3 24.Qe4

Dragon1 is convinced 24.Rc1 is much stronger; Stockfish17.1 is less sure. Note that 24.Rc1 does not threaten to immediately capture on c3 as White (as well as Black) has back-rank problems.

24...Rb8 25.g3 g6

This is good enough for equality, according to the engines, which reckon 25...Qxa2 also equalises.

White to play and lose

*****

*****

*****

*****

*****

26.Qe5??

Various moves maintain complete equality, according to the engines, including 26.Qc6, 26.Rc1 and 26.h4.

26...Qxe5 27.dxe5 c2 0-1

Morning Sunshine

Seaview

Looking inland

More mountains