Sunday, 22 November 2026

Lessons From Cap Negret VII

THE following imbalanced, but equal, position was reached after White's 29th move in my round-eight game, in which I was Black against Alberto Pastor Alonso de Prado (2045).
White has just played 29.Rf4xf7
*****
*****
*****
*****
*****
Black has two pawns for the exchange, and the game is completely equal after 29...Re8, 29...Be5 or 29...Kh7, according to Stockfish19 and Dragon1.
Instead, I played ...Qg6. I was aware White would probably reply with Qg2, but I realised my bishop is not directly endangered as Rxg3 fails to ...Qxf7, and Qxg3 fails to ...Qxg3+, Rxg3 Kxg7.
However, after 29...Qg6?? 30.Qg2, I realised that my planned ...Re8, threatening ...Re1+, is easily met by Rxg7+, so Black is not only lost, but, according to the engines, stands at about -7.00.
The game finished 30...Qd6 31.Rxg3 Kxf7 32.Rxg7+ Kf8 33.Rg8+ Kf7 34.Qg7+ Ke6 35.Qxg6+ 1-0.
LESSON: setting up a self-pin is rarely a good idea, and must be carefully calculated - and even then should probably be avoided if at all possible.

Saturday, 21 November 2026

Lessons From Cap Negret VI

I TRY to annotate games for Beau Chess as a neutral observer, seeking, with a big assist from engines, to get at the truth of what went on, not worrying about the number of question marks, etc, I award.
But looking afterwards at my round-seven game, in which I was White against Salvador Herrero Ferrando (1828), it is quite remarkable how many mistakes and blunders were committed in a short space of moves.
The mayhem began in the following position after Black's 12th move.
Black has just played the apparent-novelty 12...b6!? in what is clearly a sharp position with material and positional imbalances
*****
*****
*****
*****
*****
There followed 13.Be3? Bb7 14.Nxb6?? axb6? 15.Bxb6+ Ke7 16.a3 Bd6 17.c5?! Bb8 18.f3 Rc8 19.Nd4? d5? 20.cxd6+ Bxd6 21.Rc1? Nfd7??
In other words, I played three moves deserving a ?, one that received ??, and one that got away with ?!; my opponent played two moves that earned a ?, and one a ?? (a harsher annotator might have come close to wearing out the computer's question-mark key).
It all rather reminds me of the famous dictum - not, apparently, originally by Tartakower - that the winner of a game of chess is the player who makes the next-to-last mistake.
LESSON: if both players in a game fail to exercise basic tactical control, the outcome is almost reduced to random chance.

Friday, 20 November 2026

Lessons From Cap Negret V

IN round six I was Black against 88-year-old Guillermo Buxade Roca, who used to have a Fide rating of 2285, but is now 1875.
I firmly believe that as players age, they mainly lose their tactical skills, not their positional understanding.
So, for example, a 75-year-old rated 1800 may well be 2000 positionally, but 1600 tactically.
With juniors, as is well-known, the opposite is true, so a 15-year-old rated 1800 may well be 1600 positionally, but 2000 tactically.
This is why a common piece of advice for playing juniors is to get queens off the board as soon as possible, and to avoid wild tactical positions. 
Conversely, I believe that against seniors, it makes sense to keep queens on the board, and to embrace opportunities for hand-to-hand fighting.
However, in my game, I met 1.e4 with the French, 1...e6, and my opponent made sure the game would take on a strongly positional character by continuing 2.b3.
Queens came off at move 17, and the game was drawn in 35 moves, costing me 2.6 Fide elo.
LESSON: playing into an opponent's strengths is unlikely to do you any favours.

Thursday, 19 November 2026

Lessons From Cap Negret IV

IN round five, where I was White against Antonio Chamorro Iglesias (1793), the following position was reached after my seventh move.
I have just played 7.Nc3-d5 in response to 6...Bc8-e6
*****
*****
*****
*****
*****
One reason for playing Nd5 when I did - I could have played it earlier - is Black cannot reply 7...Nxd5.
The diagram position is reached 23 times in ChessBase's 2026 Mega database, with the most popular continuation being 7...Be7.
But six games saw my opponent's continuation, 7...Bxd5, including one game in which Black was grandmaster Ľubomír Ftáčnik (2551).
The idea of the capture is that, after 8.cxd5 Nb4, Black is threatening to win a pawn.
But White has several promising replies, including my apparent-novelty 9.d4!? (the second choice of Stockfish19 and Dragon1) and 9.Qb3, which is most popular, and is the top choice of the engines.
After the latter move, Black cannot capture on d5, which rather negates Black's whole concept, and in my game, although 9...Nbxd5 was possible, and may be best, my opponent declined to play it, perhaps because he thought it was too risky (the game's continuation of 9...exd4 10.exd4 Qa5?! 11.0-0 proved to be no picnic for Black either).
LESSON: giving up the bishop-pair to go pawn hunting before castling is usually extremely double-edged.

Wednesday, 18 November 2026

Manchester To Host The British?

THERE is a strong rumour - I put it no higher - that the ECF is in talks with a venue in the Manchester area for next year's British chess championships.
As far as I can discover, this would be a first, unless you count the 1857 British Chess Association Congress or the 1890 British Chess Association Congress (despite the same name, a different organisation).
Wikipedia also mentions the 1890 British Amateur Championship as having been held in Manchester.

ChessBase Nightmare Over (Fingers Crossed)

THANKS to advice from Paul Cooksey at the English Chess Forum, and a little digging around on the internet, I seem to have solved my problems with ChessBase changing the format of the Mega database.
The new 2cbh format means Mega cannot be used by pre-17 versions of ChessBase, including my preferred version, CB9 (actually, I would rather use CB6, but, if I recall correctly, I had to surrender that when 'upgrading' many years ago).
So, after buying the new 2027 Mega database, I had to also buy ChessBase 27 to be able to use it.
PC pointed out that, by right-clicking on the Mega icon in CB27, it is possible to convert the database from 2cbh to cbh, and so make it compatible with old versions of ChessBase.
I did that this morning, knowing, from previous similar experiences, that the database would not immediately appear in CB9.
But then I shut down my computer, restarted it, and, lo and behold, the cbh version of Mega27 was sitting nicely in CB9.
It is SO much more pleasant being able to access games in the database with a single click, rather than multiple ones, to get full player dossiers instead of cut-off ones, and to have MUCH faster searches.
And everything is on an uncluttered screen, with a light, easy-on-the-eyes background, instead of a screen with multiple bells and whistles, and way too much chrome, and a dark background - and all the data fits without having to fiddle about!
The only slight cloud was that, when I came to load a game and call up my main analysis engines, Stockfish19 and Dragon1, the latter did not appear.
It may be I accidentally deleted it somehow yesterday when trying to get ChessBase to do what I had paid for, but, whatever the reason, it was a simple mater to reinstall the engine.
All seems now to be working well ... fingers crossed.

Mega Database 2027 - Some Details

MEGA27 has 12.6million games, up from 11.74million in Mega26, including games as recent as November 2.
My personal total has risen to 1,272, up from 1,213 a year ago, and, according to ChessBase, I still score "badly" as White and "below average" as Black.
Somewhat worryingly, I see CB27 is taking up 1.35 gigabytes in my laptop, compared with 391 megabytes for CB16, and this would not be so bad if I could delete the latter, but apparently I need administrator rights to do that, an obstacle I have never come across before when trying to remove an app.

Lessons From Cap Negret III

BEFORE considering my game in round four, where I was Black against Federico Sancho Castiello (1804), have a look at a currently fashionable line from the Accelerated Dragon in the Sicilian.
1.e4 c5 2.Nf3 Nc6 3.d4 cxd4 4.Nxd4 g6 5.Nc3 Bg7 6.Be3 Nf6 7.Bc4 0-0 8.Bb3 d5!? 9.exd5 Na5 10.Qf3 Nxb3 11.axb3 Bg4 12.Qg3 Bh5!? 13.d6!?
With minor transpositions, I reached this position as Black in a 2021 correspondence game against Andrew Bussom (2296), in a 2025 weekender at Lesa, Italy, against Giulio Beretta (2143), and earlier this year in the 65+ team championships at Durrës, Albania, against Steffen Cordes (1982)
The correspondence game (engines are allowed) was drawn in 32 moves, I lost the weekender in 55 moves (despite having a winning position at move 15), and I won in Albania in 17 moves.
My game in round four at Cap Negret started 1.e4 c5 2.Nf3 Nc6 3.d4 cxd4 4.Nxd4 g6 5.Nc3 Bg7 6.Be3 Nf6 7.Bc4 0-0 8.Bb3 d5!? 9.exd5 Na5 10.0-0 Nxb3 11.Nxb3 b6 12.d6!? exd6 13.Qf3 Bg4 14.Qg3
Position in Sancho Castiello - Spanton after 14.Qg3
The position bears some similarities to the position reached after 13 moves in the three earlier games, and I very much had those earlier games in mind when I continued 14...Bh5?!, more-or-less oblivious to the very real differences between the positions.
Two of the earlier games, against Bussom and against Cordes, continued, from the first diagram: 13...Ng4 14.0-0 Be5, after which they split.
The game with Bussom saw 15.f4 Bxd6 16.Ndb5 Nxe3 17.Qxe3 Bb4, with an equal position, according to Stockfish19 and Dragon1. However, the game with Cordes was over quickly, viz 15.dxe7 Qxe7 16.f4 Bg7 17.Rfe1?? Nxe3 0-1.
Note that in neither game, nor, for that matter, in the game against Beretta, was the position of the black light-square bishop at h5 a cause for concern.
However, it certainly was a major problem for me at Cap Negret, where, from the second diagram, the game went 14..Bh5?! 15.Bd4 Re8 16.f3!? Qd7? 17.Qh4, after which White wins material.
LESSON: chess cannot be played by analogy alone - calculation rules.

Tuesday, 17 November 2026

ChessBase Update

THANKS to help from various people, including players Hans-Georg Kleinhenz and Ian Thompson, and Steffen Giehring of ChessBase, I have solved my nightmare of having paid for and downloaded the 2027 version of Mega database, but being unable to access it.
As I feared, ChessBase have changed the format from cbh to 2cbh, which does not mean much to me, except that it makes Mega27 incompatible with versions 1-16 of the ChessBase program.
Since I was using CB9, while also having CB16, my only choice was to buy a newer version, and so this afternoon I have bought and downloaded ChessBase 27.
The good news is that my important databases, Mega and Games (a personal database of my games), appear in CB27 without any trouble.
The bad news - where do I begin?
One of the first things I noticed is that, when calling up a player's dossier in classic style, it stops after giving the player's repertoire as White, ie there is no Black repertoire.
And while going through a player's tree of games, loading takes several seconds after every move - in CB9, and every other version of ChessBase I have used, loading is instant.
At the bottom of the game pane is an ugly and annoying graph of the game's progress, which gives away how a game went, and there seems no way to turn it off.
Another frustration is that it seems you can no longer simply open a database on the main screen, allowing effortless switching between databases - instead, each database opens in a separate window.
And the general colour scheme and design is horrendous when compared with the simplicity of CB9 (and, albeit to a lesser extent, CB16).
But there is some better news, in that I have loaded my main engines, Stockfish19 and Dragon1, and I have manged to create a layout that makes the board a reasonable size and minimises the screen clutter.
The result is not a patch on CB9, but it is certainly better than nothing.

Lessons from Cap Negret II

IN round three - I took a halfpoint bye in round two - I was White against Elías Pelegri Castello (1853), who offered a draw in the following position.
Black has just captured on g3
*****
*****
*****
*****
*****
*****
I recaptured with the h pawn, giving a position that is basically equal, but which Stockfish19 and Dragon1 reckons marginally favours Black.
Four moves later, the rooks came off, after which Black is slightly better, according to Dragon1 - Stockfish19 disagrees
*****
*****
*****
*****
*****
35 moves have now been played, and I am about to err with 36.f4?
*****
*****
*****
*****
*****
I was fortunate my opponent did not, over the next few moves, play the winning ...g4, and eventually I forestalled the opportunity by playing 40.fxg5! hxg5 41.g4!?, bringing about the following position.
The game is equal, but if anyone is marginally better, it is Black, according to the engines 
I kept pushing for winning chances, studiously avoiding obvious drawing lines, until eventually Black won a pawn, and should have won the game.
LESSON: playing the rating, not the position, aka known as pushing too hard in a drawn position, is frequently a recipe for ending up with a zero.

Monday, 16 November 2026

ChessBase Nightmare - Part Two

REGULAR readers may recall the problems I had two years ago after purchasing the Mega database for 2025.
These were eventually solved, thanks to help from a fellow player, Hans-Georg Kleinhenz (ChessBase did not even acknowledge my increasingly desperate pleas for assistance, never mind email any solution), who showed me how to get Mega25 working in ChessBase (I had successfully installed it, but had no idea I had to also find a setup exe).
Last year, everything went smoothly as ChessBase provided the setup exe with the purchase, as a separate download or link, if I remember rightly.
This year, unbelievably, they have changed the system again, obliging purchasers to download a product installer, and then the product, ie Mega27.
I have done this, but cannot find how to get Mega into my versions of ChessBase (I use CB9 on a day-to-day basis, but also have CB16, but do not like it as the screen is so cluttered).
Naturally, I have emailed ChessBase asking for help, but have not received an acknowledgement of my email, and, if my experience two years ago is anything to go by, I should not hold my breath waiting for assistance.
So, my question is: has any Beau Chess reader successfully downloaded the new Mega database AND got it to appear in ChessBase, and, if so, could you please tell me how you managed it?

Lessons From Cap Negret

IN round one, where I was Black against Jon Konig (1685), the following position was reached after nine moves of a Grand Prix Attack.
I have just played 9...d6?
*****
*****
*****
*****
*****
*****
*****
A thematic idea in the GPA, and also in some lines of the Closed Sicilian, is to push on with f5, even when Black apparently has the square firmly protected.
Here, f5 is already strong, eg 10.f5!? exf5 11.Nfxd4 cxd4 12.Bf4!? Be5 13.exf5!? Bxf4 14.Rxf4 Nh6 15.fxg6 hxg6 16.Qe1+!?, when White is winning comfortably, according to Stockfish19 and Dragon1.
The game went 10.Nbxd4 cxd4 11.Bb5+, which is very good for White, but even stronger, according to the engines, is 11.f5!?, eg 11...gxf5 12.exf5 exf5 13.Qe2+ Qe7 (13...Ne7 14.Ng5) 14.Qf2!?
LESSON: many openings have thematic moves associated with them (...d5 in the Accelerated Dragon, capturing on c3 in the Nimzo-Indian, sacrificing on f7 in the Italian Game), and they should be at the forefront of players' minds as long as the possibility of playing such a move remains.

Sunday, 15 November 2026

Summing Up Altea

MY score in the Cap Negret 65+ of +3=2-3 (I also took a halfpoint bye) lost 28.8 Fide elo.
Looking at night from the venue-hotel in Cap Negret towards Altea's illuminated hilltop church, nicknamed the Dome of the Mediterranean

Saturday, 14 November 2026

Cap Negret 65+ Round Nine

For pondering
Spanton (1971) - Francisco Nieto López (1764)
Sicilian O'Kelly/Maróczy Bind
1.e4 c5 2.Nf3 a6 3.c4 Nc6 4.d4 cxd4 5.Nxd4 Nf6 6.Nc3 e6
Stockfish19 and Dragon1 prefer the main move, 6...e5, one of the points of the O'Kelly Variation being White cannot reply with Nb5.
How should White proceed?
*****
*****
*****
*****
*****
7.Be3
The main line in ChessBase's 2026 Mega database runs 7.Nxc6!? dxc6!? (7...bxc6 8.e5) 8.Qxd8+ Kxd8, and now the engines recommend 9.Bf4!?, claiming the upper hand for White.
7...Bb4 8.Nxc6!?
More common are 8.Bd3 and 8.f3, but Stockfish19 reckons 8...Ne5 and 8...d5, respectively, equalise, although Dragon1 is less sure about 8...Ne5.
8...bxc6
The engines agree this is better than allowing 8...dxc6 9.Qxd8+ Kxd8 10.e5, followed by queenside castling, and they definitely dislike 8...Bxc3+?!
9.e5 Ne4 10.Qd4 Nxc3?!
Black is only slightly worse after 10...f5!?, according to the engines.
11.bxc3 Be7
FNL offered a draw in my time.
How would you assess the position?
*****
*****
*****
*****
*****
White's extra space and freer development more than compensate for having isolated and doubled pawns, according to the engines.
12.Qg4?!
Apparently a novelty, and almost certainly not a good one. The known move 12.Bd3 is better, according to the engines, and they reckon even stronger is another apparent-novelty in 12.c5!?
12...Kf8?
The engines prefer, as a kingside defensive move, 12...g6?!, but both it and the text seem to be eclipsed by 12...Qa5!?, eg 13.Qxg7 Qxc3+ 14.Ke2 Rf8 15.Rc1 Qb2+, when the engines reckon Black is slightly better.
13.c5!
Forming a sort of advanced Maróczy Bind, with pawns on e5 and c5 instead of the normal e4 and c4.
13...Qa5 14.Qd4 Rb8 15.Bd3 Rb2 16.0-0!? Rxa2
The engines at first agree this is best, but Stockfish19 comes to prefer 16...f5.
17.Rxa2 Qxa2 18.Qb4!?
Dragon1 prefers 18.Rb1, but Stockfish19 reckons the text is much stronger.
18...f5 19.Bc4
Even stronger is Stockfish19's 19.c4! (Dragon1 agrees, once shown the move), eg 19...a5 20.Qc3 a4 21.Bc2 Bd8!? 22.Bd4 (22.Ra1 Ba5!) Ba6 23.Bd3 Bg5!? 24.f4 Be7 25.Ra1 Qb3 26.Qxb3 axb3 27.Rxa6.
19...Qc2
White to play and win
*****
*****
*****
*****
*****
20.Bxa6??
This turns a winning position, according to the engines, into a lost one. Somewhat better is 20.Qb8, but 20...Kf7 is OK for Black as 21.Bxa6?? loses to 21...Bxa6 22.Qxh8 Bxf1. Almost certainly best is 20.f3, threatening to win the black queen with 21.Rc1, when the engines reckon both 20...f4?! 21.Bxf4 Qf5 22.Bg3 and 20...a5!? 21.Qa3 give White a winning position.
20...Bxa6 21.Rc1 Qe4 22.Qb8+ Kf7 23.Qxh8
White has won the exchange, but is losing the game
*****
*****
*****
*****
*****
23...f4 24.Bd4 f3 25.gxf3?
'Best', according to the engines, is 25.h3 Qg6 26.g3 Qh6 27.Kh2, but after 27...Qxc1 28.Qxh7 Qf1!? 29.Qh5+ Kh8 30.Qxf3 Bc4 31.Qg2 Qd3, Black has a bishop for two pawns, and continuing threats (the engines reckon Black's advantage is worth more than +8.00).
25...Qg6+ 26.Kh1 Be2 0-1

Friday, 13 November 2026

Cap Negret 65+ Round Eight

Inside Altea's hilltop Church of Our Lady of Consolation
Alberto Pastor Alonso de Prado (2045) - Spanton (1971)
Réti
1.Nf3 d5 2.c4 e6 3.g3 d4!? 4.e3 c5 5.d3 Nc6 6.Bg2 Nf6 7.exd4 cxd4 8.0-0 Bd6
1,238 games reach this position in ChessBase's 2026 Mega database
*****
*****
*****
*****
*****
9.b3!?
A rare continuation, although it has been played by grandmasters. The main line in Mega26 runs 9.Bg5!? h6 10.Bxf6 Qxf6 11.Nbd2 Qe7 12.a3 a5, with an equal position, according to Stockfish19 and Dragon1.
9...0-0 10.Re1 Re8 11.Bg5!? e5!?
This seems to be a novelty. The engines slightly prefer the known move 11...h6.
12.a3 a5 13.Nbd2 h6 14.Bxf6 Qxf6 15.Ne4 Qe7 16.Qc2?!
16.Qc2?! may have been a positional pawn sacrifice, but the engines are distinctly unimpressed
*****
*****
*****
*****
*****
16...Bxa3 17.c5!?
Black's dark-square bishop is cut off from the kingside, and White gets d6 as an outpost for the queen's knight, but Black has at least the upper hand, according to the engines.
17...Bb4 18.Re2 Bg4 19.h3 Bxf3!?
The engines prefer 19...Bh5, preserving the bishop-pair.
20.Bxf3 Red8
The engines suggest 20...Rec8 21.Nd6 Rc7.
21.Nd6!? Bc3
Not 21...Bxc5? 22.Nxb7 Qxb7 23.Qxc5.
22.Rf1!?
More-or-less forced.
How should Black proceed?
*****
*****
*****
*****
*****
22...Rxd6!?
The engines reckon 22...Nb4 gives Black at least the better part of equality.
23.cxd6 Qxd6 24.Bxc6!
The engines agree this is best. Not only does it swop off Black's dangerous-looking knight, it lets White quickly get both rooks into play.
24...bxc6 25.f4 exf4 26.Rxf4 Rb8
The game is completely equal, according to the engines
*****
*****
*****
*****
*****
27.Ref2 Bg1!? 28.R2f3 Bxg3!? 29.Rxf7 Qg6??
The game remains equal after such moves as 29...Re8, 29...Be5 and 29...Kh7.
30.Qg2
Black has no answer to White's threat to play 31.Rxg7+!
30...Qd6
If 30...Kh8, then 31.Qxg3 wins.
31.Rxg3 Kxf7 32.Rxg7+ Kf8 33.Rg8+ Kf7 34.Qg7+ Ke6 35.Qxg6+ 1-0

Thursday, 12 November 2026

Cap Negret 65+ Round Seven

Looking across Altea rooftops to the skyscrapers of Benidorm
Spanton (1971) - Salvador Herrero Ferrando (1828)
Maróczy Bind
1.e4 c5 2.Nf3 e6 3.c4 Bc6 4.d4 cxd4 5.Nxd4 Qb6!?
This is seventh-most popular in ChessBase's 2026 Mega database, behind ...d6, ...Bb4+, ...Qc7, ...Bc5, ...a6 and especially ...Ng6.
How should White respond?
*****
*****
*****
*****
*****
6.Nb3
All four plausible knight moves - the other three are are Nb5, Nc2 and Nf3 - have been played, and seem much of a muchness, according to Stockfish19 and Dragon1. 6.Be3? has also been tried, but presumably because the player missed the strength of 6...Qxb2.
6...Nf6 7.Nc3 Bb4 8.Bd3 Ne5!?
This is the engines' top choice.
9.0-0 Qd6!?
And so is this.
How should White proceed?
*****
*****
*****
*****
*****
10.Nb5!?
The stem game in Mega26, Viktor Korchnoi (2595) - John van der Wiel (2540), Amsterdam 1988, went 10.Bf4 0-0 11.Re1 b6 12.Bf1!? Bb7 13.Qxd6 Bxd6 14.Bg3!?, with an equal game, according to the engines (½–½, 44 moves). Not for 22 years did the same position arise again, at least in Mega26, and then Quan Liem Le (2647) - Ian Nepomniachtchi (2658), Aeroflot (Moscow) 2010, varied with 10...Bxc3 11.bxc3 Qxd3 12.Bxe5 Qxc4 13.Re1, when White had more than enough for a pawn, according to the engines (1-0, 60 moves).
10...Qxd3 11.Nc7+ Kd8 12.Nxa8 b6!?
This apparent-novelty is the top choice of the engines. Artem Kopyl (2327) - Alexander Mikhnenko (2006), Sochi (Russia) Rapid 2023, saw 12...Nxe4?, but this was well-met by 13.Qh5! (1-0, 27 moves).
13.Be3?
The engines prefer 13.f3, continuing 13...Qxd1 14.Rxd1, and either 14...Nxc4 or 14...Bb7. But their evaluations differ, fluctuating between equality and the upper hand for Black.
13...Bb7
Even stronger, according to the engines, is 13...Qxe4.
14.Nxb6??
White has to play 14.Qxd3 Nxd3, and then either 15.Nxb6 or 15.a3.
Black to play and win
*****
*****
*****
*****
*****
14...axb6?
Missing the devastating 14...Qxe4!
15.Bxb6+ Ke7
White has rook and two pawns for bishop and knight, but who stands better?
*****
*****
*****
*****
*****
Going strictly by material, using Larry Kaufman's computer-generated values in The Evaluation Of Material Imbalances (minor piece = 3.25 pawns; bishop-pair = 0.5), a rook and two pawns (5 + 1 + 1 = 7) is the same as bishop and knight (3.25 + 3.25 = 6.5) with the bishop-pair (0.5). At first glance, it might even be thought White's connected passed pawns would give White an edge, but the engines slightly prefer Black, perhaps because of the weakness of the white c and e pawns.
16.a3 Bd6 17.c5?!
The engines strongly dislike this, suggesting 17.f3 or 17.Nc5.
17...Bb8 18.f3 Rc8
Here the engines prefer 18...Qe3+, meeting 19.Kh1 with 19...Nc6!?, and 19.Rf2!? with 19...Nd3.
19.Nd4?
White should probably play 19.Re1!?, but the engines reckon Black has at least the upper hand.
19...d5?
Black is winning comfortably after 19...Qe3+ 20.Kh1 Qh6!?, according to the engines.
20.cxd6+ Bxd6 21.Rc1?
21.Qxd3 (Stockfish19) or 21.b4 (Dragon1) keeps White's disadvantage within manageable bounds.
Black to play and win
*****
*****
*****
*****
*****
21...Nfd7??
This turns a winning position into a losing one. The way to attack the loose bishop is with 21...Qa6, when White's advantage is worth almost a rook (Stockfish19) or at least almost a minor piece (Dragon1), whereas capturing on c1 or d1 gives an equal position, according to the engines.
22.Rxc8 Bxc8 23.Nc6+!?
Forced exchanges now bring about an ending in which White has the only endgame piece par-excellence, ie a rook.
23...Ke8 24.Nxe5 Qxd1 25.Rxd1 Bxe5 26.Bf4 f6
Black can save the bishop-pair with 26...Bf4!?, but White has a pleasant choice between pushing on the queenside, or grabbing the g7 pawn, eg 27.Bxg7 f6 28.g3 Be3+ 29.Kg2, and, if 29...Kf7?!, then 30.Rd3, when 30...Bc1? runs into 31.Rc3.
27.Bxe5 Nxe5
How would you assess this ending?
*****
*****
*****
*****
*****
White's advantage is roughly the equivalent of being up a minor piece, according to the engines.
28.Kf2 Nc4 29.Rc1 Ba6 30.b3!? Ne5
Not 30...Nxa3? 31.Ra1, but after the text Black threatens a fork with 30...Nd3+, a reminder that even 'simple' endgames can have tactical tricks and traps.
31.Ke3 Kd7 32.Rc5 g5!? 33.g3 h5 34.Ra5 Bb7 35.h3 Kc7 36.f4 Nc6
Not 36...Kb6? 37.fxe5! Kxa5 38.exf6, after which the white pawn queens.
37.Rc5 Kd6 38.Rc3
The engines much prefer 38.b4.
38...e5 39.fxg5 fxg5 40.h4 g4!?
Perhaps hoping to somehow create a queening threat, but it never materialises.
41.b4 Ba6 42.Rc5 Bb7 43.Rd5+ Ke6 44.Rb5 1-0
A somewhat strange time to resign, perhaps, but the engines reckon White's advantage is the equivalent of being up more than a rook and a pawn. A likely continuation was 44...Bc8 45.Rb6 Kd7 46.b5 Nd4 47.a4 Ne6 48.a5 etc.

Wednesday, 11 November 2026

Cap Negret 65+ Round Six

Old Town Altea has numerous picturesque alleys
Guillermo Buxade Roca (1875) - Spanton (1971)
French 2.b3
1.e4 e6 2.b3!?
There are are 6,172 examples of this move in ChessBase's 2026 Mega database.
2...d5 3.Bb2!? dxe4 4.Nc3 Nf6
Stockfish19 and Dragon1 also like 4...f5!?
5.Qe2 Be7 6.Nxe4 Nxe4
The engines prefer castling or 6...a5.
7.Qxe4 Bf6 8.d4 Qd5 9.Bd3 Bd7 10.Nf3!?
This natural-looking move is not in Mega26, which has 10.Qxd5, 10.0-0-0, 10.Nge2 and 10.Qe3!?
10...Qa5+
The engines prefer 10...Qxe4+ 11.Bxe4 Bc6.
11.c3
Stockfish19, but not Dragon1, prefers 11.Nd2!?
11...Bc6 12.Qe3 Nd7 13.0-0 0-0
Now both sides have castled, how would you assess the position?
*****
*****
*****
*****
*****
White's dark-square bishop is bad, but that is outweighed by White's extra space and the slightly awkward posting of the black queen, according to the engines.
14.a4 Qh5 15.Be4
I was more concerned about 15.Be2.
15...Bxe4 16.Qxe4 Qd5 17.Qxd5!?
GBR used to be rated 2285, which may be why he is happy to swop queens and steer towards a technical ending.
17...exd5
How would you assess this queenless middlegame?
*****
*****
*****
*****
*****
White's bishop is bad, but that does not seem a serious matter in this position, and anyway the bishop can be made active by playing Ba3. The engines reckon the game is equal.
18.Rfe1 Rfe8 19.Kf1 g6 20.Rxe8+ Rxe8 21.Re1 Re6!? 22.Rxe6 fxe6
How equal is this minor-piece ending?
*****
*****
*****
*****
*****
Completely, according to the engines.
23.Ke2 Kf7 24.Kd3 e5!? 25.dxe5 Nxe5+ 26.Nxe5+ Kxe5
How would you assess this bishops-of-the-same-colour ending?
*****
*****
*****
*****
*****
Black has more space, and the slightly better bishop, but those factors do not change the engines' verdict of complete equality. Or, perhaps, they are balanced by White having the smaller pawn-majority.
27.h3 Ke6 28.Bc1 h5 29.f4!? Bd6 30.g4 hxg4 31.hxg4 c6 32.Be3 a6 33.c4 Bc7 34.Bd2 Bd6 35.Be3 ½–½

Tuesday, 10 November 2026

Cap Negret 65+ Round Five

I've heard of a pub with no beer, but I've never before seen a harbour with no boats
Spanton (1971) - Antonio Chamorro Iglesias (1793)
English Symmetrical
1.c4 e5 2.Nc3 Nf6 3.g3 c5!?
An unusual way to enter the Symmetrical Variation, but there are 108 examples of the move in ChessBase's 2026 Mega database, and, with transpositions, the resulting position occurs 287 times.
How should White respond?
*****
*****
*****
*****
*****
4.Bg2 Nc6
So far, so obvious, but how should White continue?
5.e3
This is second in popularity to 5.d3, after which the main line in Mega26 runs 5...Be7 6.Nf3 0-0 7.0-0, with a slight edge for White, according to Stockfish19 and Dragon1.
5...d6
The engines suggest an apparent-novelty in 5...d5!?, the idea being to meet 6.cxd5 with 6...Nb4, when 7.e4 (the engines prefer 7.a3, 7.Nf3 or 7.d3) c4!? 8.Bf1!? b5 gives Black enough compensation for a pawn, according to Stockfish19, although Dragon1 reckons White is slightly better. Nevertheless, although the latter engine sticks with 5...d5!? as its top choice, Stockfish19 fairly quickly comes to prefer the text, or 5...g6.
6.Nge2 Be6 7.d5
More common is 7.b3!?, but that allows 7...d5!?, when the engines agree Black is better.
7...Bxd5!?
This has been played by at least one grandmaster, and scores an excellent 56% in Mega26, albeit from a small sample, but giving up Black's better bishop to go pawn-hunting, before castling, is decidedly double-edged.
8.cxd5 Nb4
How should White proceed?
*****
*****
*****
*****
*****
9.d4!?
This apparent-novelty is Stockfish19's second choice, at least for some time, but the engines agree best is 9.Qb3, when a sample line continues 9...Qb6 10.0-0 Be7 11.Rd1!? 0-0 12.d3!?, with the upper hand for White, according to the engines, but there are many reasonable alternatives along the way.
Leonardo Andrade (2238) - Ľubomír Ftáčnik (2551), Portuguese Team Championship 2010, went 9.0-0 Qd7 10.e4!? Be7 11.a3!? Nd3, with equal chances, according to the engines (0-1, 41 moves).
9...exd4!?
Perhaps Black should take the plunge with 9...Nbxd5!?, although both 10.Qb3 and 10.0-0 are promising.
10.exd4 Qa5?!
This is almost certainly a mistake. The engines suggest 10...Be7, but give White the upper hand.
11.0-0
This is much better than 11.Bd2? Nd3+.
11...Be7
Not 11...Nbxd5? 12.dxc5 Qxc5, when White is spoilt for choice, with several winning moves, including 11.Nf4, 11.Qa4+ and 11.Re1.
12.Bd2 Qc7 13.Rc1
Perhaps even stronger is 13.dxc5 dxc5 14.Nc3.
13...Na6?!
The engines suggest 13...Qd7.
14.dxc5 dxc5 15.Bf4 Bd6 16.Qa4+ Qd7 17.Qxd7+ Kxd7
How would you assess this queenless middlegame?
*****
*****
*****
*****
*****
White's bishop-pair, passed pawn and safer king more than outweigh any advantage Black may derive from having a queenside pawn-majority and fewer pawn-islands. The engines reckon White has a positionally won game.
18.Bg5!
Almost certainly not 18.Bh3+?!, as 18...Ke7! greatly reduces White's advantage, since White has difficulties defending d5 at the same time as preserving the bishop-pair.
18...Ne8 19.Bh3+
But now this is good.
19...Kc7 20.Nc3 f6?!
The engines agree with driving away the bishop, but reckon 20...h6 is the better way to do it.
21.Be3
Even stronger, according to the engines, is 21.Nb5+!? Kb6 22.Nxd6 Nxd6, and then Be3.
21...Rd8 22.Rfd1 b6 23.Nb5+ Kb7 24.a3?!
Probably too slow. Best may be 24.Bg2!?, when one line given by the engines runs 24...g5!? 25.a4 Kb8 26.f4!? h6 27.Bd2!?, with what they reckon is at least the upper hand.
24...Bb8?!
Better is 24...Nc7, maintaining the blockade.
How can White take advantage of Black's last move?
*****
*****
*****
*****
*****
25.d6!
Giving up the passed pawn to get at Black's king.
25...Nxd6 26.Bg2+ Kc8 27.Nd4 Nf7!?
This may be best, but the engines also suggest 27...g6 and 27...Rd7.
28.Nc6 Rxd1+ 29.Rxd1 Bd6?!
Perhaps best is 29...c4!? 30.Rd4 Ne5 31.Ne7+ Kc7 32.f4!? Ng4 33.Nd5+ Kc8, but 34.Rxc4+ wins back the sacrificed pawn, with a continuing attack. Note that if Black gets the king's rook into play with 29...Re8, White has 30.Bh3+ and 31.Rd7+.
30.Nxa7+ Kb8 31.Nb5 Rd8!?
This is best, according to the engines, as moving the bishop allows White's rook to occupy the seventh rank.
White to play and gain a large advantage
*****
*****
*****
*****
*****
32.Rxd6! Nxd6 33.Bf4 Kc8 34.Nxd6+ Kd7 35.Nb5 g5 36.Be3 Nc7!?
This is Dragon1's top choice. Stockfish19 very marginally prefers 36...Ke6.
37.Nxc7 Kxc7
How would you assess this ending?
*****
*****
*****
*****
*****
Two bishops are usually much stronger than a rook, even in an ending, but making progress in this position proves to take a long time. The engines reckon White's advantage is about +4.5.
38.Bf3 Kd6 39.Kf1 b5 40.Bd2 Rb8 41.Bc3 Ke6 42.Bg4+ Ke7 43.Bf5 h6 44.Ke2 Rb6 45.Bd3 c4 46.Bf5 h5 47.Kf3 Kf7 48.Bb4 Rb8 49.Bd7 Rb7 50.Bc6 Rb6 51.Bd5+ Kg6 52.Ke4 Rb8
It might seem as if White has made little or no progress, but White's advantage has grown to about +6.5, according to the engines
*****
*****
*****
*****
*****
53.Kf3
Getting cold feet about letting the rook into White's position, but there was nothing wrong with 53.Kd4, eg 53...Re8 54.Kc5!? Re2 55.Bc3 Rxf2 56.Kxb5 Rxh2 57.a4 h4 58.gxh4 gxh4 59.a5 Re2 60.a6 Re7 61.Bd4 etc.
53...f5 54.Kg2!? Re8 55.Bc6 Rb8 56.Bf3 g4 57.Be2 Re8 58.Kf1 Re4 59.f3 Rd4 60.Kf2 f4?
This simplifies White's task.
61.Bc3 fxg3+ 62.hxg3 Rd8 63.fxg4 hxg4 64.Bxg4 Kg5 65.Bf3 Rd3 66.Ke2 Rd6 67.Bd2+ 1-0

Monday, 9 November 2026

Cap Negret 65+ Round Four

Altea promenade
Federico Sancho Castiello (1804) - Spanton (1971)
Sicilian Accelerated Dragon
1.e4 c5 2.Nf3 Nc6 3.d4 cxd4 4.Nxd4 g6 5.Nc3 Bg7 6.Be3 Nf6 7.Bc4 0-0 8.Bb3 d5!?
8...d5!? dates back only to 1960 in ChessBase's 2026 Mega database, but rose in popularity after Boris Gelfand started championing it in 2016
*****
*****
*****
*****
*****
The move ...d5, without first playing ...d6, is thematic in the Accelerated Dragon, and 8...d5!? here is also the top choice of Stockfish19 and Dragon1.
9.exd5 Na5 10.0-0
A main line has yet to be established, it is probably fair to say, but this is the most popular continuation in Mega26, and scores an excellent 64%.
10...Nxb3 11.Nxb3
11.axb3!? has been played, but Black immediately regains the pawn with 11...Nxd5.
11...b6 12.d6!?
The engines agree with giving the pawn back, but saddling Black with an isolani.
12...exd6 13.Qf3 Bg4 14.Qg3 Bh5?!
Apparently a novelty, and almost certainly not a good one. Known moves are 14...Bf5 and 14...Rc8.
15.Bd4
The unsubtle 15.f3!? is also strong.
15...Re8
This is Dragon1's top choice, at least for a while, but better may be 15...h6 16.f3!? g5!?, although the engines reckon 17.f4!? gives White a positionally won game.
16.f3!? Qd7?
Reducing the protection of f6 causes Black big problems. Perhaps best is 16...Rc8!?, and if, as in the game, 17.Qh4, then 17...Rc4, as 18.g4? Rxd4! 19.Nxd4 Nxg4! 20.Qxd8 Bxd4+ 21.Kh1 Rxd8 22.fxg4 Bxg4 23.Rf4 Bxc3 24.bxc3 Bf5 sees Black emerge with two pawns for the exchange, and the much better pawn-structure, the engines reckoning Black would enjoy the upper hand. However, White improves with 18.Rae1, gaining at least the upper hand, according to the engines.
17.Qh4
Black is losing material
*****
*****
*****
*****
*****
17...Re6
There is nothing better.
18.g4 g5 19.Qxg5 Bg6
How should White, who is again a pawn up, proceed?
*****
*****
*****
*****
*****
20.f4!
This looks double-edged as every white kingside pawn advance leaves White's king a little more vulnerable, but the engines agree it is best, although Dragon1 also likes 20.Qd2!?
20...h6 21.Qh4 Rae8?
The engines prefer 21...Bxc2, when 22.f5 forces Black to give up the exchange with 22...Re5.
22.f5 Re4!? 23.Nxe4 Rxe4 24.h3
Black is already the exchange and a pawn down, and more material loss is unavoidable
*****
*****
*****
*****
*****
24...Bxf5!? 25.Bxf6 Bxf6 26.Qxf6 Bxg4 27.Kh2!?
Simplest is probably 27.Rf4, but the text is plenty good enough.
27...Bxh3 28.Qf3!? Qe6
There is no perpetual, let alone checkmate, after 28...Bxf1 29.Qxf3 Qh3+ 30.Kg1.
29.Rf2 Re3 30.Rg1+ Kf8 31.Qa8+ Ke7 32.Qxa7+ 1-0

Sunday, 8 November 2026

Cap Negret 65+ Round Three

I TOOK a halfpoint bye in the first of today's double-round games.

Spanton (1971) - Elías Pelegri Castello (1853)
Caro-Kann Modern Variation
1.e4 c6 2.d4 d5 3.Nc3 dxe4 4.Nxe4 Nd7
This is third in popularity in ChessBase's 2026 Mega database, behind 4...Nf6 and especially 4...Bf5.
5.Nf3 Ngf6 6.Ng3
There are 4,724 examples of the text in Mega26, just behind the 4,884 of 6.Nxf6!?, after which the main line runs 6...Nxf6 7.Bc4 Bf5 8.0-0 e6 9.c3 Be7, with a slight edge for White, according to Dragon1, although Stockfish19 is less sure.
6...e6 7.Bd3 Be7 8.0-0 0-0
With both sides castling early, how would you assess the position?
*****
*****
*****
*****
*****
White has more space in the centre, but this is of little consequence, according to the engines, presumably because Black is well-placed to get in a quick ...c5, without White being able to push on with d5. The engines reckon the game offers equal chances.
9.Re1 b6
This is second in popularity to an immediate ...c5.
10.Qe2 Bb7 11.Nf5!?
This is Dragon1's top choice. Stockfish18 suggests 11.Ng5!? or 11.Bf4.
11...c5!?
An apparent-novelty, and liked by the engines. Known moves are 11...Bb4!? and 11...exf5?!
12.Nxe7+ Qxe7 13.c3 Bxf3!?
This is the engines' top choice.
14.Qxf3 cxd4 15.cxd4 Qd6 16.Be3 Rfc8!?
The engines prefer 16...Rfd8 or 16...Nd5.
How should White proceed?
*****
*****
*****
*****
*****
17.Rad1
The engines give 17.Ba6 Rf8!? 18.Bb7 Rad8 19.Bf4!? Qxd4 20.Bc7 Rde8 21.Rad1 Qxb2 22.Re2, when they reckon White has more than enough for two pawns, eg 22...Qb4!? 23.Bd6 Nc5 24.Red2, but the position is hardly clear.
17...Qd5 18.Qxd5 Nxd5 19.Rc1 N7f6 20.Ba6 Rd8 21.a3 Rab8 22.Bg5 Rd7 23.Rc2 h6 24.Bh4
The engines prefer 24.Be3.
24...Nh5!?
Black is at least slightly better after 24...g5!? 25.Bg3 Nf4!?, according to the engines.
25.Bb5 Rc7 26.Rec1 Rxc2 27.Rxc2 g5 28.Bg3 Nxg3
EPC offered a draw.
How would you assess the rook-and-minor-piece ending?
*****
*****
*****
*****
*****
After ...
29.hxg3
... White has what is normally an advantageous matchup of rook and bishop against rook and knight. But here that is balanced by Black having fewer pawn islands, and in particular not having an isolani. The engines reckon the position is equal, albeit marginally preferring Black.
29...Rd8 30.Ba6 Rd7!? 31.Rc8+ Kg7 32.Bb5 Rc7 33.Rxc7 Nxc7
How would you assess the minor-piece ending?
*****
*****
*****
*****
*****
White's extra pawn island, and a lack of meaningful targets for the bishop, mean Black is slightly better, according to Dragon1, although Stockfish19 disagrees.
34.Bc6 f5!? 35.Kf1 Kf6 36.f4?!
Played to prevent ...e5, but probably a mistake.
36...Ke7 37.Ke2 Kd6 38.Bb7 Nd5
This is Dragon1's recommendation for quite some time, but Stockfish19 much prefers 38...g4!?, claiming Black is winning. Dragon1 comes to agree 38...g4!? is best, but is far less sure how strong it is.
After 38...g4!?, best play may run 39.Ke3 a5 40.Kd3 Ne8!? 41.Bc8 Nf6 42.Ke3 Nh5 43.Kf2 Kd5 44.b3 Kd6!? 45.Bb7 Nf6 46.Ke3, when the engines are enthusiastic about Black's chances, but it is not clear how Black makes progress.
39.Kf3?
Much better, according to the engines, are 39.Bxd5 and 39.fxg5!?
39...a5
The engines agree Black is winning after 39...g4+!? and 39...Nf6, but especially after the former.
White to play and gain good drawing chances
*****
*****
*****
*****
*****
40.fxg5! hxg5 41.g4!?
Preventing a favourable ...g4 seems to be the key to White's salvation.
41...Nf6 42.gxf5 exf5 43.g3 Kc7 44.Ba6 Kd6 45.Bc8 Kd5 46.Ke3?!
This may be OK, but a simpler way to a halfpoint is 46.Bxf5.
46...Ng4+ 47.Kf3!?
Dragon1 reckons this loses, but Stockfish19 disagrees. However, their evaluations fluctuate, and they agree the text is better than 47.Kd3? Nf2+. but possibly not as good as 47.Kd2!?
47...Kxd4 48.Bxf5 Ne5+ 49.Ke2 Nc4 50.b4 Nxa3 51.bxa5 bxa5
White to play and draw
*****
*****
*****
*****
*****
52.Kf3?
Eight moves draw, according to the Syzygy endgame tablebase, including 52.Bd7 and 52.Kd2, but not the text.
Black has two winning moves
*****
*****
*****
*****
*****
52...Nc4?
52...a4!? and 52...Nb5!? win, eg 52...Nb5!? 53.Kg4 a4 54.Bb1 a3 55.Kxg5 Kc3!? 56.Ba2 Kb2 57.Bf7 Nd4 58.Kf4 Nb3 etc.
White to play and draw
*****
*****
*****
*****
*****
53.Bc2!
An only-move. Dragon1 reckons is inadequate, but it draws, as Stockfish19 quickly realises. Note that the similarly motivated 53.Bd7?? loses trivially to 53...Ne5+.
53...Ne5+
Dragon1's other 'winning' move, 53...Nb2, fails to 54.Kg4.
54.Ke2 Nc4 55.Kf3
Also drawing is 55.Ba4.
55...Ne3 56.Ba4 ½–½