How Much Is 125 Euros Worth?

BACK in October-November I played in the U2000 section of the Benidorm Open at Hotel Meliá, winning 125 euros.

Prizes were to be paid by bank transfer, but problems with the Iban system meant I only received mine today.

I am accordingly £85.75 richer - Spain's "withholding tax" has made for what would otherwise appear to be a very unfavourable exchange rate (there were no bank charges).

Lessons From Benidorm VIII

MY round-nine game, in which I had black against Emilian-Paul Ioniță (1972), featured a sharp opening, decidedly out of fashion among the elite, but still popular at club level.

Because the opening is so rarely seen at the top, theory has at many points not kept up with the latest assessments of strong engines. 

The game began 1.e4 e5 2.Nc3 Nf6 3.Bc4.

Position after 3.Bc4

*****

*****

*****

*****

*****

I seem to have first had this position as Black in Guernsey 14 years ago, when I played 3...Nc6, and went on to lose. Two years later, in Gibraltar, I tried 3...c6!?, and again lost.

Since then I have only continued 3...Nxe4!?, scoring, including this Benidorm game, +5=1-3, performing 161 elo above my average rating in those games.

I thought ...Nxe4 was both well-known and best, so imagine my surprise on discovering in ChessBase's 2026 Mega database that two moves, 3...Bc5 and especially 3...Nc6, are much more popular.

Not only are those moves more popular, but they are preferred by Stockfish17.1 and Dragon1, although it should be added that they do not reckon the text is a mistake, or even dubious.

After the further moves 4.Qh5 Nd6 5.Bb3 I was aware that 5...Nc6 6.Nb5!? g6 7.Qf3 f5 8.Qd5 Qe7 9.Nxc7+ Kd8 10.Nxa8 featured in the first game of John Nunn's 1987 Batsford book, Secrets Of Grandmaster Play, co-authored with Peter Griffiths.

I had no desire to go down such a complicated line, especially as I knew that, while in the pre-computer era the compensation Black receives was regarded as very promising, engine analysis has shown ways for White to successfully hang on to the extra material.

I therefore played 5...Be7, which I had tried three times before, each opponent replying with the mainline 6.Nf3.

It seems, however, that Ioniță was aware the engines prefer, albeit marginally, 6.Qxe5, which he played quickly.

After 6...0-0 the commonest continuation in Mega26 is 7.Nge2, but Ioniță played 7.Nf3!?, which again is the engines' preference, although again they do not believe there is much difference between the two moves.

Position after 7.Nf3!?, which occurs 112 times in Mega26

*****

*****

*****

*****

*****

Black has two obvious continuations, both of which hit the white queen - 7...Nc6 and my choice, 7...Bf6!?

Of the first 10 games to reach the position after7.Nf3!? in Mega26, eight featured my choice, and the other two saw 7...c6!?

Ioniță continued with 8.Qf4!, which the engines agree gives White a slight edge (they reckon every other continuation is at least marginally favourable for Black).

After 8...Re8+ I was pleased with my position, since White must lose castling rights (9.Ne2? can be met by the winning 9...Re4!)

However, after 9.Kd1 Stockfish17.1 gives White a slight edge - the white king is safe on d1, and Black has difficulties both developing the queenside, and avoiding falling to tactics based around the vulnerability of f7, along with Black's lack of a pawn presence on the central squares.

Black's problems only gradually dawned on me, and their subtlety perhaps explains why Dragon1 reckons Black can equalise with the apparent-novelty 9...b6!? (Stockfish17.1 disagrees, reckoning the move increases White's advantage).

LESSON: going over sharp but unfashionable opening lines with modern engines is a good way to find improvements that are either unknown, or at least not known among players who have not done such work.

Lessons From Benidorm VII

IN round eight, where I had white against Miguel Ángel Fernández Alonso (1988), I accepted a draw-offer in the following position.

It's a rook-and-pawn ending in which Black's king and rook are more active than their white counterparts, but Stockfish17.1 and Dragon1 reckon White is winning

*****

*****

*****

*****

*****

The winning move is 46.Rd1, when Black's best is probably 46...Kc6.

I saw I could play 47.Rd5, preparing to menace the black kingside, but after 47...Rb1+ 48.Ke2 Rb2+ White has nothing better than 49.Rd2, when 49...Rb4 is completely equal (Stockfish17.1) or at best leaves White with a marginal edge (Dragon1).

Going back to the diagram, I naturally looked for ways to get the white king off the back rank after the white rook attacks Black's kingside. But 46.g3? Rxh2 is bad, and 46.Kg1, preparing g3, is too slow - both 46.Rd2 and 46.Rb4 seem good enough for a draw.

I cannot recall whether I seriously looked at 46.h4, giving the white king a route off the back rank via h2, but anyway 46...Rd2 holds.

The solution, as shown by the engines, is to combine the two ideas, Rd1 and h4, in a precise order, with best play running something like 46.Rd1 Kc6 47.h4! Rc2 48.Rd5 Rxc4 (48...Rc1+ 49.Kf2 Rc2+ 50.Kg1 etc, or 49...Rh1 50.h5 Rh4 51.Rg5 Kd6 52.Rxg7 Rxh5 53.Rg4 Ke5 54.c5 h6 55.Rg6 Kd4 56.Rd6+ Kxc5 57.Rxd7, when White is a protected passed pawn up, and Black has two isolanis) 49.Rf5 Kd6 50.Rxf4, when White's extra pawn and single pawn-chain give a winning advantage, according to the engines.

I accepted White's draw offer at move 45 because I genuinely thought the position was drawn, but against a lower-rated opponent I would probably have played on, hoping to induce an error.

However, I was playing someone 92 elo higher, and at the back of my mind was the thought that if I played on I might somehow be the one to throw the halfpoint away.

In other words, I was not playing the position, but rather the rating, whereas if I had played on I might have ended up winning, and at the very least I would have gained valuable experience of playing such an ending.

LESSON: every game is an opportunity for a student of chess to learn, but agreeing a draw earlier than is necessary is like leaving a lecture before it ends.

Lessons From Benidorm VI

MY round-seven game, where I had white against Malaysian junior Taj Prasshad Nambiyar Chanthiran (1518), reached the following position after my 26th move.

I have just played 26.Rbd1

*****

*****

*****

*****

*****

White has pressure down the f file and a strongly placed knight; Black controls the open file, with pressure against d3.

Stockfish17.1 gives White a slight edge, while Dragon1 reckons White has the upper hand, but the engines agree Black should kick the f5 rook with 26...g6!?, or play 26...Qd8, when the queen covers the d8-h5 diagonal as well as being well-placed to join play on the a file with ...Qa5.

Instead Black doubled up on the a file with 26...Rfa8??, allowing a decisive fork, 27.Nb6.

LESSON: Black's blunder is an example of what happens when a player focuses on the opponent's main threat (in this case, to set up a sacrificial attack against the black king), and omits to check for simple threats elsewhere.

Lessons From Benidorm V

MY round-six game, in which I had black against Singaporean junior Rahul Lakshminarasimhan (1626), featured a simple missed opportunity.

White to play and win

*****

*****

*****

*****

*****

*****

*****

The knight, which cannot move without Black suffering a catastrophic loss, is attacked three times and defended three times.

However, White can bring a fourth attacker into play long before the black king can get over to provide a fourth defender, so 22.Rc3 Kf8 23.Rfc1 is decisive.

Black can break the pin with 22...Qd6, when 23.Rfc1? can be met by 23...Ne7, but White has 23.Ba6, when 23...Rd8 again makes Rfc1 decisive.

I do not know what my opponent missed in this short tactical sequence, but perhaps he was unsure of his calculations, and so instead went for the sure win of a pawn by 22.Qa6? Qd8 23.Bxc6 Rxc6 24.Rxc6 Rxc6 25.Qxa7.

But after 25...Rc2 (not the only drawing move, eg 25...g6 is also completely equal, according to Stockfish17.1 and Dragon1) the rook on the seventh rank led to a quick draw.

LESSON: when, as in the diagram, a player is tied down and his options are limited, there is probably a strong tactical solution.

Lessons From Benidorm IV

MY round-five game, in which I had white against Spanish junior Tarek Kaji Senas (1711), reached an instructive endgame position after 31 moves.

Black has just played 31...Bh2-e5

*****

*****

*****

*****

*****

*****

*****

Material is level and both sides have doubled and isolated queenside pawns, one of which is under immediate threat. White has the only passed pawn, but Black has potential to create a passer on a flank.

Quite often in endgames, such a flank passer appears on the queenside, when it is well-known that its power to divert an enemy king can make it decisive. But that can easily also be the case with a passer on the kingside.

The position in the diagram is completely equal, according to Stockfish17.1 and Dragon1, although their first reaction on being shown it is to slightly favour Black.

I suggest this means that, although theoretically equal, the position is easier for Black to play.

But what would be the verdict if the bishops were removed? 

Same position, but without bishops

*****

*****

*****

*****

*****

The position is still completely equal, according to the engines, but it is worth watching them play it out to see why this is the case.

Best play goes something like:*

32.Kf3 Kf7 33.Kf4 Kf6 34.e5+ Ke7 35.b4!? (this freezes Black's queenside pawns, and creates a reserve tempo in the shape of b2-b3) h6 36.h4 g5!? 37.hxg5 hxg5+ 38.Kxg5 Kxe5 39.Kg6 Kd4 40.Kf6 Kc4 41.Ke7 Kxc4 42.Kd7 c5 43.Kxc7 c4 44.Kc6 Kb3 45.Kc5 ½–½

*For most moves there are two or more continuations of equal value, and I have preferred what I reckon to be the most natural, or perhaps most human.

Going back to the first diagram, it might be thought White can draw with 32.Bc3?

After 32...Bxc3 33.bxc3 White's pawn-structure has been improved, but the game is lost.

If it were White to move, the game would be drawn, more-or-less as shown above, but the sequence 32.Bc3 Bxc3 33.bxc3 costs White a tempo, and that it is all it takes to turn a draw into a loss

*****

*****

*****

*****

*****

A likely continuation is 33...g5!? (33...Kf7 also wins) 34.Kf3 h5 35.b4 Kf7 36.e5!? Ke6 37.Ke4.

Black to play and win

*****

*****

*****

*****

*****

37...c5!

After 37...g4 38.hxg4 hxg4 39.Kf4 g3 40.Kxg3 Kxe5 41.Kf3 Kd5 42.Ke3 Kc4 the white king is in time to protect c3 with 43.Kd2, and so the game is drawn.

38.bxc5

After 38.b5 c4 39.Kd4 Black creates a winning kingside passer with 39...g4.

38...c6 39.c4 h4! 40.Kd4 g4 etc.

Again going back to the first diagram, the game continuation was 32.Kf3!? Kf7 (32...Bxb2 33.Bf4) 33.Kg4 Ke6 34.Bf4!? Bxf4 35.Kxf4, resulting in a draw.

The engines reckon White has another draw with 32.Bc1!?

I rejected it for fear of getting into zugzwang, but a sample continuation shows my fear was ungrounded, eg 32...Kf7 33.Kf3 Ke6 34.Bf4, when 34...Bxf4 35.Kxf4 reaches the same drawn position as in the game.

LESSON: 'simple' endings can be just as deep, in their own way, as a middlegame slugfest.

Lessons From Benidorm III

NOTES to my round-four game - I took a halfpoint bye in round three - tell a story that will be familiar to most clubplayers.

I had black in a London System against Ireland's Pat O'Hanlon (1794), and for the first 32 moves it was a positional struggle, in which I annotated five moves with !?, but none with so much as a ?!, never mind a ?

Tactics started when White pushed his g pawn in this position, which is completely equal, according to Stockfish17.1 and Dragon1

The game lasted 14 more moves, four of which earnt a ?, one a ?! and one - the lone one of the game - an !

Clearly such annotations, even when backed by engine analysis, are subjective, but the contrast between the 32 moves of positional chess and the 14 of tactical chess could hardly be starker.

LESSON: at the highest levels chess may well no longer be "99% tactics" (if it ever was), but at lower levels 99% may be an underestimate.

Lessons From Benidorm II

MY round-two game, where I had white against Enrique Martínez Pérez (1726), gave an interesting example of why it can be dangerous to take engine assessments as gospel.

Black has just recaptured on d5

*****

*****

*****

*****

*****

*****

*****

I played 52.b4, and Black replied 52...g5, more-or-less guaranteeing a draw.

Stockfish17.1 and Dragon1 want White to play 52.Kf4, which prevents ...g5 and so wins, according to Dragon1 (it gives White's advantage as almost the equivalent of a queen).

Stockfish17.1 is a lot less certain about White's advantage, but nevertheless reckons White has the upper hand.

What neither engine appreciates is that after 52.Kf4 Black has a simple draw with 52...Bxb3!, the well-known point being White's bishop is the wrong colour for queening the h pawn.

After the further moves 53.Bxb3 Kg7 54.Kg5 Kh7 55.Kf6 Kh8 the engines stick to their assessments, ie "winning" - Dragon1; "upper hand" - Stockfish17.1, but can find nothing better than marking time with 56.Be6.

Another way of spotting the fallibility of engines is to dig out old games of yours in which you made computer-checked notes using the technology then available.

Quite often I find even a cursory look at such a game of mine with modern engines will reveal what, with the benefit of hindsight, seem rather glaring errors by older machines.

Occasionally I even get the satisfaction of seeing my original move, condemned by a version of Fritz or Junior, receiving praise from Stockfish17.1 and/or Dragon1.

LESSON: popes may be infallible, but chess engines are not.

Lessons From Benidorm

MY round-one game, where I had black against Samuel Salas Lerida (1669), reached a position after 10 moves that is not easy to judge.

I have just played 10...0-0

*****

*****

*****

*****

*****

*****

*****

The position is reminiscent of a Maróczy Bind for Black, except that the black d pawn has been swopped for White's e pawn, rather than for White's c pawn. Black has more space and is slightly more developed, while White has a slightly weakened king's position.

I cannot recall my exact thoughts, but I am fairly sure during the game I thought I had at least a slight edge.

However, Stockfish17.1 and Dragon1 reckon the game is equal (although Dragon1, if given plenty of time, comes to slightly favour Black), but with a big proviso - the 'equality' only applies if White quickly takes vigorous steps to break the bind.

Generally speaking, there are three ways to break a Maróczy Bind - by attacking it with the b pawn, the d pawn or the f pawn.

The first two options are not in play in the diagram, and are unlikely to be for quite some time, but White can instead play 11.g4!? Bg6 12.f4!, when Black has nothing better than 12...exf4, after which 13.Nxf4 gives a level game, according to Stockfish17.1, although Dragon1 slightly favours Black.

Note that 12...Nxg4!? runs into 13.f5 Bh5 14.Bxc6!? bxc6 15.Ng3, when the engines reckon White has the upper hand.

In the game White played a developing move, 11.Bg5, but after 11...h6 12.Bxf6 Bxf6 13.Ne4 Be7 the bind was intact, Black had the bishop-pair, and the knight at e4 invited Black to play the aggressive ...f5.

LESSON: I do not claim the engines' line with 11.g4!? and 12.f4! is obvious, at least not below master level, but it is an example of playing to the specifics of a position - "Black has a bind - I should be looking for a way to break it" - rather than just developing.

Summing Up Benidorm Bali

MY score in the U2000 of +3=5-0 (I also took a halfpoint bye) gained exactly two Fide elo.

I now have an 11-game over-the-board unbeaten streak, during which I have scored +5=6-0, performing roughly 68 elo above my average rating for those games.

Benidorm Bali Round Nine

FACED  a Romanian.

Emilian-Paul Ioniță (1972) - Spanton (1896)

Vienna Game

1.e4 e5 2.Nc3 Nf6 3.Bc4 Nxe4!?

This well-known combination, which relies on 4.Nxe4 being met by 4...d5, is not the top choice of Stockfish17.1 or Dragon1 - they prefer 3...Bc5 or 3...Nc6.

4.Qh5 Nd6 5.Bb3 Be7!?

This has taken over from 5...Nc6 as the most popular continuation. After 5...Nc6 the main line in ChessBase's 2026 Mega database runs 6.Nb5!? g6 7.Qf3 f5 8.Qd5 Qe7 9.Nxc7+ Kd8 10.Nxa8, which used to be regarded as a very promising sacrificial line for Black, but the engines are unimpressed, awarding White at least a slight edge.

6.Qxe5 0-0 7.Nf3!?

More prudent, and a little more popular in Mega26, is 7.Nge2, but the engines reckon the more-aggressive 7.Nf3!? is a tad better.

7...Bf6!?

Trying to exploit the fact that White's king is uncastled, but White scores 64% in Mega26 against this move, so the other way to hit the white queen, 7...Nc6. may be better.

Where should the white queen go?

*****

*****

*****

*****

*****

8.Qf4!

This is the only way to gain an advantage, according to the engines, which reckon every other move gives Black at least a marginal edge.

8...Re8+

How should White proceed?

*****

*****

*****

*****

*****

9.Kd1

Also possible is 9.Kf1, but not 9.Ne2? as that loses to 9...Re4!, eg 10.Qg3 Nf5! 11.Qh3 d5 12.d3 Rxe2+! 13.Kxe2 Nd4+ 14.Nxd4 Bxh3.

9...Nc6 10.d3

Dragon1 prefers 10.Nd5!?, but Stockfish17.1 fluctuates between the two moves.

10...Ne7!

Taking the sting out of Nd5, and making it possible to hit the white queen with ...Ng6.

11.g4?!

Probably a novelty, but Stockfish17.1 prefers the known 11.h4, and Dragon1 likes another known continuation, 11.Nd5 Nxd5 12.Bxd5.

11...Ng6 12.Qg3 Nh4?!

Black is at least slightly better after 12...Bxc3!? 13.bxc3 Qf6, according to the engines.

13.Bf4

The engines fluctuate between the text and 13.Nd5.

13...Nxf3 14.Qxf3 Be5

The engines suggest 14...Bxc3 15.bxc3 Qf6, or offering the exchange with 14...Re6!?, but anyway much prefer White.

15.Kd2?!

White is positionally winning after 15.Bxe5 Rxe5 16.h4, according to the engines.

15...Qf6 16.Nd5

An only-move.

16...Bxf4+ 17.Qxf4 Qxf4+ 18.Nxf4

Now queens are off the board, how would you assess the position?

*****

*****

*****

*****

*****

Black has difficulties completing queenside development, and so is slightly worse, according to Dragon1, but Stockfish17.1 reckons the game is equal, presumably because it does not believe White can exploit Black's difficulties.

18...c6 19.Nh3!?

Dragon1 suggests gaining space with 19.a4, while Stockfish17.1 has a similar idea, but on the kingside, with 19.h4.

19...Kf8 20.Rae1 h6 21.g5 Nf5 22.Rxe8+ Kxe8 23.Re1+ Kf8 24.g6?

24.gxh6 Nxh6 25.Ng5 d5 is equal.

24...d5?

24...Nh4 threatens both a fork at f3 and a simple capture on g6.

25.Nf4?

Correct is 25.gxf7 Kxf7, with what the engines agree is a slight edge for Black.

25...Nd4?

Threatening a fork at f3, but not simultaneously threatening to capture on g6, so again ...Nh4 is the right way to gain a large advantage.

26.Re3 Bf5 27.gxf7 Kxf7 28.c3!? Nxb3+ 29.axb3

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

*****

*****

*****

*****

*****

Usually rook and bishop work better together than rook and knight, and here there is the added factor of White having two isolanis. But the kingside pawns are not easy for Black to get at, White occupies the open file, and White has good chances of finding the knight a good outpost square. The engines agree the position is equal.

29...g5 30.Ne2 Re8 31.Rf3 Kg6 32.Nd4 Bd7 33.b4 b6 34.Rg3 Kf6 35.Rf3+ Kg6 36.Rg3 Kf6 37.Rf3+ Kg6 ½–½

Christmas Is Coming II

Nativity scene, Gran Hotel Bali

Benidorm Bali Round Eight

Spanton (1896) - Miguel Ángel Fernández Alonso (1988)

Sicilian Bb5(+)

1.e4 c5 2.Nf3 Nc6 3.Bb5 e5!?

Carlsen, Caruana and Nakamura are among those who have played this, although it is only sixth-most popular in ChessBase's 2026 Mega database.

Nevertheless, the position occurs 2,558 times in Mega26

*****

*****

*****

*****

*****

4.0-0

There is no tactical refutation as, after 4.Bxc6 dxc6 5.Nxe5!? Qd4, Black is slightly better, according to Stockfish17.1 and Dragon1.

4...Bd6!?

This is the main continuation.

5.c3 a6 6.Ba4 b5 7.Bc2 c4?!

The engines strongly dislike this, preferring the mainline 7...Nge7.

8.Qe2!?

Apparently a novelty. The engines suggest 8.a4.

8...Nge7 9.Rd1 Bc5 10.b3 cxb3 11.axb3

Stockfish17.1 reckons 11.Bxb3!? is even stronger; Dragon1 fluctuates between the two moves.

11...Qc7?!

White gets a powerful initiative after this. Black should perhaps have taken the chance to castle.

12.d4 Bd6!?

This may be best.

13.Qd2?!

The engines reckon White has a positionally won game after 13.d5, eg 13...Nd8 14.Bd3!? Rb8 15.Be3 f6 16.c4 with strong play.

13...Ng6

This is the engines' second choice, but they much prefer 13...exd4.

14.dxe5

The engines still prefer pushing on with d5.

14...Bxe5

The engines prefer sacrificing the pawn with 14...Be7!? After 15.Nd4 White threatens Nxb5, so 15...Nxd4 16.cxd4 leaves White up a pawn, albeit the extra pawn is doubled. Black could avoid losing material with 15...Qxe5!? 16.f4 Qb8, but the engines reckon both 17.Nf5 and 17.Qf2 leave White on top.

15.Nxe5 Ncxe5 16.Ba3

Black has avoided losing a pawn, but White's bishops and pressure down the d file give a large advantage

*****

*****

*****

*****

*****

16...Bb7 17.Bd6 Qb6 18.Na3 f6!? 19.c4 Bc6 20.c5 Qb7 21.Re1 0-0-0?!

Much better, according to the engines, is 21...Nf7.

22.Qa5 Rde8 23.Rab1 Qa7

Again the engines suggest ...Nf7.

24.Ra1

Even stronger is 24.Nc4! as after 24...bxc4 25.bxc4 the threat of the b1 rook to invade along the b file cannot be parried.

24...Qb7 25.Nb1!? Nf7 26.Bg3!?

The engines agree the bishop is worth preserving.

26...f5?!

The engines suggest 26...h5 or 26...Ng5, but reckon White is winning.

How should White proceed?

*****

*****

*****

*****

*****

27.f3

Best is 27.exf5! True, after 27...Bxg2!? 28.fxg6?? Rxe1+ 29.Qxe1 Ng5 White is lost, but there are stronger alternatives at move 28, including 28.f3 and 28.Nd2! The engines reckon Black should instead play 27...Rxe1+ 28.Qxe1 Nge5!?, but agree White is winning after, for example, 29.Bxe5 Re8 30.f4 Bxg2 31.Nd2.

27...f4 28.Bf2 Ng5?!

The engines much prefer 28...Nge5.

29.Qxa6 Qxa6 30.Rxa6

Now Black has weathered the storm without getting mated, how would you assess the position?

*****

*****

*****

*****

*****

White is a pawn up, has the better pawn-structure, and still has the bishop-pair - winning, according to the engines.

30...Kb7 31.Rb6+ Kc7 32.Na3 Rb8

The engines suggest giving up a second pawn for activity, eg 32...Ra8!? 33.Nxb5+ Bxb5 34.Rxb5 Ne5, but agree White is still winning.

33.Rxb8 Rxb8 34.Bd3 Ne5 35.Be2 Ra8 36.Nc2

Possible is 36.Nxb5+!? Bxb5 37.Bxb5 N(either)xf3+ 38.gxf3 Nxf3+ 39.Kg2 Nxe1+ 40.Bxe1 as the two bishops should not have major problems overcoming the rook.

36...Ne6 37.Nd4!? Nxc5?

White to play and win

*****

*****

*****

*****

*****

38.Nxb5+??

White has at least two winning moves, the most convincing being 38.Nxc6 Kxc6 39.Rc1 etc.

38...Bxb5 39.Bxb5

Black to play and draw

*****

*****

*****

*****

*****

39...Ncd3?

After 39...Nxb3 Black is very solid, and with all the white pawns on one side of the board, the position is equal, according to the engines.

40.Rd1?

White is winning again after 40.Bxd3 Nxd3 41.Rb1, according to the engines, although it would not be an easy matter.

40...Nxf2 41.Kxf2 Ra2+ 42.Kf1 Rb2 43.Rc1+ Kb6 44.Bc4 Nxc4?!

The engines reckon the game is completely equal after moves such as 44...g5, 44...h6 and even 44...Kc4.

45.bxc4 Kc5?! ½–½

Correct, according to the engines, is 45...Rd2 or or 45...g6, when they reckon White has an edge, but is not winning.

After the text I accepted my opponent's draw offer, but the engines reckon White should play on, eg 46.Rd1 Kxc4 47.h4!? Rb7 48.Rd5, which they reckon wins - so much so that they give White's advantage as being worth a piece and a pawn.

Christmas Is Coming

The season's first balcony Santa

Benidorm Bali Round Seven

FACED a Malaysian junior (born 2010).

Spanton (1896) - Taj Prasshad Nambiyar Chanthiran (1518)

English Four Knights

1.c4 e5 2.Nc3 Nf6 3.Nf3 Nc6 4.e4!?

Nimzowitsch pioneered this move, and the associated pawn-formation with a grip over d5, but a hole at d4. Botvinnik is credited with refining the idea by playing e4 before developing the king's knight, which then went to e2, in conjunction with a kingside fianchetto.

How should Black respond?

*****

*****

*****

*****

*****

4...Bb4

This was a popular reply from the beginning, along with 4...Bc5, and has today established itself as the main continuation.

5.d3 h6!?

Nakamura and other grandmasters have played this, but more popular are 5...0-0 and especially 5...d6.

6.Be2 d6 7.Bd2 Bc5!?

A known idea, but apparently a novelty in this exact position. The move is liked by Stockfish17.1 and Dragon1, along with another apparent-novelty, 7...a5!?

8.0-0 0-0

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

*****

*****

*****

*****

*****

*****

White has more space on the queenside and has a grip over d5, but a hole at d4. Both white bishops are passive, while Black's dark-square bishop is technically bad, in that Black's centre pawns are on dark squares, but is outside the pawn-chain and is the most active piece on the board. The engines reckon the game is completely equal.

9.Na4 Qe7 10.a3 a5 11.Rb1 Ba7 12.b4 axb4 13.axb4 Nd4 14.Nc3 Nxe2+!?

Black wins the bishop-pair, but swops off White's bad bishop.

15.Qxe2 Bg4 16.h3 Bh5 17.Be3 c6

Maintaining the bishop-pair with 17...Bb8!? does not look appealing, and 17...b6? is simply bad after 18.Ra1 or 18.b5!?

18.b5 Bxe3

Black is slightly better after this, according to the engines, but they reckon 18...Bd4!? is stronger.

19.fxe3!?

It is hard to say which recapture is better - the engines are split on the matter - but I thought this was the less-drawish choice.

19...c5?!

Black is slightly better after 19...Nd7 or 19...Qd8, according to the engines, but they reckon the text leaves White slightly better.

20.g4 Bg6 21.Nh4!?

Black cannot exploit this tactically, because White threatens Nxg6, which in turn threatens Nxe7+, but they reckon White should be playing on the queenside with 21.Ra1 or 21.Qb2.

21...Bh7 22.Nf5 Bxf5 23.Rxf5 Nh7 24.Qf2 Ra3!?

Counterplay, but the engines reckon 24...g6!? gives complete equality, despite the hole at f6.

25.Nd5 Qd7 26.Rd1

Black to play and lose

*****

*****

*****

*****

*****

26...Rfa8??

Black is only slightly worse after 26...g6!? or 26...Qd8, according to Stockfish17.1, although Dragon1 awards White the upper hand.

27.Nb6 Qe7 28.Nxa8 Rxa8 29.Kh2 f6 30.Qf1 Nf8 31.Ra1 Rxa1!?

The engines are not overly critical of this, although swopping rooks when the exchange down is rarely a good idea.

32.Qxa1 Ne6 33.Qa8+ Kh7 34.Qc8

How should Black proceed?

*****

*****

*****

*****

*****

34...Nf4?!

This 'clever' combination wins back the exchange, but at the cost of allowing White to swop off into an easily won pawn ending. However, Stockfish17.1 reckons White's advantage is anyway worth almost a rook, and Dragon1 is even more effusive.

35.exf4 g6 36.fxe5 gxf5 37.Qxf5+ Kg8 38.Qg6+ Qg7 39.Qxg7+

The game finished:

39...Kxg7 40.exd6 Kf7 41.Kg3 Ke6 42.Kf4 Kxd6 43.Kf5 Ke7 44.Kg6 b6 45.h4 1-0

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

Benidorm Weather

THERE has been plenty of sunshine this year, but temperatures are lower than usual - low-to-mid-teens rather than high-teens to low-20s.

Early-morning sun on Finestrat cove

Benidorm Bali Round Five

FACED a junior (born 2012).

Spanton (1896) - Tarek Kaji Senas (1711)

Spanish Exchange

1.e4 e5 2.Nf3 Nc6 3.Bb5 a6 4.Bxc6 dxc6 5.d4 exd4 6.Qxd4 Qxd4 7.Nxd4 Bd6

This is fifth-most popular in ChessBase's 2026 Mega database.

How should White respond?

*****

*****

*****

*****

*****

8.Nc3

Capablanca played the aggressive 8.f4 in what seems to have been an exhibition game in 1911, but the move is not liked by Stockfish17.1 or Dragon1, and the game was drawn.

8...Ne7 9.Be3 0-0!?

Black often castles long in the Exchange Variation of the Spanish, but the text has a specific idea, namely to speedily fully open the centre for the benefit of Black's bishop-pair.

10.f3

10.f4 has been tried, intending to meet 10...f5 with 11.e5, but the engines reckon Black is slightly better after 10...f6 or 10...Re8.

10...f5 11.exf5

The engines suggest 11.0-0-0, 11.Bf2!? or 11.Bg5!?
11...Nxf5 12.Nxf5 Bxf5 13.0-0-0

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

*****

*****

*****

*****

*****

We have the usual battle in the Exchange Variation of the Spanish between White's superior pawn-structure and Black's bishop-pair. The engines reckon Black is slightly better, or at least has the better part of equality.

13...Rae8 14.Bd2 Rf7

The engines like 14...b5, eg 15.Rde1 Rxe1+ 16.Bxe1 Re8, with a slight edge for Black (Stockfish17.1) or even the upper hand (Dragon1).

15.Rde1 Rxe1+ 16.Bxe1 Re7 17.Bd2

The engines suggest 17.Bg3.

17...Bg6 18.h3

And here they prefer pushing the h pawn two squares.

18...Bg3 19.Rf1 b5 20.a3

The engines suggest 20.Kd1 or 20.b3!?

20...a5 21.Kd1 b4 22.axb4 axb4 23.Na4

How should Black proceed?

*****

*****

*****

*****

*****

23...b3!?

A pawn sacrifice to give the light-square bishop more activity.

24.cxb3 Bd3 25.Rg1 Re2 26.Nc3

Probably not 26.Nc5?!, which works fine after 26...Bf2?? 27.Nxd3, but Black instead has 26...Bb5, when 27.b4 Bh2 28.Re1 Rxg2 wins back the sacrificed pawn, and leaves White with four isolanis, two of which are doubled (and Black still has the bishop-pair). Similarly 27.Ne4 Bh2 28.Rf1 Rxg2 29.Rf2 Rg1+ 30.Kc2 also leaves White in a mess.

26...Rf2

How should White proceed?

*****

*****

*****

*****

*****

27.Ne4

Not 27.Be1?? Bc2+ 28.Kc1 Bf4+.

27...Bxe4 28.fxe4 Bh2 29.Re1 Rxg2

How would you assess this rook-and-bishop ending?

*****

*****

*****

*****

*****

White has four isolanis - two more than Black - but has the only passer. Black has a 2-1 flank majority on the kingside, which could prove powerful in a pawn ending. The engines reckon the game is completely equal.

30.Re2

Passed pawns should be pushed? Not always, even here where the passer is supported by a rook. After 30.e5?! Kf7 31.e6+ Ke7 the pawn is firmly blockaded, and 32.Bb4+? only makes matters worse after 32...Bd6.

30...Rxe2 31.Kxe2 Be5

What should White play?

*****

*****

*****

*****

*****

32.Kf3!?

The engines reckon 32.Bc1!? also draws, but I was afraid of getting into zugzwang. Definitely bad is 32.Bc3? as 32...Bxc3 33.bxc3 leaves Black with an easily won pawn ending - the white king would have to abandon the centre (and the queenside) to deal with Black's coming kingside passer.

32...Kf7

Or 32...Bxb2 33.Bf4 etc.

33.Kg4 Ke6 34.Bf4!?

The pawn ending is drawn as long as Black cannot exchange bishops and immediately occupy e5 with the king, as would be the case after 34.Bc3?

34...Bxf4 35.Kxf4 c5 36.h4 g6

All reasonable moves leave the position completely equal, according to the engines.
37.Kg5 Kxe5 38.Kh6 Kxe4 39.Kxh7 Kf5 40.Kg7 g5 41.hxg5

After 41.h5, White queens first, but Black queens with check, and the position remains drawn.

41...Kxg5 42.Kf7 Kf5 43.Ke7 Ke4 44.Kd7 Kd3 45.Kc6 Kc2 46.Kxc5 Kxb3 ½–½

Benidorm Bali Round Four

AFTER taking a halfpoint bye in the first of today's two rounds, I faced an Irishman.

Pat O'Hanlon (1794) - Spanton (1896)

London System

1.d4 d5 2.Bf4 e6 3.e3 Bd6 4.Nf3 Bxf4

This is the most popular continuation in ChessBase's 2026 Mega database.

5.exf4 Qd6 6.Qd2 Nf6 7.Nc3 Bd7!?

This apparent-novelty is Stockfish17.1's top choice at first, although after a while it comes to view the text and 7...0-0 as of roughly equal value. Dragon1 starts by approving of castling, but comes to prefer the text, although later it switches back to castling and to 7...a6.

8.Bd3

The engines suggest 8.0-0-0!?

8...0-0 9.0-0

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

*****

*****

*****

*****

*****

The half-open e file and control of e5 are White's compensation for having doubled f pawns. The engines reckon the game is equal.

9...c5 10.dxc5 Qxc5 11.Ne5!?

By transposition we have reached the game Nicolas Eliet (2422) - Alexei Barsov (2531), Bogny-sur-Meuse (France) 2003, which continued 11...Rd8 12.Rfe1 Nc6 13.Rad1 Be8!? 14.Ne2 Nd7 (14...Ne4!? is suggested by the engines) 15.g3 g6 16.h4, at which point a draw was agreed, although the engines reckon White is at least slightly better.

11...Nc6 12.a3 d4!?

How should White respond to this pawn thrust?

*****

*****

*****

*****

*****

13.Nxd7

This, by a small margin, is the engines' top choice, but they are also happy with other moves of the knight, including 13.Ne4 and 13.Na4.

13...Nxd7 14.Ne4 Qe7 15.Rfe1 Nf6 16.Nxf6+

White probably has a slight edge after 16.f5, according to the engines, eg 16...e5 17.Qe2 Nxe4 18.Bxe4, although Stockfish17.1 is less sure than Dragon1.

16...Qxf6 17.Be4 Rac8 18.Rad1 Rfd8 19.g3 g6 20.Qe2 Rc7 21.Bd3 Rdc8 22.h4 h5 23.Qe4 a6!? 24.Rd2 Kg7 25.Kg2 Re7 26.Rde2 Na5 27.b3!? Nc6 28.Qf3 Rec7 29.Qe4 Qe7 30.b4

The engines suggest 30.a4!?, although this gives up the b4 square.

30...Qd6 31.Qf3

Here they suggest 31.Rd2 or 31.b5!?

31...Ne7 32.Re5 Nd5

After much manoeuvring by both sides, how should White proceed?

*****

*****

*****

*****

*****

33.g4?!

A sharp try, but it may be a mistake.

33.f5!? is possibly best-met by 33...gxf5, when both 34.Qxh5 and 24.Rxd5!? lead to sharp play that seems roughly balanced.

33.Rg5 Nf6 34.f5!? exf5 35.Qxf5 also gives equal chances, according to the engines.

How should Black respond?

*****

*****

*****

*****

*****

33...f6

Best may be 33...hxg4 34.Qxg4 Rxc2! 35.Bxc2 Rxc2 36.f5 Ne3+! 37.R5xe3 dxe3 38.f6+!? Kxf6! 39.Rxe3, after which Black emerges a pawn up, but the line is very sharp and certainly hard to envisage from the diagram.

34.R5e4

Not 34.Rxe6?? Nxf4+ etc, but Stockfish17.1 suggests 34.Rxd5!?, claiming 34...Qxd5 35.Qxd5 exd5 36.f5!? gxf5 37.Bxf5 gives White full compensation for the exchange, although Dragon1 is not wholly convinced.

34...hxg4 35.Qg3

Not 35.Qxg4?? f5 36.Rxe6 Qxe6! (this is even stronger than 36...Nxf4+, as after 37.Kf1 Nxe6 White has 38.Bxf5 Nf4 39.Bxc8) 37.Rxe6 fxg4 38.Rxg6+ Kf7 39.Rxg4 Rg8.

35...f5

Possibly 35...Qc6 is stronger.

36.Re5?

Correct is 36.Rxe6!, when 36...Nxf4+?? fails to 37.Kf1. However Black has 36...Ne3+! 37.R1xe3 Qd5+, eg 38.Kg1 dxe3, although after 39.Qxe3 White has enough compensation for the exchange, according to the engines. Dragon1 slightly prefers 38.R3e4?, but Stockfish17.1 shows this is wrong, thanks to the long line 38...fxe4 39.Re5 (39.Rxe4 Re8) exd3+! 40.Rxd5 dxc2 41.f5 c1=Q 42.Qe5+ Kh6, and only now does Dragon1 realise White does not have a draw as 43.fxg6 can be met by 43...Qc6, eg 44.Kg3 d3! 45.Qg5+ Kh7 46.h5 Re8 47.h6+ Kg8 48.Qf5 Rce7, after which the Black king is safe enough, and Black is up a rook for at most two pawns.

36...Rc3

Even stronger, according to the engines, are 36...Re7 and 36...Rxc2!? 37.Bxc2 Rc3, when 38.Bd3 runs into 38...Rxd3! 39.Qxd3 Nf4+ and 40...Nxd3.

37.Kg1

The engines more-or-less come to agree this is the best try, as after 37.Rxe6 Ne3+! 38.R1xe3 Qd5+ 39.Kh2 dxe3 40.Qxe3 they reckon White does not have enough for the exchange, especially if Black finds 40...Rh8!

37...Rxa3?

Probably best is 37...Re8.

38.Rxe6 Qxf4

White to play and equalise

*****

*****

*****

*****

*****

39.Qg2

The engines give 39.Rd6! Qxg3+ 40.fxg3 Nxb4 41.Ree6 Rc7 42.Rxg6+ Kf8 43.Bxf5 Rxg3+ 44.Kh2 Rh3+ 45.Kg1 Rxh4, when Black is three pawns up, but White seems to have enough for a draw, eg 46.Rd8+ Ke7 47.Rxd4 Nc6 48.Rf4 Rh8 49.Re6+ Kd8 50.Rd6+ Ke8 51.Rxg4, after which White has won back two pawns and remains very active. Black could instead try 40...Ne3, but 41.Rxd4 Nxc2 42.Rd7+ Kh6. However 43.Ree7 Rh8 (only move) 44.Kh2 Nxb4 45.Bxf5! gxf5 46.Rd6+ Kh5 47.Rg7 forces Black to take a draw by repetition by 47...Ra2+ 48.Kg1 Ra1+ etc. Naturally there are many alternatives in these lines, but 39.Rd6! seems to hold up in all of them.

After 39.Qg2, it is Black to play and win

*****

*****

*****

*****

*****

39...Qf3?

Winning is 39...g3, eg 40.Qxg3 Qxg3+ 41.fxg3 Nxb4, when Black is two pawns to the good, and the black king does not face serious threats. Trickier is 40.fxg3 as Black is losing unless he finds 40...Rxc2! 41.gxf4 Rxg2+ 42.Kxg2 Nxf4+ and 43...Nxe6.

40.Qh2?

After 40.Qxf3 gxf3 41.Rd6 Nf4 42.Kh2 White is two pawns down, but seems to have enough activity to hold, eg 42...Rc7 43.Re5 Kf7 44.Rxd4 Nh5 45.Rd6. But the position is far from easy to play, for example winning back a pawn by 42.Rxd4? (instead of playing 42.Kh2) can be met by 42...Rxd3!, when 43.cxd3? fails to 43...Ne2+ etc, and 43.Rxd3 Nxd3 44.cxd3 gives a rook-and-pawn ending in which White is 'only' a pawn down, but every white pawn is isolated, and Black's advantage is equivalent to being about a rook up, according to the engines.

40...Nf4 41.Re7+ Kh6 42.h5 Nh3+ 43.Kf1 Rxd3!

This is strongest.

44.cxd3 Rc2

Other moves also win.

45.R7e2 Qxd3 46.hxg6 f4 0-1

Benidorm Bali Round Two

Spanton (1896) - Enrique Martínez Pérez (1726)

Modern/Pirc

1.Nc3 g6 2.e4 Bg7 3.d4 c6 4.f4 d6 5.Nf3 Nf6 6.Be2 0-0 7.0-0

With both sides castling early, how would you assess the position?

*****

*****

*****

*****

*****

White has more space and a slight lead in development, which together give the upper hand, according to Stockfish17.1 and Dragon1.

7...a5!?

This is Stockfish17.1's top choice, for a while, and is anyway preferred over the much more popular 7...b5, which the engines agree is well-met by 8.e5. Dragon1 also prefers the text over 7...b5, but to a lesser degree, and eventually both engines settle on 7...Nbd7!? as best.

8.a4 Qb6 9.Kh1 Na6 10.Ra3!? Bd7 11.Rb3 Qc7 12.Be3 Ng4 13.Bf1 Nb4 14.Qd2 d5

The engines dislike this, preferring 14...Rad8 or 14...Nf6.

How should White proceed?

*****

*****

*****

*****

*****

15.h3

The engines much prefer 15.e5.

15...dxe4!

This equalises, according to the engines.

16.Nxe4 Nf6 17.Nc5 Bc8

The engines give 17...Bf5!, the point being 18.g4 can be met by 18...Ne4 (not 18...Bxc2?? 19.Rxb4 etc) 19.Nxe4 Bxe4, when they reckon Black is at least slightly better. White should probably play 18.c3, when 18...b6 19.cxb4 bxc5 seems equal.

18.c3 Nbd5 19.f5!?

White gets reasonable compensation for this pawn sac, but I probably should have gone for my other main thought of 19.c4!?, and after 19...Nb4, best may be 20.Ne3.

How should Black respond?

*****

*****

*****

*****

*****

19...b6!?

This zwischenzug is Black's best chance for an edge.

20.Bh2 Qd8 21.Nd3

21.fxg6!? is probably asking too much of the position after 21...bxc5, eg 22.gxh7+ (the engines reckon this is is better than taking on f7) Nxh7 (22...Kh8!?) 23.dxc5 Bf5!? 24.Ne5 Be4!?, with a complicated position that the engines reckon favours Black.

21...Bxf5 22.Nfe5 Ne4!? 23.Qc1 Qc8?!

The engines suggest 23...Rc8 or 23...f6, claiming a slight edge for Black.

24.c4 Nb4 25.Nxb4 axb4 26.g4!? Be6 27.Bf3 Bxe5!?

This looks risky, but Dragon1 reckons it is best. Stockfish17.1 also likes the text, but reckons 27...f6 is as good.

28.Bxe5 Nd6 29.Qh6 f6 30.Bxd6 exd6 31.d5!?

31.Rxb4 regains the pawn and gives complete equality, according to the engines.

31...cxd5 32.cxd5?

Capturing on b4 is best.

Black to play and win

*****

*****

*****

*****

*****

32...Bf7?

Missing 32...Qc4!, forking the white rooks.

33.Rfc1 Qd7 34.Rxb4 Ra5

There seems nothing wrong with 34...Rxa4.

35.Qd2 Rc8?!

Capturing on  a4 is still playable, and is better than the text.

36.Rxc8+

Possibly stronger is 36.Rc6!?, with a slight edge, according to Dragon1, although Stockfish17.1 reckons the position is equal.

36...Qxc8 37.Rxb6 Rxa4

How would you assess this late-middlegame?

*****

*****

*****

*****

*****

The engines reckon the position is completely equal.

38.Rc6

Not 38.Rxd6? Ra1+, eg 39.Kg2 Qc5 40.Rd8+ Kg7 41.Qf2 Qc7, and White's rook drops, or 39.Bd1 Qb8! 40.Rxf6 Qb3 etc.

38...Qb8 39.Qc3?

Probably best is 39.Kg2, with equality, according to the engines.

Black to play and win

*****

*****

*****

*****

*****

39...Kg7?

Missing 39...Ra1+, followed by 40...Qa7, eg 40.Kg2 Qa7 41.Kg3 Qg1+ 42.Bg2 Re1 etc.

40.g5 Qd8?

Black has to play 40...Rf4, according to the engines, and after 41.gxf6+ probably has to continue with 41...Kh6!?, as 41...Rxf6 42.b4 gives White a dangerous queenside initiative.

White to play and win

*****

*****

*****

*****

*****

41.gxf6+

This wins a pawn, but not the game, whereas 41.b4! wins comfortably, according to the engines, as Black cannot cope with pressure against f6 and d6 as well as the advancing passed pawn.

41...Qxf6 42.Rxd6 Qd4!

There is no win after this.

43.Kg2 Rb4 44.Kg3 Rc4 45.Qxd4 Rxd4 46.Rd7 Kf6 47.b3 h5 48.h4 Rb4 49.Bd1 Rd4

EMP offered a draw in my time.

50.Bc2 Rxd5!?

The engines do not like this because they do not understand the coming bishop-and-pawn ending. Stockfish17.1 reckons the move gives White the upper hand, while Dragon1 reckons White's advantage after the text is worth more than a rook, but they are both wrong.

51.Rxd5 Bxd5 52.b4

This allows complete equality, but so does 52.Kf4 as Black has 52...Bxb3!, which the Syzygy endgame tablebase confirms is a draw (White's bishop is on the wrong colour complex for promoting an h pawn).

52...g5 53.b5 Bb7 54.Bd1 Kg6 55.Bf3 gxh4+ 56.Kf2 Bc8 57.b6 Kg5 58.Kg2 h3+ 59.Kh2 h4 60.Bd5

EMP offered a draw.

60...Kf5 61.Bc6 Ke5 62.Bd7 Bb7 63.Bxh3 Kd6 64.Bg2 ½–½