Lessons From Mariánské Lázně: Look At The Whole Board

IT is easy in the heat of battle to focus on the one part of the board where all the action seems to be.

Position in my round-six game after I played 47.Kg1-h2

*****

*****

*****

*****

*****

*****

*****

*****

*****

My opponent played 47...Qf4+ and offered a draw, which I accepted.

In some ways that was understandable as he is rated 218 Fide elo lower, and I was threatening both his h and g pawns.

But winning, according to Stockfish15 and Komodo13.02, is 47...Qxa3.

This does not win a pawn - White replies by capturing on h4 or g7 - but it wins because Black gets a fast-moving passed b pawn while White does not have a passer.

A sample continuation is 48.Bxh4 Qe7 49.Bf2 b5 50.Qg3 (not 50.g4? as 51...Qe4 leads to mate) b4 etc.

Lessons From Mariánské Lázně: Loose Pieces Drop Off

GRANDMASTER John Nunn popularised the phrase "loose pieces drop off," sometimes abbreviated to LPDO.

Position in my round-five game after Black captured on d5

*****

*****

*****

*****

*****

I was thinking about Nunn's aphorism as I considered my reply in this game, but that did not stop me playing 16.Bg5?! as I felt it was important to prevent a black rook going to d8, and at the same time I was developing a piece.

Stockfish15 and Komodo13.02 reckon getting the queen off the e file, while also attacking d5, is better, which is why they suggest 15.Qf3.

After my move the game continued 16...h6 17.Bh4 Rfe8, at which point I probably should have played 18.Qb5 or 18.Qb3.

Instead I chose 18.Qd2, which gives the white knight a second defender, but Black took advantage of the loose bishop with 18...Qg4.

I unloosed the bishop, as it were, by playing 19.Bg3?, only to be hit by 19...Bxd4 20.Qxd4 Re1+ 0-1.

Better were 19.h3!? Qxh4 20.Nf3 Qb4 21.Qxd5 Re2 and 19.Nf3 Nf4 20.Bg3 Ne2+ 21.Kh1 Qc4, although in both cases the engines give Black a large advantage.

Lessons From Mariánské Lázně: Beware Complacency

IN round four I was under pressure for much of the game before swopping off into a fairly even ending.

Position after White played 35.Nc5-e6

*****

*****

*****

*****

*****

White has the better minor piece for cooperating with a queen, but I felt my king was safe behind a barrier of black pawns.

In fact the king is technically in front of the barrier, but is behind the pawns from the viewpoint of the white queen and knight.

What I failed to properly consider is that the king only has three vacant squares to which it can potentially move - g7, g5 and h5 - and two of these are covered by the knight.

When a king has only one move, and that move would take it away from its pawn-cover, alarm bells should be ringing.

Unfortunately for me the bells only rang after I played 35...Qb4?? - several moves draw including, as my opponent pointed out, 35...a5 - and I had to spend a miserable minute or so waiting for White to double-check that Black cannot avoid mate after 36.Qd8.

Lessons From Mariánské Lázně: Keep A Cool Head

GRANDMASTER David Bronstein once stated: "To lose one's objective attitude to a position nearly always means ruining your game."

I have not managed to find where he said this - it may be from his book on the 1953 candidates'  tournament - but I do believe many chess players fall into one of two  camps: the unreasonably optimistic and the unreasonably pessimistic.

I am probably more often in the former camp, and I think my 2066-rated opponent in round three at Mariánské Lázně was the same.

Position after I played 12...Rff7!?

*****

*****

*****

*****

*****

White has a promising-looking attack, but there appears to be nothing immediately decisive.

Probably best is 13.exf6, but 13...Bxf6 leaves Black with a solid position, although Stockfish15, but not Komodo13.02, gives White a slight edge.

The game saw the aggressive 13.b4?!, which is strongly disliked by the engines, presumably because it drives the knight to a square from which it is better placed to defend the black king, and because it also weakens the white queenside.

After 13...Ne6 White should probably develop with 14.Nd2, although the engines reckon both 14...fxe5 and 14...a5 give Black at least a slight edge.

Instead White played 14.f4, the idea being to meet 14...fxe5 with 15.f5!?

However after 15...gxf5! - clearly best, according to the engines - White did not have enough for two pawns.

He continued 15.Qxe5? - better is 15.Bxf5 - but 15...Bd6 16.Qe1 Qf6 17.Be3 Qe5 left White with little option but to play 18.Qg3+, after which queens came off and Black emerged a pawn up.

To reiterate: objectivity is an important chess skill that helps a player find the correct plan, or at least makes it less likely a player will go seriously wrong.

Central London League

PLAYED for Battersea against Pimlico Knights on board three (of five) in division one last night.

Federico Bastianello (2002) - Spanton (1974)

Nimzowitsch-Larsen

1.b3

Pushing the b pawn one square on move one is rarer than one might imagine from its reputation as a 'nuisance' opening. I have faced it 19 times (ignoring casual games), and that number would be a lot lower if I had not had black against Ian Calvert on six occasions. My score of +5=6-8 is a not-very-good 42%. When a sample size is small, I think it is often more relevant to use a merit-points system I devised for comparing opening performance over many years when ratings fluctuate. You score a merit point if you beat someone higher rated, or draw with someone at least 160 elo higher, and a demerit if you lose to someone lower rated, or draw with someone at least 160 elo lower. On those rare occasions when my opponent has the same rating as me, both a merit and a demerit are possible. By this system my score against 1.b3 is 3-4, which again is nothing to shout about.

Black has a wide choice of replies to 1.b3, but I have come to believe there are three main plans:

A) Blockade. Black plays 1...e5 - the most popular move in ChessBase's 2023 Mega database - and sets out to blunt the white dark-square bishop, which can be expected to go to b2. This is similar to a common plan against White's kingside fianchetto, ie to play ...d5 and ...c6. A major difference is that ...f6 compromises Black's kingside, and indeed ...f6 is rarely part of Black's plan. The mainline in Mega23 runs 1...e5 2.Bb2 Nc6 3.e3 Nf6 4.Bb5 (fighting for the e5 square) Bd6!? 5.Na3 Na5!? 6.Be2!? a6 7.0-0 with an unclear and strategically complicated position.

B) Counterplay. White's first move is about dominating the central dark squares, ie d4 and e5. Black can, as it were, ignore this and play to dominate the central light squares, ie d5 and e4. This is the plan I adopted in the game.

C) Symmetry. Black copies White's moves until the opportunity arises to vary favourably. This is based on the idea that moving second can be advantageous in chess since the opponent has to commit first. It is not very practical to try this against 1.e4 or 1.d4, but it is reasonably popular against the English, and can also be used against the Nimzowitsch-Larsen. The mainline of this system in Mega23 goes 1...b6 2.Bb2 Bb7 3.e3 e6 4.Nf3 Nf6 5.Be2 Be7 6.0-0 0-0 7.c4 c5 8.d4, and now Black usually varies with 8...cxd4, meeting the main recapture 9.Nxd4 with 9...a6!?, although Stockfish15 and Komodo13.02 reckon White has a slight edge. Note that Black can maintain the symmetry for at least one more move with 8...d5!?

1...d5 2.Bb2 Nf6 3.e3 Bf5 4.Nf3 Nbd7 5.Nh4!?

The main move in Mega23 is 5.Be2, but the engines prefer the text, with Stockfish15 claiming it gives White a slight edge (Komodo13.02 disagrees).

How should Black reply?

*****

*****

*****

*****

*****
5...Bg6

All four games to reach the diagram in Mega23 saw the text played, but there is something to be said for 5...Bg4!?, as will soon become clear.

6.f4!?

Two of the four players in Mega23 played this. The engines prefer winning the bishop-pair with 6.Nxg6, meeting 6...hxg6 with 7.g3!?, while the strongest of the four players, a 2440, effectively ignored what was happening on the kingside by continuing 6.c4!?

6...Bh5!?

By no means forced, especially as after ...

7.Be2 Bxe2 8.Qxe2

... White has gained a tempo (6.f4!?) on the line 5...Bg4!? 6.Be2 Bxe2 7.Qxe2. However it is by no means certain the tempo is a useful one, although playing f4, to help control e5 and possibly half-open the f file, is thematic in the Nimzowitsch-Larsen.

8...e6

There are 15 games with this position in Mega23, and in all but one of them it is Black to play

*****

*****

*****

*****

*****

9.d3 Bd6 10.Nf3!?

This stops a quick ...e5, but that is probably not a big threat, eg 10.0-0 e5?! 11.Nf5 justifies White's flanking manoeuvre with the king's knight.

10...Qe7 11.Ne5 Bb4+!?

How should White meet the check?

*****

*****

*****

*****

*****

12.Nd2

Stockfish15 marginally prefers 12.c3, although Komodo13.02 reckons the text gives White a slight edge.

12...Ba3 13.Bxa3 Qxa3 14.Ndf3 Qb4+ 15.Qd2 Qxd2+ 16.Kxd2 Nxe5 17.Nxe5 Nd7 18.Nxd7 Kxd7

After an interesting opening, the game has quickly liquidated to an unusual ending with all the rooks and pawns still on the board, but nothing else (apart from kings)

*****

*****

*****

*****

*****

There are 3,221 examples of this ending in Mega23, meaning it occurs 0.03% of the time. Usually, as far as I can tell from looking at sample examples, the ending arises with at least one half-open file. The absence of such a file makes a draw more likely.

The game finished:

19.g4 h5 20.g5 g6 21.h4 a5 22.a4 Kd6 23.c4 Ra6 24.Kc3 c5 25.e4 dxe4 26.dxe4 Rd8 27.Rad1+ Ke7 28.Rxd8+ Kxd8 29.Rd1+ Ke7 30.e5 Rc6 ½–½
Pimlico Knights won the match 3.5-1.5

Advantage White?

THE next round of matches in the Central London League is due to be played tonight at the league's community-centre venue in Pimlico.

Scheduled are two matches in division one, two in division three and one in division four.

Although division four is played over four boards, the other divisions are five-a-side, which makes winning the toss more important than usual.

But how much more important? Or, to put it another way, should the outcome between two evenly matched teams be 3-2 or 2.5-2.5?

As a ballpark figure, whites on average across large databases score 55%, and blacks score 45%.

So the side winning the toss is expected to score 3 x 55% and 2 x 45%, while the other team is expected to score 2 x 55% and 3 x 45%.

That means the team with white on odd boards is expected to beat equally strong opponents by a score of 2.55-2.45.

In other words, assuming my maths are correct, winning the toss in a five-board match is a lot less important than many of us think.

Lessons From Mariánské Lázně: King Safety

GRANDMASTER Nigel Davies is fond of emphasising that the most important factor in chess is king safety.

In some ways this is a truism since the only way to win a game of chess, short of the opponent resigning, is to deliver checkmate - 'merely' capturing every enemy piece, apart from the king, is not enough.

Most decisive games actually end with resignation, but only because the player resigning thinks checkmate is inevitable.

Games between complete novices, however, nearly always end with one of two results - checkmate or stalemate.

Draws in such games are theoretically possible, but rarely happen before the board is reduced to bare kings.

Even bare kings may not be enough, if a story told me by an arbiter at the British championships is true.

He reckons he saw a game in which two very young juniors got down to bare kings, but continued playing.

The arbiter left them to it and went to look at other games, only to return a few minutes later when he saw the boys resetting the pieces.

"That was a a draw, then?" the arbiter half-said/half-asked.

"Oh no," replied one of the boys, "my opponent managed to promote his king to a queen and so won."

The story is probably apocryphal - after all, how to you mate with a lone queen? - and anyway is something of a digression from the point of this post, which about king safety.

The critical position in my round-two game occurred at move 12.

I have just played 12.d5!?

*****

*****

*****

*****

*****

12...e4?!
Black is roughly equal, or at least only slightly worse, according to Stockfish15 and Komodo13.02, after 12...Bxf3 13.Bxf3 Nd4 or 12...Ne7.

The text wins the exchange after 13.Nxe4 Bxa1 but loses the game thanks to the chronic weakness of the black king, now it no longer has a fianchettoed bishop.

The game lasted a further 20 moves, but the result, at least according to the engines, was never in doubt.

Lessons From Mariánské Lázně: Many A Slip 'Twixt Cup And Lip

ONE of the features that makes chess so exciting, at least to those who play the game, is the same thing as makes it so frustrating - one careless move after hours of play can turn a win into a loss.

My opponent in round one of the Seniors 50+ tournament suffered in this way.

Position after I played 22...cxd4

*****

*****

*****

*****

*****

*****

*****

Material is arguably level - I write 'arguably' because the bishop is probably better than the knight, although I do not believe that is completely clear.

But the key to the position is the weakness of the d pawn.

White cannot immediately capture it as the white queen is attacked, so clearly White must move the queen with the intention of picking up the pawn next move.

Both 23.Qd2 and 23.Qd1 are good for this purpose.

True, Black could reply 23...Rd8, when 24.Bxd4?! is a little problematic as it walks into a pin.

Stockfish15 and Komodo13.02 reckon 24.Bxd4?! is playable, but much better, according to them, is 24.Rd3, after which White will have no problems grabbing the pawn.

However, the game saw 23.Qa1?, which at first glance may seem good as it also avoids a pin and sets up 24.Bxd4.

But the queen, although it looks powerful on the long dark diagonal, has been drastically decentralised.

After 23...Nd5 the game has gone from winning for White (Stockfish15), or at least the upper hand for White (Komodo13.02), to winning for Black (Stockfish15), or at least the upper hand for Black (Komodo13.02).

White has no good way to defend f4, eg 24.Rf3 is met by 24...Rc2, and if 25.Bxd4 then 25...Qc7 is decisive as 26.Bb2 is met by 26...Qc6, when Black threatens 27...Nxf4! Similarly 26.Rf1 fails to 26...Qc6.

For a short while Komodo13.02 reckons 24.f5 equalises, but again 24...Rc2 is very strong.

Another try is 24.Qf1, but the engines reckon neither 24...Qc5 nor 24...Ne3 can be adequately met.

The game continued 24.Bxd4 Nxf4 25.Qf1 (there is nothing better) Rc1! 0-1

The lesson from this game is one we all know but most of us find hard to follow: keep concentrating to the very end, particularly looking out for the opponent's threats, not just your own.

Borrespondence Chess?

IN my experience most chess players' views of correspondence chess fall into three groups.

There are those who wrongly think engines are banned, and that such a ban cannot be enforced.

The latter would be correct, except that engines are allowed in most forms of correspondence chess, for example at the Fide-recognised International Correspondence Chess Federation.

The second group is composed of people who know engines are allowed, but think their use makes the whole exercise boring and pointless

The third group is those who play the game, engines and all, and this group includes me as I play at the ICCF, mainly for various England teams.

It is true that the vast majority of games finish as draws, and I rarely post them on this blog.

But occasionally a game avoids such a fate, or has some other point of interest.

Juan Borges Medina (Cuba - 2246) - Spanton (England - 2280)

Carlos Flores Gutiérrez Team Tournament Final

Board Five

Spanish Berlin

1.e4 e5 2.Nf3 Nc6 3.Bb5 Nf6 4.d3 Bc5 5.0-0 Nd4 6.Nxd4 Bxd4 7.c3 Bb6 8.d4!?

The main moves in ChessBase's 2023 Mega database are 8.Nd2 and 8.Bg5, which are almost exactly equally popular although the former scores 13 percentage points higher. The text has been played by grandmasters, but is roughly half as popular as the mainlines, although its score of 60% is 10 percentage points better than 8.Nd2.

8...0-0

White has a large advantage after 8...Nxe4? 9.Re1, according to Stockfish15 and Komodo13.02.

9.a4!?

The engines like this obscure move - only one example in Mega23 - and 9.dxe5.

9...a6 10.Bd3 d6 11.a5 Ba7 12.Bc2 Re8

How should White proceed?

*****

*****

*****

*****

*****

13.d5!?

This may be an improvement on Ghazal Hakimifard (2296)  - Frank Satzberger (2131), Swiss Championship (Samnaun) 2022, which saw 13.Re1, when the engines reckon 13...d5 may give Black a slight edge.

13...c6 14.h3 Bd7 15.c4 cxd5 16.cxd5 Rc8 17.Nc3 h6 18.Bd2 Nh7!? 19.Ba4 Rf8 20.Rc1 f5 21.Bxd7 Qxd7 22.Be3!? Bxe3 23.fxe3

Now all bishops have been exchanged, how would you assess the position?

*****

*****

*****

*****

*****

White has more space on the queenside but Black has a slight kingside initiative. The position is completely equal, according to the engines.

23...f4 24.exf4 Rxf4 25.Rxf4 exf4 26.Qf3?!

The engines prefer 26.Ne2, continuing 26...Rf8 27.Nd4 Ng5 28.h4!? Nxe4 29.Qf3, claiming White has full compensation for a pawn.

26..Rf8 27.h4!?

Not so much preventing 27...g5, rather 27...Ng5. However, better may be 27.Rf1, and if 27...Ng5 the engines reckon 28.Qh5!? (threatening 29.h4 Nh7 30.e5 with an initiative) Qc7 29.Qg4 Qxa5 30.h4 Qb6+ 31.Rf2 Nh7 32.Qe6+ Kh8 33.e5 again gives White good compensation for a pawn.

27...Qd8 28.Qg4 Qxa5 29.Qe6+ Rf7 30.Qxd6

How would you assess this position?

*****

*****

*****

*****

*****

White has equalised material and has a pair of dangerous-looking passed pawns. On top of this, the black king seems a little vulnerable, but the key to the position is the white king, which really is vulnerable.

30...Rf6 31.Qb8+ Nf8 32.Rf1!?

Not 32,Qxb7? as Black has 32...Qc5+ with a big attack after both 32.Kh2 Qf2 and 32.Kf1 f3. However perhaps better is 32.Qa7, although 32...Rb6 followed by ...Rxb2 is good for Black, and if 33.e5 then 33...Qc5+, eg 34.Kf1 (best, according to the engines) f3! 35.gxf3 Qe3 36.Rc2 Ng6 gives a strong attack, one line running 37.e6 Nf4 38.e7 Kf7 39.Qb8 Qxf3+ 40.Ke1 Qh1+ 41.Kd2 Qh2+ 42.Kd1 Qg1+ 43.Kd2 Qd4+ 44.Ke1 Kxe7, after which the black king can escape checks from the white queen, but the white king has no shelter.

32...Qb6+ 33.Rf2 Qe3 33.h5 b5 34.Qe5

How should Black proceed?

*****

*****

*****

*****

*****

34...f3!

A pawn sacrifice that gives the black rook a half-open f file and the knight a fine square at f4, while also reducing the white king's cover.

35.gxf3 Ng6! 36. Qb8+

Even worse is 36.hxg6 Qe1+ 37.Kg2 Rxg6+ etc. Best, but losing, may be the engines' 36.Qxf6!? gxf6 37.hxg6.

36...Rf8 37.Qg3 Nf4 38.d6 b4 39.d7?!

This is hopeless, but 39.Nd5 Qe1+ 40.Kh2 Nxd5 41.exd5 Rf5 also wins easily for Black.

39...bxc3 0-1

Mariánské Lázně Game Nine

Manfred Korth (1980) - Spanton (1895)

English Symmetrical

1.c4 c5 2.Nc3 Nc6 3.g3 g6 4.Bg2 Bg7 5.e4!? d6 6.Nge2

Position after 6.Nge2

White is playing what is known as Botvinnik's set-up, ie with a hole at d4, the king's knight at e2 and a fianchettoed light-square bishop. This is a refinement of a set-up Aron Nimzowitsch liked, which was with the king's knight on f3 instead of e2 and the light-square bishop sometimes developed to e2 rather than g2. It is mildly controversial in this position as, according to Tony Kosten, who dedicated much of a book to the subject, Botvinnik's system works best when Black has played ...e5. Andrew Soltis, another English Opening expert, reckoned Botvinnik's system works best when Black has played ...Nf6. Black has done neither in the diagram, and according to Komodo13.02 is slightly better, although Stockfish15 calls the position equal.

6...e6 7.0-0 Nge7 8.d3 0-0 9.Rb1 f5!?

The mainline in ChessBase's 2023 Mega database runs 9...b6 10.a3 Bb7 11.b4 Qd7 12.Be3 Nd4 13.Qd2 with a position the engines reckon is roughly equal, but favours, if anyone, Black.

10.a3 a5 11.Be3 Nd4

This is Black's normal reaction in this system once White has placed the dark-square bishop on e3 (with the queen's knight on c3). However, it lets White reply 12.b4, which may be why the engines prefer 11...fxe4 12.dxe4 e5!?, after which the backward d pawn can be masked by ...Nd4.

12.b3!?

This seems somewhat passive. A normal continuation is 12.b4 axb4 13.axb4 with at least a slight edge for White, according to the engines.

12...Nec6 13.Qc1!?

This supports Bh6 while protecting c3. 13.Qd2 does the same, but then White has to worry about the consequences of a later ...Nf3+.

13...Rf7 14.h4!?

The engines are not keen on this, although they disagree as to what White should play. One interesting suggestion by Stockfish15 is 14.Nxd4!?, the point being 14...cxd4 does not win a piece as White has 15.Bg5.

14...Qf8 15.Nf4?!

The engines strongly dislike this, preferring, for example, 15.Qd2!? with what they assess as an equal position.

15...fxe4

This is fine, but probably stronger is the engines' 15...e5 16.Nfd5 f4, and if 17.gxf4 then Black has 17...Qd8!?

How should White recapture?

*****

*****

*****

*****

*****

16.Bxe4!?

This is the engines' top choice. Also OK is 16.dxe4, but 16.Nxe4? loses to 16...Rxf4! thanks to the fork threat 17....Ne2+.
16...e5 17.Bd5??

This combination is bad materially and even worse positionally. After ...

17...exf4 18.Bxf7+ Qxf7 19.Bxf4

... White has rook and pawn for bishop and knight, an exchange that slightly favours Black from a material view if other things are equal, especially here where Black gets the bishop-pair into the bargain. But the real problem is the weakness of the white king's position.

19...Bh3 20.Rd1 Nf3+ 21.Kh1 Qf5!? 22.Ne4

Best, according to the engines, is 22.Qe3 Qg4 23.Ne2, but they reckon 23...Nxh4 gives Black a huge advantage.

Black to play and win

*****

*****

*****

*****

*****

22...Bf1! 0-1

Summing Up Mariánské Lázně

MY score of +5=1-3 in the Seniors 50+ gained me 16.4 Fide elo, and I won the prize for being third-best among those born before 1959.

Flashing my winnings in Mariánské Lázně's subtly named Irish Pub

A Brit Of All Right

THE English Chess Federation has announced this year's British Chess Championships will be held at Leicester's De Montfort University.

The senior 50+ and 65+ championships will apparently be run at the same time, with each being seven-round events.

Provisional details are at the ECF website.

Mariánské Lázně Game Eight

Spanton (1895) - Jiří Tejkal (1816)

Sicilian Bb5(+)

1.e4 c5 2.Nf3 Nc6 3.Bb5 g6 4.c3 Bg7!?

This is easily the most popular move in ChessBase's 2023 Mega database, although in my, admittedly limited, praxis opponents have generally preferred 4...Nf6 (nine examples of the latter, six of the text, including this game, and one of 4...d6).

5.d4 cxd4 6.cxd4

What should Black do about White's classical centre?

*****

*****

*****

*****

*****

6...d5?

There are seven examples of this mistake in Mega23, including by a 2696, albeit at blitz. Stockfish15 and Komodo13.02 suggest 6...Nf6 7.Nc3 and now 7...d5, although they reckon 8.Ne5!? Bd7 9.Nxd7 gives White a slight edge.

7.cxd5 Qxd5 8.Nc3 Qe6+

Cem Kaan Gokerkan (2470) - Kirill Alekseenko (2696), Lichess Blitz 2021, went 8...Qh5 9.d5 a6 10.Ba4 b5 11.dxc6 bxa4 12.Qa4 with a large advantage for White (1-0, 55 moves).

9.Be3 a6 10.d5 Qg4 11.h3 Qb4

If 11...Qxg2? then 12.Bxc6+ bxc6 13.Rh2 traps the black queen.

How should White proceed?

*****

*****

*****

*****

*****

12.Bd4?!

This is probably enough for an advantage but almost certainly better is 12.Ba4 b5 13.dxc6 Bxc3+ (or 13...bxa4 14.0-0) 14.bxc3 Qxa4 15.Qxa4 bxa4 16.0-0-0, when material is level but White's large lead in development is decisive, according to the engines, which also like 12.Bxc6+ bxc6 13.0-0.

12...Bxd4 13.Qxd4!?

The engines for quite some time prefer 13.Nxd4 but come to see the text as almost on a par.

13...Qxd4 14.Nxd4 axb5 15.dxc6 bxc6 16.Nxc6 Bd7?

Better is 16...Bb7, which transposes to the game after 17.Nxb5 Bxc6.

17.Nxb5?!

Sacrificing the knights to win the black rook. This leads to interesting play, but White had a simpler way to an advantage, according to the engines, with 17.Ne5, eg 17...Rb8 18.Nxd7 Kxd7 19.0-0-0+, although they prefer giving up the b5 pawn with 17...Nf6!? 18.Nxd7 Nxd7 19.Nxb5 0-0.

17...Bxc6 18.Nc7+

Where should the black king go?

*****

*****

*****

*****

*****

18...Kf8?!

The king will be safe on the kingside, but it would not be in great danger on the queenside and is probably needed to stop the white passed pawns. The engines give 18...Kd7 19.Nxa8 Bxa8 (not 19...Bxg2? 20.Rd1+, after which the white knight escapes, eg 20...Kc6 21.Rg1 Bxh3 22.Rd8) 20.0-0 Nf6 21.b4, which is the same as the game but with the black king on d7 instead of far from the action on f8.

19.Nxa8 Bxa8

This is best, according to Stockfish15, although Komodo13.02 for a while marginally prefers 19...Bxg2!? However White has 20.0-0-0! with what Stockfish15 reckons is a winning advantage but Komodo13.02 evaluates as only a slight edge for White.

20.0-0 Nf6 21.b4

How should Black combat White's passers?

*****

*****

*****

*****

*****

21...Kg7

Stockfish15 at first marginally prefers 21...Ke8 but comes to marginally prefer the text. Komodo13.02 is not keen on either king move, preferring 21...e6 and 21...Nd5. The one thing they agree on is White is slightly better.

22.b5 Rb8 23.a4 Bd5!?

The engines agree this is probably not best, but find it difficult to choose between 23...Ne4, 23...Ne8, 23...Nd5 and 23...e6.

24,Rfb1 Ne4 25.b6 Bb7 26.a5 Ba6?!

The engines' 26...Nc5 is probably better, but with advantage to White.

27.Re1 Nd6

The e pawn cannot be saved.

28.Rxe7 Nb5?!

The engines suggest 28...Nb7 but agree White is winning, eg 29.Rc7 h6 30.Ra3 with Rf3 to come. Indeed Stockfish15 reckons Black should sacrifice the knight with 29...Nxa5!? 30.Rxa5 Rxb6, which at least keeps the game going.

29.Rd1 Bc8

Or 29...Rf8 30.Rdd7 (30.b7 also wins) Bc8 31.Rd5 Ba6 32.b7, when the threat of 32.Rdd7 followed by Rxf7+! cannot be successfully met.

30.Re8 1-0

Mariánské Lázně Seniors 50+ Round Nine

WON in 22 moves with black this morning against Germany's Manfred Korth (1980).

Royal Trumps

Statues commemorating the 1904 meeting in Mariánské Lázně between Britain's Edward VII (foreground) and Austria-Hungary's Franz Joseph - the former apparently tried to woo the latter from his alliance with Germany

Mariánské Lázně Game Seven

Stef van Kesteren (1862) - Spanton (1895)

Scotch

1.e4 e5 2.Nf3 Nc6 3.d4 exd4 4.Nxd4 Qf6!?

This can have independent significance, but after ...

5.Be3 Bc5

... the game has transposed to normal 4...Bc5 lines, except Black has avoided the variation 4...Bc5 5.Nb3. Whether Black should be trying to avoid that variation is another matter.

6.c3 Nge7 7.Bc4 0-0 8.0-0 b6!?

This is fourth-most popular in ChessBase's 2023 Mega database. The mainline runs 8...Bb6 9.Na3 d6 10.Nac2 with an equal position, according to Stockfish15 and Komodo13.02.

How should White respond?

*****

*****

*****

*****

*****

9.Nxc6

The 8...b6!? variation is fairly new - the oldest game in Mega23 is from 1992 - and no consensus has emerged over how White should proceed. The main move in Mega23 is 9.f4!?, but the slightly surprising 9...d5!? seems a good reply, eg 10.exd5 Na5 11.Bd3 Bb7 12.Nc2 Nxd5 13.Bxc5 bxc5, which occurred in Johan-Sebastian Christiansen (2506) - Vladimir Vusatiuk (2321), Pardubice (Czechia) 2018, and was slightly better for Black, according to the engines (but ½–½, 30 moves). The engines flick between several ninth moves for White, including the text, which is second-most popular in Mega23.

9...Qxc6 10.Bxc5 Qxc5 11.Qxd4!?

This is the point of White's play - inviting Black to give White a classical centre with 11...Qxd4 12.cxd4, although it is debatable how strong that would be in the absence of queens and four minor pieces.

11...Bb7!? 12.Qxc5

Almost certainly not 12.Qxd7?! Qxc4 13.Qxe7 Rfe8, after which Black regains the sacrificed pawn and has an initiative based on better development.

12...bxc5

How would you assess this queenless middlegame?

*****

*****

*****

*****

*****

White's pawn-structure is better, but Black has pressure against e4 and no weaknesses that are easy to get at. The engines reckon the position is equal.

13.Nd2 Rfe8 14.Rae1!?

Alexandre Danin (2598) - Vladislav Nozdrachev (2463), Kaluga (Russia) 2013, went 14.Rfe1 Ng6 15.b4!? a5!? 16.a3 acb4 17.axb4 Rxa1 18.Rxa1 cxb4 19.cxb4 Kf8 with a draw in 35 moves. The idea of the text, presumably, is to support kingside expansion.
14...Nc6 15.Bd5

White may be slightly better after 15.f4, according to the engines.

15...Rab8 16.Bxc6?!

This is probably bad, not so much because the white bishop was better than the black knight, although it was, but because the only minor pieces left on the board after the trade are a black bishop and a white knight, and the former is the better piece.

16...Bxc6 17.b3 d6 18.f3 a5

Who stands better?

*****

*****

*****

*****

*****

The position is equal, according to the engines, but I believe Black's position is much easier to play, ie Black will attack on the queenside, while it is not clear what White can do to win the initiative.

19.c4 Rb7 20.Re3 f6 21.Rfe1 a4 22.bxa4?!

White temporarily gains a pawn, but after Black recaptures on a4 White will be under pressure.

22...Ra8 23.Rb3 Rba7 24.Reb1 Kf7 25.Rb8 Rxa4 26.Rxa8 Rxa8 27.Rb2 Ke6 28.Kf2 Ke5 29.Ke3 Ra3+ 30.Nb3 Bb7 31.g3 Ba6 32.Rc2 Ra4 33.Nd2 Rb4?!

According to the engines, Black has a large advantage after 33...f5.

34.h4 Bc8 35.a3 Ra4 36.Rc3 Ba6 37.f4+ Ke6 38.g4 g6 39.f5+ Kf7 40.fxg6+ hxg6 41.Kf4 Bc8 42.Nf3 Be6 43.Nd2 Ke7 44.g5 f5?

Black should hold tight with, for example, 44...Kf7, after which it is Black, if anyone, who has a small edge. After the text Black is probably lost.

45.exf5 Bxf5 46.h5 Kf7 47.h6 Ra8?!

This almost certainly makes matters worse, although 47...d5 48.Ke5 is not a great improvement.

48.Rb3 Ke7 49.Ne4 Bxe4 50.Kxe4 Ke6 51.Rf3 Ra4

Desperately trying for counterplay, but this is one rook-and-pawn ending that is definitely not drawn.

The game finished:

52.Rf6+ Kd7 53.Rxg6 Rxc4+ 54.Kd3 Rh4 55.Rg7+ Ke6 56.h7 Kd5 57.Rxc7 Rh3+ 58.Ke2 Ke4 59.g6 Rh2+ 60.Kf1 Ke3 61.Kg1 Rh6 62.Re7+ Kd4 63.Re6 Kd5 64.g7 Rxe6 65.g8=Q 1-0

Mariánské Lázně Seniors 50+ Round Eight

WON in 30 moves with white against Jiří Tejkal (1816).

Plus Ça Change, Plus C'est La Même Chose

The modern world meets the classical

Mariánské Lázně Game Six

Spanton (1895) - Johann Loibl (1677)

1.Nc3/Scotch

1.Nc3 e5 2.Nf3 Nc6 3.d4 exd4 4.Nxd4 Bc5!?

This is second in popularity to 4...Nf6 in ChessBase's 2023 Mega database.

5.Nf5!?

The text, 5.Nxc6 and 5.Be3!? are the top choices of Stockfish15 and Komodo13.02.

5...Qf6 6.e4

There are 365 examples of this position in Mega23, normally being reached via a Scotch move-order: 1.e4 e5 2.Nf3 Nc6 3.d4 exd4 4.Nxd4 Bc5 5.Nf5!? Qf6 6.Nc3

*****

*****

*****

*****

*****

6...Nge7 7.Be3 Bb4!? 8.Nxe7 Bxc3+!? 9.bxc3 Qxe7

Now the flurry of tactics is over, how would you assess the result of the opening?

*****

*****

*****

*****

*****

White has the bishop-pair on a board with an open centre but lots of pawns. White also has more space in the centre, but three isolated pawns, two of which are doubled. The engines reckon chances are even.

10.Bd3 d6 11.0-0 Ne5!?

This may be a novelty. The engines slightly prefer castling.

12.Rb1

The engines reckon 12.f4!? Nxd3 13.cxd3 gives White an edge.

12...Nxd3 13.cxd3 b6 14.Rb5!?

The rook embarks on an interesting excursion, but perhaps developing with 14.Re1 or 14.Qd2 is better.

14...Bd7 15.Rg5 f6 16.Rg3 0-0-0!?

More or less committing to castling long.

How should White respond?

*****

*****

*****

*****

*****

17.f4

The engines give 17.h4!? 0-0-0 18.c4 with an unclear position, but one they reckon is equal.

17...0-0-0

Black is better after 17...h4, according to the engines.

18.f5 Be8!?

The engines still like ...h4.

19.Qb3 h4 20.Rgf3?

Necessary seems to be 20.Rh3 or 20.Rg4.

How can Black exploit White's last move?

*****

*****

*****

*****

*****

20...Bh5?!

Missing the strength of 20...d5, when 21.exd5? loses to 21...Bh5. The engines flick between several moves, but eventually more or less settle on 21.d4, when 21...dxe4 is good, but best seems to be 21...h3!?, eg 22.exd5 hxg2 23.Re1 Qe4 with strong play.

21.R3f2 d5 22.Qa4?!

Probably better is 22.c4!?, and if 22...dxe4 then 23.d4!?

22...Kb7 23.Re1 Be8 24.Qb3 Bf7 25.exd5?!

This helps Black's play in the centre, but White has major problems in any case.

25...Rxd5 26.c4 Re5 27.d4 Re4

Also strong is winning a pawn by 27...Bxc4 etc.

28.Rff1 Qd7?

This lets White off the hook. The engines like 28...h3, but even more 28...Bxc4.

29.Qd3?

Both 29.d5 and 29.Bf2 leave Black with a much smaller advantage, according to the engines.

29...Rhe8 30.Bf2 Qc6

This is probably enough for an edge, but Black can grab a pawn with 30...Qxf5, and if 31.Bxh4 then the engines point out 31...Qg6! is very good, for example 32.Bg3? loses to 32...Rxe1, while 32.Rxe4 Qxe4 33.Qxe4+ Rxe4 34.Bf2 is met by 34...Bxc4, when White has nothing better than the passive 35.Ra1.

31.Rxe4 Rxe4?!

Probably better is 31...Qxe4, when White seems to have too many weaknesses to defend.

32.Rc1 Qe8 33.Qf3 Kb8 34.h3!?

The engines like this safety-first move - White's back rank is more vulnerable than it might appear at first glance.

34...Bh5 35.Qa3 Re2 36.c5 Bf7 37.cxb6 axb6 38.Qc3 Qd7 39.a3!?

The engines prefer grabbing the h4 pawn, or protecting the a pawn with 39.Rxc2.

39...Bd5 40.Qd3?

It was necessary to defend the second rank with 40.Rc2.

40..Ra2

Opposite-coloured bishops are well known for their drawing tendencies in endings, but in middlegames that is rarely the case, and here it could not be more obvious who has the better bishop

*****

*****

*****

*****

*****

41.Qc3 Qd6 42.Ra1 Rxa1+ 43.Qxa1 Qf4 44.Qe1 Qg5

White cannot prevent mate and defend all the weak white pawns.

45.Qe8+ Kb7 46.Qg6 Qc1+ 47.Kh2

Now 47...Qxa3 wins, eg 48.Bxh4 Qe7 followed by pushing the b pawn. Instead the game saw:

47...Qf4+ ½–½

Mariánské Lázně Seniors 50+ Round Seven

LOST in 65 moves with black against Netherlands' Stef van Kesteren (1862).

Familiarity Breeds Contentment

An overnight scattering of snow has given Mariánské Lázně what is for me a more-familiar look

Mariánské Lázně Game Five

Spanton (1895) - Jan Bartoš (FM 2133)

Sicilian Bb5(+)

1.e4 c5 2.Nf3 d6 3.Bb5+ Bd7 4.Bxd7+ Qxd7 5.0-0 Nc6 6.Qe2!?

This has been played by grandmasters, but is only fifth-most popular in ChessBase's 2023 Mega database.

6...g6 7.c3 Bg7 8.Rd1

White is ready to play d4 - how should Black respond?

*****

*****

*****

*****

*****

8...e5!?

This is the main move in Mega23 and is the choice of Stockfish15 and Komodo13.02.

9.Na3 Nge7 10.Nc2

Robert J Fischer - Friðrik Ólafsson, Mar del Plata (Argentina) 1960, went 10.d4!? cxd4 11.cxd4 exd4 12.Nb5 0-0 13.Nfxd4 d5 14.Nb3 a6 15.Nc3 d4 16.Na4 with equality, according to the engines (but 1-0, 42 moves). I rejected 10.d4!? because of 10...exd4 11.cxd4 Nxd4 12.Nxd4 Bxd4?!, missing that 13.Nb5 will restore material equality, eg 13...Nc6?! 14.Bf4 etc. The engines reckon Black should play 12...cxd4 13.Nb5 Nc6 14.Bf4 Be5, as in Nicolas Rossolimo - Fischer, US Championship (New York) 1962, which continued 15.Bh6 0-0-0 16.f4 Bf6 (½–½, 43 moves), or give the pawn back by 14...0-0!?, when the engines reckon both 15.Nxd6 and 15.Bxd6 leave Black slightly better.

10...0-0 11.d4 cxd4 12.cxd4 exd4 13.Ncxd4 Nxd4 14.Nxd4 d5

How should White continue?

*****

*****

*****

*****

*****

15.exd5!?

This may be a novelty. The engines prefer a known move, 15.e5, and another apparent novelty, 15.Nf3!?

15...Nxd5 16.Bg5?!

This is Komodo13.02's choice for a short while, but both engines quickly come to dislike the move. They suggest 16.Qf3 with approximate equality.

16...h6 17.Bh4 Rfe8

Possibly stronger is the engines' 17...Nf4, eg 18.Qe3 g5 19.Ne2 (not 19.Bg3? Bxd4 20.Rxd4 Qxd4!) Nxe2+ 20.Qxe2 Qe8 21.Qxe8 Rfxe8 22.Bg3 Re2 with a good game.

18.Qd2?

Other queen moves, eg 18.Qb5, which is probably best, keep White's disadvantage to a minimum.

18...Qg4 19.Bg3?

If 19.Nf3 then 19...Nf4 20.Bg3 Ne2+ is strong, eg 21.Kh1 Qc4, threatening ...Nxg3 followed by ...Re2, while the engines' suggested 19.h3 Qxh4 20.Nf3 Qb4 21.Qxd5 also allows ...Re2. But both these lines are less catastrophic than the text.

19...Bxd4 20.Qxd4 Re1+! 0-1

Mariánské Lázně Seniors 50+ Round Six

DREW in 47 moves with white against Austrian Johann Loibl (1677).

Singing The Town's Praises

Mariánské Lázně's neo-baroque colonnade, shown with the town's singing fountain in the foreground, was built in 1888-9 with cast iron from nearby Blansko to a design by Viennese architects

Mariánské Lázně Game Four

Piotr Cap (2067) - Spanton (1895)

New London

1.d4 d5 2.Bf4 e6 3.e3 Bd6 4.Bg3 b6!?

Hinting at, but not committing to, going for an exchange of Black's bad bishop for White's good one.

5.Nf3 Nf6 6.Bd3

How should Black proceed?

*****

*****

*****

*****

*****

6...Bb7

Since White has spent a tempo on moving the light-square bishop, it might be thought this is the time to go for the aforementioned exchange. However 6...Ba6!? 7.Bxa6 Nxa6 7.Qe2 may favour White. My main analysis engines, Stockfish15 and Komodo13.02, suggest continuing 7...Qc8!?, which is not in ChessBase's 2023 Mega database. One line they give continues 8.a3 0-0 9.Nbd2 Qb7!? 10.c4, after which the black queen's knight looks awkwardly placed, but the engines reckon the position is equal.

7.0-0 0-0 8.Nbd2 Ne4 9.c4 Nxg3 10.hxg3

How would you assess the position now Black has won the bishop-pair?

*****

*****

*****

*****

*****

The bishops are a long-term plus and the white kingside is damaged. But Black has no immediate threats and White has hopes of a queenside initiative. Black also has to be wary about a Greek gift sacrifice on h7. Stockfish15's verdict is complete equality, while Komodo13.02 gives White a slight edge.

10...Nd7 11.Rc1 c5!?

The engines agree this is no time to be afraid of possible hanging pawns, ie pawns on d5 and c5 unsupported by pawns on the e or b files.

12.cxd5 exd5 13.b3!?

Preventing ...c4, but there are other ways to do this. The drawback with the move is the white queen can no longer quickly come to the queenside, for example to the a4 square, which means Black can play ...Rc8 without worrying about the safety of the a7 pawn.

13...Rac8 14.Qe2

Not just developing the queen - White wants to get rid of Black's bishop-pair.

14...Qe7 15.Ba6 Bxa6 16.Qxa6 Nf6

How would you assess the position now Black no longer has the bishop-pair?

*****

*****

*****

*****

*****

The bishop-pair has gone, but in achieving this White has swopped his good light-square bishop for Black's bad one (bad because of a black central pawn being fixed on a light square). In the resulting position Black has more space on the queenside, but White has pressure against the Black queenside. The engines reckon the position is completely equal.

17.Rfe1 Ne4 18.Nxe4 dxe4 19.Nd2

How should Black proceed?

*****

*****

*****

*****

*****

*****

19...Rfd8?!

Developing the king's rook may look right at first glance, but this was a good time to break the pin on the c pawn, for example by 19...Rcd8.
20.Rc4

Indirectly threatening e4.

20...f5 21.b4 Rb8

Almost certainly not 21...b5?! True, 22.Qxb5 Rb8 looks fine for Black, but 22...Rc2! 23.c4 (what else?) Qxb5 Bxb4 24.Qxf5 wins for White, according to the engines.

22.bxc5 bxc5 23.Nb3 Rb6 24.Qa5?!

The engines reckon White has the upper hand after 24.Qa4.

24...cxd4 25.Qd5+ Kh8?!

Probably better is 25...Qf7, after which 26.Rxd4 Qxd5 27.Rxd5 Rf8 holds.

26.Rxd4 Rf8 27.Rc1 Be5 28.Rdc4 Rd8?!

More accurate seems to be 28...Rd6.

29.Rc8! Rbd6 30.Qa8?!

The engines give 30.Qc5, which appears to win the a7 pawn, eg 30...Qe6 31.Rxd8+ Rxd8 32.Qxa7, or 30...Qe8 31.Rxd8 Rxd8 32.Qxa7.

30...g6 31.Rxd8+ Rxd8 32.Rc8 Rxc8 33.Qxc8+ Kg7 34.Nc5 Kh6!?

This is probably fine - completely equal, according to Komodo13.02 - but Stockfish15 prefers other moves, eg 34...Qd6.

35.Ne6

Stockfish15 at first gives White the upper hand after this, but later only a slight edge at best. The position is still completely equal, according to Komodo13.02.

Black to play and lose

*****

*****

*****

*****

*****

35...Qb4?? 36.Qd8 1-0

Mariánské Lázně Seniors 50+ Round Five

LOST in 20 moves with white against Fide master Jan Bartoš (2133) in the second of today's double-round games.

Mariánské Lázně Seniors 50+ Round Four

LOST in 36 moves with black against Poland's Piotr Cap (2067) in the first of today's double-round games.

Getting The Abbey Habit

Karl Reitenberger of Teplá Abbey is credited with founding Mariánské Lázně as a spa town on unpromising marshland in the early 1800s, although the area had been inhabited, sparsely, since at least the Middle Ages

Mariánské Lázně Game Three

Hartmut Moritz (2066) - Spanton (1895)

Ponziani

1.e4 e5 2.Nf3 Nc6 3.c3 Nf6!?

This is easily the most popular move in ChessBase's 2023 Mega database, although it scores eight percentage points lower than 3...d5.

4.d4 Nxe4!?

Magnus Carlsen preferred the more common 4...exd4 in a 2013 win over Yifan Hou at Tata Steel (Wijk aan Zee, Netherlands).

5.Bd3!?

The mainline in Mega23 runs 5.d5 Ne7 6.Nxe5 Ng6 7.Nxg6 hxg6 8.Bd3 Nf6 9.0-0 with maybe a slight edge for White, according to Stockfish15 and Komodo13.02.

5...d5 6.Nxe5

The engines slightly prefer this over 6.dxe5 Bg4.

6...Nxe5 7.dxe5 Nc5?!

The knight was well-placed on e4. The c5 square is not bad either, but Black is spending a tempo driving the white light-square bishop to arguably a better square. The engines suggest 7...Bc5, claiming equality, one point being 8.Bxe4 can be met by 8...Qh4. Instead White should probably castle.

8.Bc2 Be7 9.0-0 0-0 10.Qh5!?

This may be a novelty. A known continuation is 10.b4 Ne4 11.f3, when White was at least slightly better, according to the engines, in Andrew Morley (2065) - Renāte Vidruska (2045), 4NCL Team Championship Online 2020, although Black won in 46 moves.

10...g6 11.Qe2 f6 12.Bh6 Rf7!? 13.b4?!

The engines strongly dislike this, claiming it leaves Black better. They suggest 13.exf6 Bxf6 and either 14.Nd2 or 14.Re1, when Stockfish15 gives White a slight edge but Komodo13.02 calls the position equal.

13...Ne6 14.f4

The engines prefer 14.Nd2 or 14.exf6, but give Black at least the upper hand.

How should Black respond?

*****

*****

*****

*****

*****

14...fxe5

This seems best, although the engines also like 14...a5!? But forcing an exchange of dark-square bishops with 14...Bf8?! throws away most of Black's advantage, according to the engines, while they reckon the passive 14...Nf8? gives White at least the upper hand after 15.Nd2.

15.f5!?

This is White's idea, whereas recapturing with 15.fxe5?! lets Black swop a pair of rooks and so relieve the pressure.

What should Black play?

*****

*****

*****

*****

*****

15...gxf5!

This is clearly best, according to the engines. If 15...Nf4 then 16.Bxf4 exf4 17.fxg6 hxg6 18.Bxg6 Rg7 is much better for Black, despite the exposed black king, according to the engines. However they reckon sacrificing the exchange with 16.Rxf4! exf4 17.fxg6 hxg6 18.Bxg6 is much less clear, eg 18...Rf6? 19.Qh5 wins for White. Instead Black should probably play 18...Qd6!? 19.Bxf7+ Kxf7 20.Qh5+ Qg6 21.Qxg6+ Kxg6, when 22.Bxf4 leaves White a pawn up but Black gets strong play with the bishop-pair, eg 22...Bf6 23.Kf2 (23.Bxc7? Bf5) Be6 24.Be3 a5 25.bxa5 Rxa5, although the line is not very forcing and White remains with an extra pawn.

16.Qxe5?

Better is 16.Bxf5, when 16...Nf4!? 17.Bxf4 exf4 (capturing the light-square bishop lets White equalise with 18.Bxe5, according to the engines) 18.Bxc8 leaves White a pawn down, whichever way Black recaptures, but Black has pawn weaknesses and an exposed king.

16...Bd6 17.Qe1

This seems best. Certainly not 17.Qxd5?? Bxh2+ etc.

17...Qf6

Even stronger, according to the engines, is 17...f4.

18.Be3 Qe5 19.Qg3+

This seems necessary as 19.g3? f4 is much worse for White, according to the engines.

19...Qxg3 20.hxg3 Bxg3!? 21.Bxf5 Nf8?!

Not 21...Ng7? 22.Bxh7+!, but the engines' 21...Be5 and 21...Nd4!? are probably better.

22.Bc2?

Exchanging on c8, or dropping the bishop to d3 from where it covers the king's rook, is better.

22...Rxf1+?

This is winning but much better is 23.Bxh2+!

23.Kxf1 Bd7?!

Hoping to play 24...Bb5+, but that is little more than a check. Best seems to be 23...Be5, one point being 24.Bd4?! runs into 24...Bxd4 25.cxd4 Ne6.

24.Na3?!

Preventing 24...Bb5+, but the knight has few prospects on a3. The engines give 24.Bb3 c6 (24...Bb5+ 25.Kg1 is not a problem for White) 25.Nd2, albeit with the upper hand for Black.

24...Ng6 25.Rd1 Re8 26.Bc5?!

Objectively better is 26.Bf2 Rf8 27.Rd2, but Black can simplify at will.

26...Bg4 27.Rc1 Nf4 28.Bb3 b6 29.Bf2

There seems nothing better, but White is completely lost.

29...Bxf2 30.Kxf2 Nd3+ 31.Kg3 Nxc1 32.Kxg4 Nxb3 33.axb3 c6 (0-1, 50 moves)

Mariánské Lázně Seniors 50+ Round Three

WON in 50 moves with black after being downfloated against Germany's Hartmut Moritz (2066).

Golden Oldie

Mariánské Lázně's golden age is generally recognised as being in the second half of the 19th century, when many of its more-interesting buildings were erected

Mariánské Lázně Game Two

Spanton (1895) - Miloslav Macho (1692)

Alekhine's Defence

1.e4 Nf6 2.e5 Nd5 3.c4 Nb6 4.d4 d6 5.exd6

This is the main move in ChessBase's 2023 Mega database, and also the most successful percentagewise.

5...cxd6!?

Larry Kaufman in The Chess Advantage In Black And White (Random House 2004) wrote: "Some commentators now give this a ?!" However it remains the most popular move in Mega23 and was used by Magnus Carlsen in a 2020 rapidplay win over Levon Aronian. As far as I can discover Carlsen, had previously just tried 5...exd6.

6.Nc3 g6 7.Be3 Bg7 8.Be2

Aronian played 8.h3!? against Carlsen, and their game continued 8...0-0 9.Nf3 Nc6 10.d5!? Na5 11.Bd4 Bxd4 12.Qxd4 e5!? 13.dxe6 Bxe6 14.Ne4?! (14.Be2!? Nbxc4 15.Bxc4 Nxc4 16.0-0 gives good compensation for a pawn, according to Stockfish15 and Komodo13.02) Nc6 15.Nf6+ Kh8 16.Qf4 Bf5 with an unclear position that the engines reckon favours Black (0-1, 37 moves).

The mainline in Mega23 runs 8.Rc1 0-0 9.b3 Nc6 10.d5 Ne5 11.Be2 f5!? (the engines strongly dislike this, while reckoning White has the upper hand anyway) 12.f4 Ng4 13.Bxg4 fxg4 14.Nge2, scoring a whopping 73% for White.

8...Nc6 9.Nf3 Bg4 10.b3 0-0 11.0-0

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

*****

*****

*****

*****

*****

White has more space in the centre but Black's minor pieces are more active than the white ones. The engines reckon the position is equal, but after various continuations are inputted, Stockfish15 comes to give White a slight edge.

11...e5

The engines follow the most popular line in suggesting 11...d5 12.c5 Nc8, after which Stockfish15 gives White a slight edge but Komodo13.02 calls the game equal.

12.d5!?

Played with the coming exchange sacrifice in mind. Sergei Tiviakov (2520) - Arnaud Hauchard (2420),  Torcy (France) 1991, went 12.dxe5 dxe5 13.c5!? Nc8 14.h3 Bf5 15.Ng5 Nd4 16.Bc4 with a slight edge for White, according to the engines (1-0, 39 moves).

12...e4?!

Komodo13.02 reckons 12...Bxf3 13.Bxf3 Nd4 is completely equal, but Stockfish15 gives White a slight edge.

13.Nxe4!

White is probably also better after 13.dxc6, but the engines agree the text is clearer.

13...Bxa1

Otherwise Black is a pawn down without compensation.

14.Qxa1

Who is better, and, more importantly, by how much?

*****

*****

*****

*****

*****

White has a pawn and the bishop-pair for the exchange, which is often reckoned to be sufficient compensation. But here the added weakness of the black king's position means White is the rough equivalent of a rook ahead, according to the engines.

14...Ne5

The engines suggest 14...f5!?, but White has several very good continuations, including the sharp 15.Bh6 Qe7 16.Nxd6 Bxf3 17.gxf3!? Ne5 (17...Qxe2?? 18.Qg7#) 18.f4 Ned7 19.Bf3 Rab8 20.Nxb7! Rf7 21.d6, after which White has three pawns and the bishop-pair for the exchange.

15.Nxe5 Bxe2 16.Re1

Even stronger, according to the engines, are 16.Ng4!? and 16.Nc6!?

16...dxe5 17.Rxe2

This is almost certainly better than trying to get fancy with 17.Qxe5?!, when 17...Nd7 muddies the waters, eg 18.Qc3 Bh5 19.Bh6 f6 20.Bxf8 (20.h3!?) Qxf8, after which White has the better coordinated pieces but is down a bishop for two pawns.

17...Nd7

Protecting the f6 square, while not giving up the e6 square, which would be the case after 17...f6.

18.Bh6 Re8 19.c5 f5

Possibly better is the engines' 19...Qh4!?, eg 20.Bg5 Qh5 21.Qd1 f5, but 22.c6 is very strong as moving the black knight is answered by the forking Nf6+.

20.Nd6 Qh4

How should White continue?

*****

*****

*****

*****

*****

21.Qc1!?

This is enough to maintain White's winning advantage, but best-play, according to the engines, runs 21.Nxe8 Rxe8 22.c6 bxc6 23.dxc6 Nb6 24.Qc1.

21...f4

Material is level after 21...Nxc5 22.Nxe8 Rxe8 23.Qxc5 Qxh6, but 24.d6 wins, according to the engines.

22.Nxe8 Qxh6

Or 22...Rxe8 23.Bxf4.

23.Nc7

Other moves also win.

23...Rc8 24.d6 Qh5 25.f3 e4?!

This does not help, but 25...g5, which the engines reckon is best, loses to, for example, 26.Qc4+ Qf7 27.Qe4 (threatens the simple 28.Qxb7) Rb8 (27...Nxc5 28.Qxe5 Nd7 29.Qxg5+) 28.b4, after which Black is 'only' a pawn down, but seems helpless.

26.Rxe4 Nxc5?

Better is 26...Qxc5, but 27.Qxc5 Nxc5 leaves White with several winning continuations, eg 28.Re5 Nd7 29.Re7 Rd8 30.Ne6.

27.Qc4+ Kh8 28.Qd4+ Kg8 29.Re7

Checkmate cannot be sensibly stopped.

The game finished:

29...Ne6 30.Rxe6 Qf5 31.Re4 Qa5 32.d7 Qxc7 33.Re8+ 1-0