Bavarian International Round Four

Jens Weichelt (2042) - Spanton (1858)
Réti
1.Nf3 c5 2.e3 g6 3.d4 d5 4.Be3 Nc6 5.0-0 cxd4 6.exd4 Bg7 7.c3

This pawn-formation could arise from the Exchange Variation of the Caro-Kann

*****

*****

*****

*****

*****

7...Nf6 8.Bf4 0-0 9.Nbd2 Nh5!? 10.Be3 Qd6 11.Re1!?

This seems the best way of meeting Black's threat to win the bishop-pair.

11...Nf4 12.Bf1 a6

How should White proceed?

*****
*****
*****
*****
*****

13.Ne5!?

How should Black respond?

*****

*****

*****

*****

*****

13...Nh5

After 13...Bxe5 14.dxe5 Qxe5 I was worried about 15.Qf3. However Stockfish16 and Komodo14.1 continue 15..g5. albeit giving a slight edge to White.

14.Ndf3 f6 15.Nxc6

The engines prefer 15.Nd3.

15...bxc6 16.b4!? Re8 17.Nd2!?

The c5 square beckons.

What should Black play?

*****

*****

*****

*****

*****

17...e5!?

Aggressive, and perfectly reasonable, if followed up correctly.

18.g4!? exd4 19.cxd4 Nf4??

This loses a piece, Correct is 19...Bxg4! 20.Qxg4 f5 with ...f4 and at least equality to come.

20.Bxf4 Rxe1 21.Bxd6 Rxd1 22.Rxd1 (1-0, 32 moves)

Venue

THE tournament, which ended up with 518 players, is being held at Gut Kaltenbrunn on the outskirts of Gmund.

It is a converted 14th-century farmhouse, formerly owned by Tegernsee Abbey, providing plenty of playing-room and associated facilites.

Gut Kaltenbrunn viewed from across the Tegernsee

Biergarten

Entrance to the inner courtyard

Food, beer and analysis tent

One of two playing halls

Bavarian International Round Three

Spanton (1858) - Andriy Manucharyan (2058)

French Winawer

1.e4 e6 2.d4 d5 3.Nc3 Bb4 4.Ne2!?

This seems to have been first played by Géza Maróczy in 1924. Alexander Alekhine used the move to beat Aron Nimzowitsch in 1931, and Emanuel Lasker similarly scored over José Capablanca four years later, the same year Alekhine won with it against Max Euwe in their first world championship match. The move scores a decent 56% in ChessBase's 2023 Mega database.

How should Black respond?

*****

*****

*****

*****

*****

4...dxe4

This is easily the commonest reply in Mega23, but 4...Nf6!? and 4...Nc6!?, which have also been played hundreds of times, score much higher percentagewise.

5.a3 Bxc3+

This is popular, but the main move in Mega23 is 5...Be7.

6.Nxc3 Nc6

Nimzowitsch played 6...f5!? against Alekhine at Bled (Yugoslavia) 1931. The game continued 7.f3 exf3 8.Qxf3 Qxd4!? 9.Qg3 Nf6!? 10.Qxg7 Qe5+ 11.Be2 Rg8 12.Qh6 Rg6 13.Qh4, when Stockfish16 awards White the upper hand, but Komodo14.1 is at first less keen on White's position, albeit coming to give White a slight edge (1-0, 19 moves).

7.Bb5 Ne7 8.Be3

The mainline in Mega23 runs 8.Bg5 f6, and then 9.Be3, with an equal game, according to the engines.

8...0-0 9.Nxe4?!

Stockfish16 prefers the most popular move, 9.Qd2, continuing 9...f5!? 10.0-0-0 with an unclear position. Komodo14.1 suggests 9.Bxc6!? Nxc6 10.Qd2 Qe7!? 11.Nxe4, rating the position as equal.

9...f5 10.Ng5 f4 11.Bd2

Not 11.Nxh7? Kxh7 12.Qh5+ Kg8 13.Bd3 as Black has 13...Nf5, while 11.Qh5?? h6 leaves two white pieces en prise.

11...Qd5!

Forcing White to give up the bishop-pair.

12.Bxc6 Nxc6 13.Nf3

How should Black proceed?

*****

*****

*****

*****

*****

13...Qe4+?!

Komodo14.1 likes the simple 13...Nxd4 14.Nxd4 Qxd4 15.Bc3 Qe4+ 16.Qe2 Qxe2+ 17.Kxe2, reckoning 17...e5! gives Black the upper hand.

Stockfish16 gives 13...e5!? 14.dxe5 Bh3!, the idea being 15.gxh3? fails to 15...Nxe5 16.Nxe5 Qxh1+ 17.Ke2 Qe4+ 18.Kf1 Qxe5. Better is 15.c4!, but 15...Qe4+ 16.Qe2 Qxe2+ 17.Kxe2 Bxg2 is good for Black.

14.Qe2 Qxc2 15.0-0!?

This seems a slight improvement over the known move 15.Rc1, after which 15...Qxb2 favours Black, according to the engines.

15...Bd7

Grabbing a second pawn with 15...Qxb2?! can be met by the forcing 16.Qc4 Qb6 17.Rab1 Qa6 18.Qxa6 bxa6 19.Rfc1, with what the engines reckon is at least enough compensation for two pawns.

16.Rfc1 Qg6 17.Rc5 Rae8 18.Bc3

What should Black play?

*****

*****

*****

*****

*****

18...e5!?

The engines approve of this returning the pawn for an attack, although they reckon 18...h6 is also equal.

19.dxe5 Bg4

Interesting is 19...Bh3, but White is fine replying 20.Qf1 (but not 20.Ne1? as both 20...Bxg2 and 20...f3 are strong).

20.Qc4+ Kh8 21.Ne1 f3 22.g3 Rf5?!

Black is slightly better after 22...Be6!?, according to the engines.
23.Qe4

Even stronger, according to the engines, is 23.Nd3, with Nf4 to come.

23...Re7?!

Preparing to gang up on the e pawn, but it is almost certainly too slow. The engines suggest 23...Rf7!?, although after 24.Qxg6 hxg6 the white king is no longer in danger, and White's pieces are a little more active.

24.Rd1 h5?

The engines recommend covering the back rank with 24...Re8!?, but much prefer White.

White to play and win

*****

*****

*****

*****

*****

25.h4

White is better after this, but the engines point out the forcing 25.h3! Bxh3 26.Rxc6! bxc6 (or 26...Qxc6) 27.Qh4, forking bishop and rook. White emerges with bishop and knight for rook and pawn, and has a large advantage, according to the engines.

25...Qe8 26.Nd3 Rf8

Black has no adequate answer to the threat of Nf4-d5, eg the engines reckon best is 26...Nd8!?, but 27.Nf4, followed by doubling rooks on the d file wins, and if 27...c6, to stop this, then 28.Nxh5! is strong and 28.e6!! is even stronger.

27.Nf4 Bf5 28.Qxf3 Kg8

The engines prefer 28...Bg4, but 29.Ng6+ Kg8 30.Qd5+ (30.Qxf8+!? is also good) Be6! 31.Qe4! wins the exchange.

29.Qd5+ Kh7 30.e6! Bg4

Or 30...Bxe6, when 31.Qd3+ Kg8 32.Rxh5 wins as 32...g6 loses to 33.Rh8+ Kf7 34.Qxg6#.

31.Qe4+ g6

White to play and lose

*****

*****

*****

*****

*****

32.Rd7??

White is roughly the equivalent of a queen up after 32.Rdd5, according to the engines, but other moves also win, including the simple 32.Re1.

The game finished:

32...Rxd7 33.Nxh5? Rd1+ 34.Be1 Qxe6 0-1
AM: "Sorry."

Gmund

GMUND is often referred to as Gmund am Tegernsee to differentiate from other places with the same name, gmund being commonly used in German to indicate a settlement at a rivermouth.

In this case Gmund is situated where the River Mangfall starts, the river being an outflow from the lake of Tegernsee.

The town, although it was then only a small village, is first mentioned in a 1075 report from nearby Tegernsee Abbey.

Railside meadow on Gmund's outskirts

Unlike many Bavarian tourist spots, murals are rare

But the train station is picturesque

And so are many other buildings

Fly-fisherman on the Mangfall

Churches with distinctive onion-style domes

Gasthof Herzog Maximilian has existed, albeit under various names, since at least 1339, guests including a Russian tsar and an Austrian emperor

Bavarian International Round Two

FACED a German-registered junior (born 2009).

Tobias Smith (1648) - Spanton (1858)

QGD Exchange

1.d4 d5 2.c4 e6 3.Nc3 Nf6 4.cxd5 exd5 5.Bg5 c6 6.e3 Bd6!?

Clearly the black dark-square bishop is more actively placed on d6 than after the main move, 6...Be7, but the question is whether White can take advantage of the pin on the black king's knight

*****

*****

*****

*****

*****

7.Qc2!?

This is much less popular than 7.Bd3, although the lines may easily transpose, but, for what it is worth, the text scores 64% in ChessBase's 2023 Mega database, while 7.Bd3 scores 58%.

7...h6!?

Perhaps Black should castle before playing this move, although it has been tried by players rated over 2400.

How should White respond?

*****

*****

*****

*****

*****

8.Bh4

This is the commonest continuation in Mega23, and it has been played by Grigoriy Oparin (2683), albeit at blitz, but probably critical is 8.Bxf6, the point being 8...Qxf6 runs into 9.Nxd5! etc. That leaves 8...gxf6, after which it is doubtful if Black's bishop-pair makes up for having a smashed kingside and problems finding a safe haven for the king.

8...0-0 9.Nf3 Re8 10.Be2!?

In the postmortem TS said he rejected 10.Bd3 because of the reply 10...Bg4. But White can boldly meet this with 11.h3!?, when 11...Bxf3 12.gxf3, followed by castling long, gives White promising kingside attacking chances.

10...Be6 11.0-0 Nbd7 12.Rab1 a5 13.a3 Qe7!?

This may be a novelty. Pedro Ramón Martínez Reyes (2386) - Aryan Tari (2656), Chess.com Blitz 2022, went 13...Qb8 14.b4 axb4 15.axb4 b5!? with equality, according to Stockfish16 and Komodo14.1 (0-1, 44 moves).

14.Na4 g6 15.Rfc1 Bf5 16.Bd3 Be4!? 17.Bxf6!?

Opting for simplification. Certainly not 17.Nd2? Bxh2+! 18.Kxh2 Ng4+ 19.Kh3 Qe6 etc.

17...Nxf6 18.Bxe4 Nxe4 19.Nc5!? Bxc5!?

This seems best. After ...

20.dxc5 a4

... Black has strong pressure against c5.

21.Nd2

How should Black proceed?

*****

*****

*****

*****

*****

21...Qf6

Probably not 21...Nxf2?! 22.Kxf2 Qx3+ 23.Kf1, even though Komodo14.1 reckons Black has sufficient compensation after 23...d4, 23...g5 and 23...Qf4+. However Stockfish16 disagrees, reckoning White is winning, and Komodo14.1 is not so keen on its suggestions once they have been inputted on the screen.

22.Nxe4 Rxe4 23.Qc3 Kg7 24.f3!?

This creates a new weakness, at e3, but relieves the pressure on c5.

24...Re5 25.b3

The engines prefer sitting tight with 25.Rc2, although they reckon 25...Rae8 may give an edge.

25...axb3

The engines give Black at least a slight edge after 25...Rae8!?

26.Rxb3 Ra7 27.Rcb1 Re7 28.Qd4 Qxd4 29.exd4 Re2

What should White play?

*****

*****

*****

*****

*****

30.R1b2?!

The game is equal after 30.R3b2 or 30.Rxb7, according to the engines.

What should Black play?

*****

*****

*****

*****

*****

30...Rxa3! 31.Rxb7?

White is a (backward) pawn down, but with drawing chances, after the engines' 31.h3. The problem with the text is that after ...

31...Ra1+ 32.Rb1 Raa2

... Black has unchallenged control of the seventh rank.

33.g4

The best try, according to the engines.

While I was thinking about my next move, TB offered a draw, but the engines reckon White is roughly the equivalent of a rook down.

33...Rxg2+ 34.Kh1

Even worse is 34.Kf1, when Black has a very pleasant choice between 34...Rxh2, forcing 35.Kg1, transposing to the game, and the even stronger 34...Raf2+ 35.Ke1 Rxh2, when the threat of ...Ra2 followed by mate cannot be reasonably met.

The game finished:

34...Rxh2+ 35.Kg1 Rhg2+ 36.Kh1 Rgd2 37.R7b6 Rxd4 38.Rxc6 Rc2 39.Rc7 Rd3 40.Rd7 Rxc5 41.Kg2 Rc2+ 42.Kg3 Rcc3 0-1

Autumnal Colours

MY previous visits to Tegernsee have been in the spring, when I have enjoyed lots of sunshine but also occasional snow.

This year there is no sign of snow, and the forecast was for rain most days.

But so far, apart from an overnight downpour, the days have been dry and largely sunny, although temperatures drop quickly in the evening.

Perhaps the biggest change, however, is seeing attractive autumnal colours.

Bavarian International Round One

Spanton (1858) - Matthias Schnieder (1594)

Pirc

1.e4 g6 2.d4 Bg7 3.Nc3 d6 4.f4 Nf6 5.Nc3 Bg4!?

There are 672 examples of this move in ChessBase's 2023 Mega database, although it trails in popularity 5...c6, 5...c5 and especially 5...0-0.

6.h3!?

In the postmortem MS said this is unnecessary as Black is more-or-less obliged to play ...Bxf3 as soon as the white queen moves. However the text is the main continuation in Mega23 and is quite liked by Stockfish16 and Komodo14.1.

6...Bxf3 7.Qxf3

Black has to deal with the threat of 8.e5

*****

*****

*****

*****

*****

7...c6 8.Be3 a5

Aimed at discouraging 0-0-0.

How should White proceed?

*****

*****

*****

*****

*****

9.Bd3

Six games in Mega23 reached the position in the second diagram. Three continued with the text, and three saw 9.0-0-0!?, which is preferred by the engines.

9...Na6 10.a3

The engines suggest 10.e5 or 10.0-0-0!?

10...0-0 11.0-0 Nd7!?

This is probably a novelty. The engines like the known-move 11...e6, albeit much preferring White.

12.Rad1 Qc8!? 13.f5 c5?!

The engines are not particularly concerned about this, but after ...

14.Nd5 Qd8 15.c3

... White's kingside play looks menacing, while the white queenside is not under pressure.

15...Nc7 16.g4!?

The engines prefer avoiding doubled pawns by 16.Nxc7 Qxc7 17.g4 (Stockfish16) or 16.Bc4!? Nxd5 17.Bxd5 (Komodo14.1).

16...Nxd5 17.exd5 cxd4 18.Bxd4 Ne5

The knight is a monster on e5, but the engines prefer 18...Bxd4+!?, claiming 19.cxd4 leaves White only slightly better.

19.Qe3 Qd7 20.Rf2 b5 21.Rdf1 b4!? 22.axb4 axb4

What should White play?

*****

*****

*****

*****

*****

23.cxb4!?

The engines like the text and 23.c4 almost equally.

23...Nxd3 24.Qxd3 Bxd4+ 25.Qxd4 f6

Stockfish16 claims 25...Qa7!? equalises, while Komodo14.1 reckons it leaves White with the upper hand.

26.Qd3?!

Black is under more pressure after 26.fxg6 hxg6 27.Qe4.

26...g5 27.h4 h6 28.h5?

Too slow - Black now gets a strong initiative on the queenside

*****

*****

*****

*****

*****
28...Rfb8 29.Qb3?

Now White is lost, according to the engines. Counterplay with 29.Qe4 is called for.

29...Ra4

The b4 pawn must fall, equalising material and leaving White with three weak pawns (b2, d5 and g4).

30.Rg2 Raxb4

Even stronger, according to the engines, is 30...Rbxb4.

31.Qc3 Qb7 32.Re1!?

This is probably better than passively defending d5 with 32.Rd1.

32...Rc8!?

Can you find a strong combination for White?

*****

*****

*****

*****

*****

33.Rxe7??

The engines give 33.Qe3, but after 33...Kf8 Black has the safer king, can pick up the b pawn more-at-less at will and  according to the engines, is positionally winning.

Black to play and win

*****

*****

*****

*****

*****

33...Qb6+??

Winning is 33...Rxc3 34.Rxb7 Rc1+ etc.

34.Qe3 Rxb2 35.Rxb2 Rc1+ 36.Kg2 Qxb2+ 37.Qe2 Rc2 38.Re8+ ½–½

Good Walking

MY hotel is on the northeast shore of the lake, while the venue, Gut Kaltenbrunn, a former monastery and now a large hostelry, is on the northwest shore.

It is less than a half-hour walk between the two, with good views across the Tegernsee.

Early morning

Since I had plenty of time, I walked on to Bad Wiessee, which is proabaly the prettiest of the lake's towns.

The tourist-information office was taking tournament-registration confirmations, and I knew there would be places to have an espresso.

This carving seems new - I cannot remember it being in Bad Wiessee last year

The latest list of entries has 520 players, including 13 grandmasters.

Lakeside Chess

ARRIVED in the Bavarian town of Gmund am Tegernsee this evening after quite an adventure.

The journey via Heathrow and Munich airports should have been relatively painless.

But due to multiple railworks and other delays, I had to catch five trains and two buses in addition to my flight from London.

I am in Gmund for the 26th edition of a tournament that this year has attracted 534 entries, including 15 grandmasters, although I seem to be the only player from the UK.

There are nine rounds over nine days, starting tomorrow, with all rounds at 16:00, apart from the final round at 10:00.

The time limit is 40 moves in 90 minutes, with 30 minutes to finish, and a 30-second increment throughout. Pairings will be accelerated for the first five rounds.

Tegernsee is the name of a town and a valley, but is best known as the name of a four-mile-long lake that Germans have used for recreation for hundreds of years.

I have played at towns beside the lake three times before, but in a seniors event held in the spring.

This will be a new experience, partly prompted because the tournament has been highly recommended to me by several German players.

Battersea Season 2023-4

DATE.....EVENT..........................COL..RATING..OPP'S RATING..SCORE..SEASON PERF

24/10/23 London League              B      1870         2102                     =            2102

26/10/23 Central London League W     1870         2118*                    =            2110

*Corrected rating.

Central London League

PLAYED for Battersea on board three (of five) against HMC 2 in division one last night.

Spanton (1870) - Mei-Xian Eunice Hng (2003*)

Maróczy Bind

1.e4 d6 2.c4!? Nf6 3.Nc3 c5 4.Nge2 Nc6 5.d4 cxd4 6.Nxd4 Bd7 67Be2 g6 8.Be3 Bg7

This position, which occurs 1,759 times in ChessBase's 2023 Mega database, normally arises from a Sicilian Accelerated Dragon move-order

*****

*****

*****

*****

*****

9.0-0 0-0

We have now reached a position occurring 13,456 times in Mega23.

10.Rc1

This is a popular alternative to the mainline, which runs 10.Qd2 Nxd4 11.Bxd4 Bc6 12.f3 with a slight edge for White, according to Stockfish16 and Komodo14.1.

10...Rc8 11.Qd2 Re8 12.b3 a6 13.h3 Qa5

Now both players have connected their rooks, how would you assess the position?

*****

*****

*****

*****

*****

White's extra space gives at least a slight edge, according to the engines. However White has to deal with Black's immediate tactical threat of 14...Nxe4!, winning a pawn, eg 15.Nxe4 Qxd2 etc or, more complicatedly, 15.Nxc6 Qxc3! 16.Qd1! Qb2 17.Rc2 Nc3! 18.Qd3 Nxe2+ 19.Rxe2 Qc3 20.Qxc3 Bxc3.

14.Rfd1 Nxd4 15.Bxd4 b5

How should White meet this challenge to the Bind?

*****

*****

*****

*****

*****

16.Bxf6!?

Seeking to take advantage of the unprotected status of the black queen. Wolfgang Glaser (2120) - Aleksandar Vučković (2280), German Team League 1996, went 16.f3 bxc4 17.bxc4?! Be6 with a slight edge for Black, according to the engines (0-1, 41 moves). The text is probably a novelty.

16...Bxf6 17.Nd5 Qxd2 18.Nxf6+ exf6 19.Rxd2 Rxe4 20.Bf3 Re6 21.c5!?

As my opponent pointed out afterwards, 21.Bb7 Rc7 22.Bxa6?! loses the bishop to 22...d5!? 23.Bxb5 Bxb5. True, White gets three connected passed pawns after 24.Rxd5, but the engines reckon Black is much better.

21...Be8!?

Possibly better is 21...Rc7 22.c6 Kf8, when White has full compensation for a pawn, but no more, according to the engines.

22.c6 Rc7 23.b4 Kf8 24.Rd3?!

Hoping to go after the a6 pawn, but this is almost certainly too slow. The engines give 24.Bd5 Re5 25.f4 Re3 26.Bf3 Re6 27.Rdc2, claiming White may have a tiny edge.

24...d5! 25.Rxd5?!

Probably better is 25.Bxd5, eg 25...Rd6 26.Be4 Rxd3 27.Bxd3 Rxc6 28.Rxc6 Bxc6 with a completely equal position, according to the engines.

25...Rexc6 26.Rcd1 Rc1 27.Kf1 Rxd1+ 28.Rxd1 Rc4 29.Rd6 Rxv4 30.Rxa6 Rb2 31.Ra7 f5 32.Bd5 Rd2 33.Bb3 h5 34.Rb7 ½–½

Black is at least slightly better in the final position, according to the engines, but was well behind on the clock

HMC won the match 3-2.

*Fide rating - no ECF

Pleasing Finish To Mallorca

Spanton (1864) - Barry Foran (1592)

Calvià U2350

Caro-Kann Panov-Botvinnik Attack

1.e4 c6 2.d4 d5 3.exd5 cxd5 4.c4 Nf6 5.Nc3 e6

This is the main move in ChessBase's 2023 Mega database, although also very popular is 5...Nc6.

6.Nf3 Be7

Again this is the main move in Mega23, but the more-active 6...Bb4 is also often seen, the chief continuation being 7.cxd5 Nxd5 8.Bd2 Nc6 9.Bd3 0-0!? 10.0-0 (not 10.Bxh7+?? Kxh7 11.Ng5+ Kg8 12.Qh5 Nf6) Be7!? 11.a3!? with a slight edge for White, according to Stockfish16 and Komodo14.1.

7.c5!?

Anatoly Karpov & Mikhail Podgaets, in Caro-Kann Defence: Panv Attack (Batsford 2006), wrote of this move: "Never very popular, and now a virtually forgotten continuation. The plans of the two sides are clear. White intends to construct a pawn-chain b4-c5-d4 and on the basis of this carry out an advance on the queen's flank, while Black will set about undermining this pawn-chain (...a5 and ...b6) and also undertake counterplay in the centre: ...Ne4 and ...f5."

7...0-0 8.Bf4

Karpov & Podgaets: "Containing more ideas than 8.Bg5, but also not without its downsides. The bishop, standing on f4, can come under the tempo-gaining attack ...g5."

8...b6 9.b4 a5

Karpov & Podgaets: "A higher ranking move than 9...bxc5."

10.a3

How should Black proceed?

*****

*****

*****

*****

*****
10... bxc5?!

Karpov & Podgaets mention 10...Ne4, but concentrate on what is the mainline in Mega23, namely 10...axb4 11.axb4 Rxa1 12.Qxa1 Nc6 13.Qa4 bxc5!? They reckon White can accept the piece sacrifice as "the variations show the risk is hardly excessive," but they also state that 14.bxc5 is a quiet way to get an "approximately equal game." The engines agree with these assessments.

The problem with the text is White is better placed to use the opened b file.

11.bxc5 Bd7 12.Rb1 Ne4 13.Nxe4 dxe4 14.Ne5 Nc6 15.Rb6 Nxe5 16.Bxe5

What should Black play?

*****

*****

*****

*****

*****

16...f5?

This protects the loose e pawn but fatally weakens Black's other e pawn. The engines recommend 16...Bf6 or 16...Qc8, but give White the upper hand.

17.Bc4 Qc8 18.Qb3 Qxc5?

Not 18...Kf7? 19.d5, so Black should probably shuffle the king into the corner and accept White can go a pawn up.

The game finished:

19.dxc5 Qxc5 20.Bd6 Qd4 21.Bxf8 Kxf8 22.0-0 1-0

New Season, New Hopes

PLAYED for Battersea on board five (of eight) against Streatham & Brixton in London League Division One last night.

James Toon (2102) - Spanton (1870)

English Symmetrical

1.c4 c5 2.Nf3 Nc6 3.g3 g6 4.Bg2 Bg7 5.0-0 d6 6.Nc3 e5 7.d3 Nge7

Black has a Botvinnik setup, the key points being the V pawn-formation in the centre and the position of the king's knight at e7 - there are 3,320 examples of this position in ChessBase's 2023 Mega database

*****

*****

*****

*****

*****

8.a3 0-0 9.Rb1 f5!?

More popular, and preferred by Stockfish16 and Komodo14.1, is 9...a5.

10.Bd2

Not 10.b4? as Black can reply 10...e4, although there are six examples of White playing the move in Mega23, including one by a 2245.

10...b6 11.b4 Bb7

How should White proceed?

*****

*****

*****

*****

*****

12.Ne1!?

This may be a novelty, at least in this exact position. Mega23 has two games with 12.Ng5 and two with 12.Nd5. According to the engines the former should be met by 12...Qd7, with equality, and the latter by 12...Nxd5 13.cxd5 Nd4, again with equality (Komodo14.1) or possibly a slight edge for White (Stockfish16). The engines give best-play as 12.b5!? Na5 13.e4 Qd7, claiming White has the upper hand (Stockfish16) or at least a slight edge (Komodo14.1).

12...Qd7 13.Nc2

The game has transposed to a known position in which 13...Rae8 and 13...f4 have been played, but I preferred ...

13...Nd4

... as White is hard-pressed to avoid an exchange of light-square bishops.

14.Bxb7

Preserving the bishop by Bh3 or f3 looks artificial, while 14.e4!? Nxc2 15.Qxc2 f4 is also problematic.

14...Qxb7 15.Ne3!? f4 16.Ned5 Nxd5 17.Nxd5 f3 18.exf3 Nxf3+ 19.Kg2

The white king's position has clearly been weakened, but the knight at d5 is strongly placed for defence and attack

*****

*****

*****

*****

*****

19...Rf5 20.Be3 Raf8!?

Preceding this move with 20...cxb4 may be stronger.

21.bxc5 Rh5

Threatening mate-in-one.

22.h4 Qd7?!

After 22...dxc5 23.Bxc5 Rf7 Black has full compensation for a pawn, according to the engines.

23.Rh1?

White has at least the upper hand after 23.cxb6, according to the engines, eg 23...Qg4 can be met by 24.Rh1, while 23...Rxh4!? seems to come unstuck against 24.Rh1!

23...dxc5 24.Bg5 h6 25.Bf6!?

How should Black respond?

*****

*****

*****

*****

*****

25...Bxf6

After 25...Nd4!? the black king's rook appears trapped thanks to 26.Bxg7 Kxg7 27.g4. However one continuation by the engines runs 27...Rf3! 28.Ne3 Qb7 29.Kg1!? Rf7 30.gxh5 Nf3+ 31.Kf1 Nd2+ 32.Ke1! Qxh1+ 33.Kxd2 Qxh4!? 34.f3 Qxh5 35.Qe2 Qxf3. although they reckon White's knight is better than Black's three pawns. If in this line 28.gxh5, then 28...Qf5! can to lead to a perpetual, eg 29.Qe1 Rxd3 30.Rh2 Qg4+ 31.Kh1 Qf3+ etc.

Instead the engines give 25...e4! 26.dxe4 Rxd5! 27.Qxd5+ Qxd5 28.exd5 Rxf6, claiming a slight edge for Black.

26.Qxf3 Rf5

Not 26...Bxh4? as White is on top after 27.Qe2.

27.Qe4 Bg7 28.Rhf1 g5?!

The engines give 28...h5 or 28...R8f7, but much prefer White.

29.hxg5

Even stronger seems to be 29.Ne3, forcing the black king's rook from f5, and then taking on g5 at a time when Black has to recapture with the h pawn, eg 29...R5f7 30.hxg5 hxg5 31.Nxd5 with what the engines reckon is a positionally won game.

29...Rxg5 30.f3 Rf7

Interesting is the engines' 30...Rf4!? as 31.Nxe4 exf4 gives Black decent compensation for the exchange. However the engines reckon White is better after 31.Qe2!? Rd4 32.Rh1.

31.Rh1 Qa4!?

This pawn raid was the main point behind 30...Rf7.

32.Rbf1 Qxa3 33.Rf2 Qa4 34.Rh4 Qd7

The black queen has grabbed a pawn and returned home, but White is still slightly better, according to the engines

*****

*****

*****

*****

*****

35.f4 Rg4!?

This is Komodo14.1's top choice. Stockfish16 prefers 35...exf4 36.Rhxf4 Rxf4 37.Rxf4, which frees the bishop but leaves the black king looking very exposed.

36.Rxg4 Qxg4 37.Ne3 Qd7 38.f5 Qd4 39.Nd5 Qxe4+ 40.dxe4 Bf6 ½–½

In the final position White has at least full compensation for a pawn, according to the engines, but JT was behind on the clock

Streatham & Brixton won the match 5-3.