Showing posts with label Benko Gambit. Show all posts
Showing posts with label Benko Gambit. Show all posts

Wednesday, 15 February 2023

Beat The ... Benko Gambit

HERE is another piece I wrote in the covid panic but but did not publish.

In this series I am looking at the statistically best way to play against popular opening lines.
The numbers are drawn from the 2021 edition of ChessBase's Mega database, ignoring, where possible, those results that include very few games and so are statistically insignificant.

The Benko Gambit: 1.d4 Nf6 2.c4 c5 3.d5 b5!? is especially popular at club level.
White scores a disappointing 52% by accepting with 4.cxb5, but an impressive 60% by declining with 4.Nf3.
Position after 4.Nf3
Black has six popular replies.

A) 4...g6 (2,378 games)
After 5.Qc2 the line splits.
A1 5...bxc4 6.e4, after which the line splits again.
A1.1 6...d6 7.Bxc4 Bg7 8.0-0 0-0 9.Re1!?, when A1.1a 9...Nbd7 10.h3 Nb6 11.Na3!? scores 79% for White, albeit from a small sample, and A1.1b 9...Ba6 10.Na3 Nfd7!? 11.Rb1 scores 80% for White, albeit from a small sample.
A1.2 6...Bg7 7.Bxc4, when A1.2a 7...d6 is a transposition to A1.1, and A1.2b 7...0-0 8.0-0 d6 is also a transposition to A1.1.
A2 5...d6 6.e4 Bg7 (6...bxc4 is a transposition to A1.1) 7.cxb5!?, after which the line splits again.
A2.1 7...a6 8.Nc3 0-0 9.a4, when A2.1a 9...axb5 10.Bxb5 Na6 11.h3 scores 88% for White, albeit from a small sample, and A2.1b 9...e6 10.dxe6 scores 75% for White.
A2.2 7...0-0 8.Nc3 a6 is a transposition to A2.1.
A3 5...Bg7 6.e4, after which the line splits again.
A3.1 6...d6 is a transposition to A2.
A3.1 6...0-0 7.cxb5 a6 8.a4 scores 58% for White, albeit from a small sample.

B) 4...Bb7 (1,350 games)
After 5.Nbd2 the line splits.
B1 5...bxc4 6.e4 e6 7.dxe6,  after which the line splits again.
B1.1 7...dxe6 8.Bxc4 scores 75% for White, albeit from a small sample.
B1.2 7...fxe6 8.e5 Nd5 9.Nxc4 scores 57% for White, albeit from a small sample.
B2 5...e6 6.dxe6!? fxe6 (there are no games with 6...dxe6!? in Mega21) 7.e4!? scores 75% for White, but from just two games. Those games continued 7...Nxe4 8.Nxe4 Bxe4, when Stockfish13 and Komodo12.1.1 like the 9.Ng5 of Dorian Rogozencko (2490).

C) 4...bxc4 (770 games)
After 5.Nc3 the line splits.
C1 5...d6 6.e4, after which the line splits again.
C1.1 6...g6 7.e5!? dxe5 8.Nxe5 Bg7 9.Bxc4 0-0 10.0-0, when C1.1a 10...Nfd7!? 11.Nxf7! Rxf7 12.d6 scores 76% for White, and C1.1b 10...Bb7 11.Re1 scores 80% for White.
C1.2 6...Nbd7 7.Bxc4 g6 8.0-0 Bg7 9.a4!? 0-0 10.h3 Rb8 11.Re1 scores 57% for White, albeit from a small sample.
C2 5...g6 6.e4 d6 is a transposition to C1.1.

D) 4...d6 (604 games)
After 5.Qc2!? the line splits.
D1 5...g6 is a transposition to A2.
D2 5...bxc4 6.e4 g6 is a transposition to A1.1.

E) 4...e6 (435 games)
After 5.Bg5 the line splits.
E1 5...exd5 6.cxd5, after which the line splits again.
E1.1 6...d6 7.e4 a6 8.a4, when E1.1a 8...Be7 9.Bxf6!? Bxf6 10.Qc2 scores 71% for White, and E1.1b 8...b4 9.Nbd2 Be7 10.Be2 scores 83% for White, albeit from a very small sample.
E1.2 6...h6 7.Bxf6 Qxf6 8.Qc2, when E1.2a 8...d6 9.e4 a6 10.a4 b4 11.Bd3!? scores 80% for White, albeit from a small sample, and E1.2b 8...Be7 9.e4 Qb6!? 10.Be2 0-0 11.0-0 scores 67% for White, albeit from a very small sample.
E1.3 6...Qa5?! 7.Nc3!? Ne4 8.Bd2 Nxd2 9.Nxd2 scores 82% for White.
E2 5...Qa5+ 6.Nbd2!?, after which the line splits again.
E2.1 6...bxc4 7.Bxf6 gxf6 8.e4 Rg8 9.g3 scores 79% for White, albeit from a small sample.
E2.2 6...Ne4?! 7.b4! Qxb4 8.Rb1 Qc3 9.Rb3 Qa5 10.Rxb5 Qc3 11.Rb3 Qa5 12.Bf4 scores 100% for White, albeit from a small sample.
E3 5...h6 6.Bxf6 Qxf6 7.Nc3 b4 8.Nb5, after which the line splits again.
E3.1 8...Kd8 9.a3!? scores 83% for White, albeit from a small sample.
E3.2 8...Na6 9.e4 e5 10.a3 scores 79% for White, albeit from a small sample.
E4 5...b4 6.e4 d6 7.a3!? scores 65% for White, albeit from a small sample.
E5 5...bxc4 6.Nc3, after which the line splits again.
E5.1 6...d6 7.e4, when E5.1a 7...e5 reaches a position where both 8.Bxc4 and 8.Nd2!? score 56%, albeit from small samples, and E1.5b 7...Be7 8.Bxc4 scores 68% for White.
E5.2 6...Qb6 7.e4!? scores 100% for White, albeit from a small sample.

F) 4...b4 (380 games)
After 5.b3!? the line splits.
F1 5...g6 6.Bb2 Bg7 7.e4 scores 80% for White, albeit from a small sample.
F2 5...d6 6.Bb2 g6 7.Nbd2 Bg7 8.g3 0-0 9.Bg2 a5 10.0-0 scores 75% for White, albeit from a very small sample.

Wednesday, 12 February 2020

Happy Birthday

WHILE playing in the 4NCL over the weekend at Maidenhead, Berks, I browsed the Chess & Bridge bookstall.
Events with bookstalls are much rarer than they used to be, but I still enjoy seeing what is new.
On this occasion I came away with something (relatively) old: American Grandmaster - Four Decades Of Chess Adventures by Joel Benjamin.
The book, whose title is presumably meant to remind prospective readers that Benjamin is a US-born grandmaster, not a Johnny-come-lately foreign transfer, was published by Everyman Press in 2007.
My copy is a 2008 reprint, reduced in price from £14.99 to £7.50, and, from what I have read so far, contains just the right mix of career history, anecdotes and annotated games.
I am regarding it as a birthday present to myself, today being my 63rd birthday.
A red, white and blue American Grandmaster
Here is one of the book's earlier games. Notes in italics are by Benjamin.
Jay Whitehead (2325) - Benjamin (2475)
US Championship (Greenville, Pennsylvania) 1983
Benko Gambit
1.d4 Nf6 2.c4 c5 3.d5 b5 4.Nd2!?
One of those anti-Benko systems that enjoyed its 15 minutes of fame.
The move has since been played by Kramnik, Akopian, Dreev and other strong GMs.
4...d6 5.e4 bxc4 6.Bxc4 g6 7.b3 Bg7 8.Bb2 0-0 9.Ngf3 e5!?
There is only one earlier game in ChessBase's 2020 Mega database, and that saw 9...Nbd7, which became the main line. Benjamin's choice was later played by Ivanchuk.
10.dxe6 fxe6 11.e5?!
This doesn't really make sense because Black can pin. 11.0-0 Nc6 is roughly equal.
11...Nh5
11...Ng4 12.h3 Nh6 13.Qc2 d5 gives Black a comfortable edge.
Stockfish10 and Komodo10 agree with Benjamin's analysis, although White went to win when the moves were played in Baris Esen (2571) - Johnatan Bakalchuk (2272), EU Championship 2015.
12.Qc2 d5 13.Bd3 Nd7 14.g3 Bb7 15.0-0 Rc8 16.Rac1
How should Black proceed?
*****
*****
*****
*****
*****
16...Nf4!?
Benjamin gives this !!?, saying: I can only marvel at the optimism of my past self. Black will only get a pawn for a piece, but White's position is tricky to play.
The tournament bulletin awards the move ?! (I have sort of split the difference), recommending 16...Qe7.
17.gxf4 Rxf4 18.Rfe1 Qf8
An exclamation mark is given to White's 18th move by the bulletin; to Black's 18th move by Benjamin.
19.Qd1 Bh6 20.h3
White has no time to reposition his bishop: 20.Bf1 d4 21.Bg2 Rg4 22.h3 Rxg2+ and wins [eg 23.Kxg2 Bxd2 24.Qxd2 Qxf3+ and mates].
20...Rf7?!
With the bishop uncovered, White still can't reorganise.
The engines much prefer 20...Qe7, eg 21.Bf1 (the move Benjamin's choice is aimed at preventing) Rcf8 22.Bg2 d4 with an unclear position that the engines reckon narrowly favours White.
21.Qe2?!
Better, it seems, is the engines' 21.Rc2, eg 21...Bxd2 22.Nxd2 Rxf2, when Black has a second pawn for his piece, but 23.Qg4 gives White the upper hand. Black can perhaps improve with 21...Qe7, but the engines' 22.Nh2 looks good, eg 22...Rcf8 23.Ndf3, and the white king seems to have adequate defences.
21...Qe7 22.Rc2
[Robert] Byrne pointed out an interesting regrouping: 22.Nf1!? Rcf8 23.N1h2, since if now 23...Bxc1 24.Bxc1 White gains control of squares on the kingside.
The engines reckon the above line is problematic for White if Black, instead of grabbing the exchange, plays 23...Rf4.
22...Rcf8 23.Nh2??
Whitehead finally loses his patience. After 23.Rf1 Black's attack is still held at bay.
The engines agree 23.Rf1 is much better than the text, but reckon Black is at least equal after 23...d4 or 23...Rf4.
23...Rxf2 24.Qxf2 Rxf2 25.Kxf2 Qh4+
The white pieces have no footholds. It's just a matter if time before Black picks one off.
26.Ke2 Qxh3 27.Nhf3 d4 28.Rf1
Not 28.Be4? Bxe4 29.Nxe4 d3+.
28...Qg4 29.Nc4 Bf4 30.Rf2 Bg3 31.Rf1 Bxe5 32.Ncd2 Bf4 33.Rg1 Qh5 34.Be4?
But 34.Ne4 Kg7 would win in due time.
34...Bxe4 35.Nxe4 d3+ 0-1