The move 7. 0-0 is interesting and seems to have had some success lately.
Does anyone have any suggestions about how to best play against it?
French Advance: Milner-Barry Gambit
Sort:
Brazilian GM Luis Supi butchered the Uruguayan IM Rodi with the modernized Milner-Barry. Supi delayed as much as he can move the Queen's knight.
Here the game: https://www.chess.com/pt/events/2022-sales-oliveira-masters/09/Supi_Luis_Paulo-Rodi_Luis_Ernesto
The only times the MBG does well for white, is when black is careless. The truth is that white is just worse out of the opening (about -0,60 after 8 moves) and even has to be careful to not lose, let alone have an advantage. All black has to do is play a couple of careful but sensible moves and then he's just clearly better.
Don´t pick an opening variation and hope your opponent will be careless. Just pick a decent variation. Play 3. N.c3 or 3. N.d2 for starters and actually try to put black under pressure.
7.dxc5 is almost half a pawn stronger and more importantly, performs at 58:38 in 35k games versus 7.0-0's 52:45 at 86k games.
if you're under 2000, this is the most promising line. 8...Nge7? is the main line at 71% wins to 26% losses in 806 games and 8...a5's 54% to 39% stats in 682 games aren't much better. i'd switch to this "low hanging fruit" line myself.
1.e4 e6 2.d4 d5 3.e5 c5 4.c3 Qb6 5.Nf3 Nc6 6.Bd3 Bd7 7.dxc5 bxc5 8.Qe2
8...Nge7? 9.b4!
9...Nxb4 10.cxb4 Bxb4 11.Nbd2 +5.4
9...Bxb4 10.cxb4 +4.2!
10...Nxb4 11.Nc3 +4.2
10...Qxb4+ 11.Nbd2 +4.8
8...a5 9.Nbd2
9...a4 10.b4 axb3 11.Nxb3 +0.6
9...Nge7? 10.Nb3
10...Ng6 11.h4 +1.9
10...a4 11.Nxc5 Qxc5 12.0-0 +1.4
3.Nc3 and 3.Nd2, as mentioned, perform 4% and 5% better for a head start on even more promising lines, but i chose to stay in your gambit
1983B-Boy escreveu:
7.dxc5 is almost half a pawn stronger and more importantly, performs at 58:38 in 35k games versus 7.0-0's 52:45 at 86k games.
if you're under 2000, this is the most promising line. 8...Nge7? is the main line at 71% wins to 26% losses in 806 games and 8...a5's 54% to 39% stats in 682 games aren't much better. i'd switch to this "low hanging fruit" line myself.
1.e4 e6 2.d4 d5 3.e5 c5 4.c3 Qb6 5.Nf3 Nc6 6.Bd3 Bd7 7.dxc5 bxc5 8.Qe2
8...Nge7? 9.b4!
9...Nxb4 10.cxb4 Bxb4 11.Nbd2 +5.4
9...Bxb4 10.cxb4 +4.2!
10...Nxb4 11.Nc3 +4.2
10...Qxb4+ 11.Nbd2 +4.8
8...a5 9.Nbd2
9...a4 10.b4 axb3 11.Nxb3 +0.6
9...Nge7? 10.Nb3
10...Ng6 11.h4 +1.9
10...a4 11.Nxc5 Qxc5 12.0-0 +1.4
3.Nc3 and 3.Nd2, as mentioned, perform 4% and 5% better for a head start on even more promising lines, but i chose to stay in your gambit
After 6. Bd3, the move is cxd4
QUOTE: After 6. Bd3, the move is cxd4
6...Bd7 appears in 150,000 games vs 101,000 for 6...cxd5, but I'll get some theory for that too.
Lastrank wrote:
The move 7. 0-0 is interesting and seems to have had some success lately.
Does anyone have any suggestions about how to best play against it?
I should have given all the moves of the variation I was inquiring about.
1. e4 e6 2. d4 d5 3. e5 c5 4.c3 Nc6 5. Nf3 Qb6 6. Bd3 cxd4 7. 0-0.
I haven't looked at this in a long time, but I thought that 6...Bd7 is an inaccuracy. 6...cxd4 is more precise. After 6...Bd7 White can play 7.dxc5 Bxc5 8.Qe2 (or maybe O-O) with some advantage.
Lately, after 1.e4 e6 2.d4 d5 3.e5 c5 4.c3 Nc6 5.Nf3 Qb6 6.Bd3 cxd4 Some White players have been trying 7.O-O !? MVL played this against PonKratov in the FIDE Grand Swiss. Clarissa Yip won a couple of games in the US Women's Championship in 2021 with a similar approach
OK... a lot more involved than the first variation without as much good news, but very playable
1. e4 e6 2. d4 d5 3. e5 c5 4. c3 Qb6 5. Nf3 Nc6 6. Bd3 cxd4 7. O-O
7...dxc3 8.Nxc3 56:40 at 3,212 +0.5
8...Bd7 9.Re1 55:41 at 376 +0.2
9...Bb4? +1.5
10.a3 56:41 at 313 +1.2
10.Be3 54:43 at 391 +1.2
10.Ng5 61:38 at 112 +0.2 with initiative maybe?
9...Nge7 10.Be3 74:24 at 364 +0.2
10...Qc7? 11.Nb5 96% at 105 +8.5
10...Qd8 +0.2
11.Bc5 76:24 at 17 =
11.Bg5 73:27 at 15 +0.3
10...d4? 11.Nxd4 Nxd4 12.Bxd4 Qxd4? 13.Nb5 Qb6 14.Nd6+ Kd8 15.Nxf7+ Ke8? 16.Nxh8 g6
17.Qf3 +9.2
10...Qxb2? 11.Nb5 93% at 40 +6.5
9...a6 10.Be3 56:43 at 260 +0.5
10...Qc7? 11.Rc1 64:36 at 61 +1.1
11...Nge7 12.h4 +1.3
11...Bb4 12.a3 bxc3 13.Rxc3 Nge7 14.Bc5 +1.9
10...Qd8 11.Rc1 Nge7 12.Na4 Ng6 13.Bb6 +1.1
10...d4? 11.Nxd4 Nxd4 12.Qg4 +1.0 (12.Bc2!? +1.2)
10...Qxb2? 11.Na4 Qa3 12.Nb6 Rd8 13.Rb1 +1.7
9...Bc5? 10.Na4 66:30 at 140 +1.8
10...Bxf2+? 11.Kf1 74% at 58 +4.9
10...Qa5 11.Nxc5 Qxc5 12.Be3 Qa5 13.a3 Nge7 14.b4 +1.6
8...Nge7 9.Re1 Ng6 10.h4 +0.6
10...Bc5? 11.Na4 +1.9
11...Bxf2+ 12.Kf1 +2.4
11...Qa5? 12.h5 Nf8 13.Nxc5 Qxc5 14.a3 +3.3 untested
10...Be7 11.g3 a6 12.h5 Nf8 13.Be3 is +0.6 but untested
10...Qc7? 11.h5 Ngxe5 12.Nxe5 Nxe5 13.Nb5 +5.8
10...Bb4? 11.h5 Nge7 12.a3 Bxc3 13.bxc3 +1.7
8...a6 9.Be3 57:38 at 149 +1.1, but 9.Re1 +0.1 scores slightly better
9...Qc7 10.Rc1 63:33 at 51 +1.3
10...Nge7 11.b4! +1.6
10...Bd7 11.Na4 +1.1
9...Qd8 10.Na4 Nge7? 11.Bb6 Qd7 12.Nc5 +7.8
9...Qxb2? 10.Na4 Qa3 11.Nb6 Rb8 12.Qc2 +2.3
8...Bb4? 9.Be3! 69:29 at 124 +1.1
9...Qc7 10.Nb5 67:30 at 33 +1.2
10...Qd8 +1.4
11.a3 +1.3 and 46:46 at 13
11.Nfd4 Bf8 12.Qb3 Bd7 13.Rfc1 +1.4 and untested
10...Qd7? +2.6
11.a3 86% at 7 +1.5
11.Qa4 Nh6 12.Rac1 a6 13.Nfd4 +2.6 is untested
10...Qb8 +1.8
11.a3 83% of 6 games +0.8
11.Nfd4 Nh6 12.Qa4 Be7 13.Nxc6 +1.8 but untested
9...d4? 10.Nxd4 Nxd4? 11.Bxd4 +6.3
9...Qd8 10.Rc1 +1.0
9...Bc5? 10.Na4 +4.6
7...Bd7 8.Qe2!? 57:39 at 900 games =
8...dxc3 9.bxc3 =
9...Nge7 =
10.Bf4 Ng6 11.Bg3
11...Be7 12.Nbd2 0-0 13.Rab1 is 73:27 in 11 games and =
11...Bc5 12.Nbd2 0-0 13.Rab1 Qc7 14.c4 =
10.Nbd2 Qc7 11.Nb3 Nc8 12.a4 = (stockfish)
9...Bc5 10.Nbd2 Nge7 11.Nb3 Ng6? 12.Nxc5 Qxc5 13.Rb1 +0.8
9...a6 +0.3
10.Be3 is -0.7, but won 4 out of 5 games
10.Nbd2 Nge7 11.Nb3 Nc8 12.Re1 is +0.3 and is stockfish's untested choice
8...Nge7 9.Nbd2 =
9...dxc3 10.bxc3 Qc7 11.Nb3 Ng6 12.Re1 Be7 13.Nbd4 =
9...Ng6 10.Nb3 dxc3 11.bxc3 =
what computers see and people see are two different things. i see a lot of lines where the lower evaluated one actually performs better. i just saw that very phenomenon finishing the new theory up...
1. e4 e6 2. d4 d5 3. e5 c5 4. c3 Qb6 5. Nf3 Nc6 6. Bd3 cxd4 7. O-O dxc3 8.Nxc3 Bd7 9.Re1 Bb4? +1.5
10.a3 56:41 at 313 +1.2
10.Be3 54:43 at 391 +1.2
10.Ng5 61:38 at 112 +0.2 with initiative maybe? down a pawn and kicking much butt.
there's one line i really want to find again where one side is evaluated at something like +20, but is actually only winning something like 35% of games. i trust performance stats better than any computer OR grandmaster evaluation! the stats tell you what people are actually going to play and what happens after. no time gets wasted on GM lines you'll never see over the board.
i'm going to kick the hornets nest in a minute showing how gambits are actually stronger than main lines under 2000 despite all the WRONGness trying to warn amateurs away from gambits. my stats in the monte carlo exchange are already winning without theory! i was never able to beat the french before, but my stats there and in king's gambit are at least +5%. nothing is as satisfying as smacking a smug 1700+ down in under 20 moves with an "unsound gambit" they've no respect for.
Here is a modern twist to the Gambit
https://www.chessgames.com/perl/chessgame?gid=2097003