Busted French line?

[jdart]

I have a bunch of opening books. Hiarcs book recommends 5. Nf3. A line from that book, also found in the Chessbase French Powerbook 2018, goes:

1. e4 e6 2. d4 d5 3. Nc3 Nf6 4. e5 Ng8 5. Nf3 Ne7 6. Bd3 Nbc6 7. O-O Nb4 8. Be2
Nbc6 9. Qd3

with very good stats for White.

[supersharp77]

I have the results of a long analysis with Oct 5 SF dev, nn-baeb9ef2d183.nnue, 450 GB of cache, most important 7-piece EGTBs. The TCEC game that SF lost was:
[Event "TCEC Season 19 - Superfinal"]
[Site "https://tcec-chess.com"]
[Date "2020.10.01"]
[Round "5.1"]
[White "LCZero v0.26.3-rc1_T60.SV.JH.92-190"]
[Black "Stockfish 202009282242_nn-baeb9ef2d183"]
[Result "1-0"]
[BlackElo "3564"]
[ECO "C11"]
[GameDuration "04:28:25"]
[GameEndTime "2020-10-01T09:34:42.529 UTC"]
[GameStartTime "2020-10-01T05:06:17.274 UTC"]
[Opening "French"]
[PlyCount "178"]
[Termination "adjudication"]
[TerminationDetails "TCEC win rule"]
[TimeControl "7200+10"]
[Variation "Steinitz variation"]
[WhiteElo "3542"]
[pgn]
{WhiteEngineOptions: Protocol=uci; MoveOverheadMs=1000; WeightsFile=J92-190; Threads=2; Backend=demux; BackendOptions=backend=cudnn-fp16,custom_winograd=true,res_block_fusing=true,(gpu=0),(gpu=1),(gpu=2),(gpu=3); NNCacheSize=20000000; MinibatchSize=320; MaxPrefetch=160; MaxCollisionEvents=917; MaxOutOfOrderEvalsFactor=2.4; CPuct=2.147; CPuctAtRoot=2.147; CPuctBase=18368; FpuValue=0.443; CPuctFactor=2.815; PolicyTemperature=1.607; SmartPruningMinimumBatches=600; MovesLeftMaxEffect=0.2; MovesLeftThreshold=0.0; MovesLeftSlope=0.004; MovesLeftQuadraticFactor=0.0; MovesLeftScaledFactor=1.0; MovesLeftConstantFactor=0.0; VerboseMoveStats=true; SyzygyPath=/home/syzygy; LogFile=lc0.log; Ponder=false; UCI_ShowWDL=true; UCI_ShowMovesLeft=true; StrictTiming=true; CommandLineOptions=--show-hidden;, BlackEngineOptions: Protocol=uci; Threads=172; Hash=131072; SyzygyPath=/home/syzygy/; Use NNUE=true; Contempt=24; Move Overhead=1000; UCI_ShowWDL=true; OwnBook=false; Ponder=false;}
1. e4 {book, mb=+0+0+0+0+0,} e6 {book, mb=+0+0+0+0+0,}
2. d4 {book, mb=+0+0+0+0+0,} d5 {book, mb=+0+0+0+0+0,}
3. Nc3 {book, mb=+0+0+0+0+0,} Nf6 {book, mb=+0+0+0+0+0,}
4. e5 {book, mb=+0+0+0+0+0,} Ng8 {book, mb=+0+0+0+0+0,}
5. a3 {d=14, sd=41, mt=73356, tl=7136644, s=84575, n=6202827, pv=a3 Nd7 Nce2 c5 c3 Ne7 Nf4 c4 h4 h6 Qg4 Qa5 Be2 Nb8 Bd1 Bd7 Nh5 Nf5 Bc2 Nc6 Ne2 O-O-O Kf1 Qb6 Rb1 Kb8 Qf3 Ka8, tb=0, h=0.0, ph=0.0, wv=0.43, R50=50, Rd=-11, Rr=-11, mb=+0+0+0+0+0,}
b6 {d=48, sd=56, mt=353388, tl=6856612, s=88559721, n=31295854446, pv=b6 Nce2 Ba6 c3 c5 Nf3 Qd7 h4 Nc6 Rb1 h5 g3 g6 Bh3 Bh6 O-O Bxc1 Rxc1 Nh6 Re1 Bxe2 Qxe2 c4 Ng5 Ne7 Rb1 Nef5 Qd2 b5 Kg2 Qe7 f3 Rb8 Bxf5 Nxf5 Rh1 Nh6 Qf4 a5 g4 Kd7 Rhf1 Kc6 Kg3 Kb7 b3 Ka6 bxc4 bxc4 Rxb8 Rxb8 Nxe6 Qxe6 Qxh6 Rb3, tb=316, h=94.3, ph=0.0, wv=0.87, R50=50, Rd=-11, Rr=-11, mb=+0+0+0+0+0,}
6. h4 {d=15, sd=59, pd=Nd7, mt=134040, tl=7012604, s=83749, n=12342874, pv=h4 c5 Nf3 h6 Be3 Nc6 h5 Nge7 Bd3 c4 Be2 Nf5 Bf4 Be7 Qd2 Bd7 Bd1 Nh4 Nxh4 Bxh4 Ne2 Bg5 c3 f6 Bxg5 hxg5 f4 g4 f5 exf5 Ba4 Nxe5 Bxd7+ Nxd7 Ng3 Nf8 Nxf5 g6 Qe2+ Kd7 Ne3 Qe7 Qxg4+ f5 Qf3 Re8 Kf2 Qe4, tb=0, h=0.0, ph=0.0, wv=0.44, R50=50, Rd=-11, Rr=-11, mb=+0+0+0+0+0,}
c5 {d=42, sd=65, pd=Nce2, mt=82962, tl=6783650, s=92912244, n=7708092694, pv=c5 Nf3 Nc6 Ne2 Ba6 c3 Qd7 Rb1 h5 g3 g6 Bh3 Bh6 O-O Bxc1 Rxc1 Nh6 Re1 Bxe2 Qxe2 c4 Ng5 Ne7 Qd2 Nef5 Rb1 b5 Kg2 Qe7 f3 Rb8 Bxf5 Nxf5 Rh1 Nh6 Qf4 Kd7 g4 a5 Rhe1 Kc6 Kg3 a4 Re2 Kb7 Kg2 Ka7 Kg1 Rb7 Rg2 Ng8 Qg3 Nh6 Rf1 Qc7 Ra1 Qd7, tb=3, h=36.2, ph=0.0, wv=0.86, R50=50, Rd=-11, Rr=-11, mb=+0+0+0+0+0,}
7. Nf3 {d=16, sd=53, pd=c5, mt=74470, tl=6948134, s=79386, n=14488678, pv=Nf3 h6 h5 Ne7 Ne2 Qd7 c3 a5 b3 Ba6 Rh3 a4 b4 Bc4 Nd2 Nbc6 Nxc4 dxc4 Nf4 cxd4 Bxc4 Nxe5 Be2 Rc8 Qxd4 Qxd4 cxd4 Nc4 Ra2 Nd6 Bb2 Nd5 Nxd5 exd5 Re3+ Be7 Re5 Nc4 Rxd5 Nxb2 Bb5+ Kf8, tb=0, h=0.0, ph=50.0, wv=0.45, R50=49, Rd=-11, Rr=-11, mb=+0+0+0+0+0,}
h6 {d=50, sd=76, pd=Nf3, mt=881032, tl=5912618, s=85729264, n=75530224925, pv=h6, tb=11891, h=99.7, ph=50.0, wv=0.92, R50=50, Rd=-11, Rr=-11, mb=+0+0+0+0+0,}
8. h5 {d=16, sd=54, pd=h6, mt=143645, tl=6814489, s=79824, n=17523138, pv=h5 Ne7 Ne2 Qd7 c3 a5 b3 Na6 Rh3 Nc7 Nf4 Ba6 Bxa6 Nxa6 Bd2 Nc7 a4 c4 Bc1 cxb3 Nd2 Nc6 Nxb3 Be7 Qg4 O-O Rg3 Ne8 Kf1 Bg5 Ne2 f6 f4, tb=0, h=0.0, ph=66.6, wv=0.50, R50=50, Rd=-11, Rr=-11, mb=+0+0+0+0+0,}

1-0
[/pgn]


If 9... Nec6 is played like in the game, the evals go bad, down to -1.22 and white probably can win it.

So the conclusion is that black has several different viable moves and this French line is not busted. However, it will be a tough game for black to hold.

Well some new moves seemed to arise in this contest....Whites Move 5. A3 ! Is a nice wrinkle to eliminate the winawer possible pinning move Bb4..Black should continue Ba6 and exchange Bishops quickly... ...Blacks Move 5 c5!? is ineffective and too slow as white easily strengthens his D4 square with NF3... Nc3-e2... and c3. A critical problem with b6 (French Defense) is it somewhat weakens the queenside white squares and also b6 blocks Blacks queen critical access to the Qb6 and Qa5 dark squares..meanwhile White piles up attacking possibilities on the Kingside...Black must act quickly to Keep the balance in such closed positions...

[mmt]

I have a bunch of opening books. Hiarcs book recommends 5. Nf3. A line from that book, also found in the Chessbase French Powerbook 2018, goes:

1. e4 e6 2. d4 d5 3. Nc3 Nf6 4. e5 Ng8 5. Nf3 Ne7 6. Bd3 Nbc6 7. O-O Nb4 8. Be2
Nbc6 9. Qd3

with very good stats for White.

The latest SF dev would respond with 5... c5:
[d]rnbqkbnr/ppp2ppp/4p3/3pP3/3P4/2N2N2/PPP2PPP/R1BQKB1R b KQkq - 2 5
67/91 4,212,234,424k -0.90 1. ... c5 2.Be3 Nd7 3.dxc5 Nxc5 4.Nd4 Ne7 5.f4 a6 6.Qd2 Nc6 7.Qf2 Nxd4 8.Bxd4 Qc7 9.a3 Bd7 10.0-0-0 b5 11.h4 Rc8 12.g4 Qb7 13.b4 Ne4 14.Nxe4 dxe4 15.Rh3 Bc6 16.Be2 Be7 17.f5 Bd5 18.Qf4 Kd7 19.Rd2 Rhd8 20.Re3 Kc7 21.f6 gxf6 22.exf6+ Bd6 23.Be5 Qb6 24.Bxd6+ Rxd6 25.Rc3+ Kb8 26.Rxc8+ Kxc8 27.g5 Rd8 28.Kb2 Qc7 29.Qe3 Kb7 30.Rd4 Rg8 31.Bd1 Qe5 32.Bf3 h6 33.Bxe4 Bxe4 34.Qxe4+ Qxe4 35.Rxe4 hxg5 36.Rg4 Kc6 37.Kb3 Rh8 38.hxg5 Rh3+ 39.c3 Rf3 40.Rh4 Rg3 41.Rh7 Rxg5 42.Rxf7 Rf5 43.Ra7 Kb6

[alchemist888]

and 9... Qd7 instead of Nec6

[d]rnbqkb1r/p3npp1/1p2p2p/2ppP2P/3P4/P4N2/1PP1NPP1/R1BQKB1R b KQkq - 2 9
68/75 18:23:46 2,000,997,225k 30,214k -1.05 1. ... Qd7 2.Rh3 Bb7 3.c3 a5 4.Rh4...

Rh3-Rh4 maneuver looks very strange. At least unnatural.
I really like your deep analysis. Thank you so much!
But we know that Stockfish is able to quickly change it's evals, even after a long analysis. SF is able to agressive pruning. We have often seen that evals can change dramatically already on the next move on the board.
Also all evals in your analysis are pretty high. Some of them require careful analysis to draw a conclusion.