ForumHelpSearchMy ProfileSite InfoGuests InfoRepertoireLinks
The good news is that this month, we have a couple of live tournament with slow (or at least moderately slow) time controls to look at: Tata Steel (Wijk aan Zee). I’ve chosen a few French Defense games from this and the chess festival in Salamanca. Last month I referred a bit to Pentala Harikrishna’s recent book giving a detailed White repertoire with 3 Nc3, and this month I’m keeping an eye out for lines which are recommended in that and two other recent products: Dmitry Kryakvin’s two volume French repertoire book for Chess Stars (The Modern French Defence) and Wesley So’s repertoire with the Tarrasch (3 Nd2) for Chessable, which is given in the second part of his 1 e4 repertoire series. In this and future columns I’ll be using recent games which feature their recommendations, at least when doing so is useful and informative.

Download PGN of February ’21 French games

>> Previous Update >>

Tarrasch, Guimard Variation Mainline 4 Ngf3 Nf6 5 e5 Nfd7 6 Nb3 a5 7 a4 Be7 [C04]

Since Kryakvin recommends playing the Guimard Variation, 3 Nd2 Nc6, (note that he also has a short section on 3...c5), let me analyse a few recent games with it. The most popular line is 4 Ngf3 Nf6 5 e5 Nfd7. Drygalov, A - Santiago, Y, Titled Tuesday 12-15 2020, provides a good example of the old main line 6 Nb3 a5 7 a4 Be7 8 Bb5:

After 8...Na7 (8...Ncb8 is in the notes) 9 Bd3 b6, there are various ways for both sides to reorganize. In the notes, I’ve illustrated the state of play by using several previous examples. The game itself is interesting, but as a Blitz game it not surprisingly gets out of hand.

Tarrasch, Guimard Variation 4 c3 e5 [C04]

Gadimbayli, A - Andreikin, D, Titled Arena Jan 2 2021 is another rather silly Blitz game, but bears notice because of the strength of the Black player and as a reminder that White’s 4 c3 is a serious weapon. After 4...e5 5 exd5 Qxd5 6 Ngf3 exd4 7 Bc4, Andreikin played the odd 7...Qa5?!:

This could have turned out badly (White missed an early win of a piece, among other things), so I’ve given notes about Black’s better alternatives.

Steinitz Variation 5 f4 c5 6 Nf3 Nc6 7 Be3 a6 8 a3 [C11]

Giri, A - Firouzja, A, Wijk aan Zee 2021, was an extremely interesting game (at a normal time control!) that tested the variation 3 Nc3 Nf6 4 e5 Nfd7 5 f4 c5 6 Nf3 Nc6 7 Be3 a6 8 a3:

We’ve seen this increasingly often over the past two years, and Firouzja’s straightforward solution is critical.

Steinitz Variation 5 f4 c5 6 Nf3 Nc6 7 Be3 Be7 8 Qd2 0-0 9 Be2 [C11]

In the main line 7...Be7 8 Qd2 0-0 9 Be2 (9 dxc5 is also popular), the move 9...b6 has been dismissed in a couple of sources, but may still be a valid option. In Skripchenko, A - Paehtz, E, Salamanca 2021, there followed 10 h6 f6 11 dxc5:

and here Black sprung the novelty 11...fxe5. This had been rejected due to 12 Nxd5, but players had perhaps neglected to look further. In fact, it seems fully playable.

Winawer Main 7 Qg4 0-0 8 Bd3 Nbc6 9 Bg5 [C18]

Just when the main lines after 3 Nc3 Bb4 4 e5 c5 5 a3 Bxc3+ 6 bxc3 Ne7 7 Qg4 0-0 8 Bd3 Nbc6 9 Qh5 are looking respectable again for Black, the annoying 9 Bg5 is causing new problems. After 9...Qa5 10 Ne2 Ng6 11 0-0, the 3500 powerhouse computers LCZero and Stockfish played two games from this position, each playing one with White and one with Black. I don’t know what the rules of the match were, but both played 11...Qc7 as Black:

This may be a new move, at least I can’t find it in any of my databases (true, my correspondence databases haven’t been updated recently). The main idea is to be that after ...c4/Bxg5/...fxg6, the queen is well-placed to come to the defensive square f7. In LCZero 0.27.0d-Tilps-dje-mag - Stockfish 20210113, TCEC 20 Superfinal 2021, White played 12 Ng3 c4 13 Bxg6 fxg6 14 h4 Qf7 and a protracted struggle ensued, with both sides doing little for most of 170 moves, but White occasionally trying to make progress with a pawn break or exchange. Eventually drawn.

In Stockfish 20210113 - LCZero 0.27.0d-Tilps-dje-magi TCEC 20 Superfinal 2021, the play quickly took on a more dynamic tone after 12 a4 f5?!:

This led to a long forcing sequence, ending up with White having not only a theoretical advantage but the only real winning chances. The play is instructive, and it may be that Black can defend, but regardless of the objective assessment, 12...c4 is preferable for the practical player.

After all is said and done, 11...Qc7 may be playable for a computer engine, but if White plays well, Black has to defend indefinitely. You could have this as a reserve weapon, but the last time I analysed it (see the Archives), 11...c4 was more fun and looks objectively better as well.

Exchange Variation 4 Nf3 Nc6 5 Bd3 Bg4 6 c3 Bd6 [C01]

Tari, A - Caruana, F, Wijk aan Zee 2021, featured the line 3 exd5 exd5 4 Nf3 Nc6, Kryakvin’s recommendation (as well as various books over the past 20 years). Instead of the most critical 5 Bb5, Tari chose the solid 5 Bd3, leading to 5...Bg4 6 c3 Bd6 7 0-0 Nge7 8 Nbd2 Qd7 9 b4:

Here Black could take a chance with an aggressive move such as 9....0-0-0, 9...f6, or 9....Ng6, but chose the safe 9...0-0, after which Black has ideas of ...Bf5 or ...Ng6 with comfortable equality. The game became unbalanced, but didn’t stray far from equality.

King’s Indian Attack 3 Nd2 Nf6 4 Ngf3 b6 5 c3 [C00]

Topalov, V - Paehtz, E, Salamanca 2021, started out poorly with an improvised opening in which Topalov blundered and Paehtz missed her chance. The game opened 1 e4 e6 2 d3 d5 3 Nd2 Nf6 4 Ngf3 b6 5 c3 dxe4 (5...Be7 and 5...c5 are more popular) 6 dxe4 Bc5:

White played 7 Bb5+ Bd7 8 Bf1?? here, and Black played 8...Bc8, perhaps expecting a repetition. But 8...Ng4! would have been practically winning. After the smoke cleared, a hard battle with an interesting ending ensued, so I’ve included the game anyway.

Till next month, John

>> Previous Update >>

Please post you queries on the French Forum, or subscribers can write to me at if you have any questions or queries.