ForumHelpSearchMy ProfileSite InfoGuests InfoRepertoireLinks
This month I’ll emphasize two variations which are quite popular but to which I haven’t paid enough attention. The first is in the Tarrasch Variation with 3 Nd2 c5 4 exd5 Qxd5, where White plays 5 dxc5. The other is the Armenian Variation of the Winawer, 3 Nc3 Bb4 4 e5 c5 5 a3 Ba5. We also see Magnus Carlsen on the Black side of a Poisoned Pawn(!), and some typical games with 3 Nc3 Bb4 4 exd5 exd5 5 Bd3.

Download PGN of January ’18 French games

>> Previous Update >>

Tarrasch Variation with 3...c5 4 exd5 Qxd5 5 dxc5 Qxc5 [C07]

After 3...c5 4 exd5 Qxd5, the move 5 dxc5 has attracted a lot of grandmaster interest over the past few years, at least in part because the main lines with 5 Ngf3 cxd4 6 Bc4 are holding up so well for Black. White hopes that his active piece play will lead to some advantage. The recent game Jones, G - Adhiban, B, Wijk aan Zee 2018, can serve as a starting point. After 5 dxc5, Black chose the recapture 5...Qxc5. In notes to a game from a few years back, Jones said that this is superior to 5...Bxc5, but see below. In this game, both sides developed naturally and reached this standard setup:

Such positions have led to equal play in numerous contests, and that was true here until Black became too ambitious and left his queenside undefended.

A miniature win by Robson over a much lower-rated opponent in Robson, R - Arjun, U, Colombus 2017, demonstrates one feature of the 5 dxc5 Qxc5 line: that Black does burn up time by playing moves such as ...Qxc5-c7, ...Nbd7, and ...a6, so he has to be ultra-careful to finish his development as quickly as possible.

Here 10...0-0 would be fine, but 10...a6?! 11 a5 Ng4? was way too slow and the game ended in short order.

Tarrasch Variation with 3...c5 4 exd5 Qxd5 5 dxc5 Bxc5 [C07]

The recapture 5...Bxc5 has been criticized, in particular by Renier Castellanos’ in his repertoire for White. But that assessment hasn’t really been borne out by practice, and I’m not convinced that White retains any real advantage. In Naiditsch, A - Le Quang Liem, Riyadh 2017, the main line 6 Ngf3 Nf6 7 Bc4 Qc6 8 Qe2 0-0 9 Nb3 arose:

Here Black played 9...Nbd7 10 Bf4 Bd6, which should have equalized.

In Shahaliyev, I - Ahmadzada, A, Baku 2018, instead of 9...Be7 or 9...Nbd7, Black went into 9...b6!? 10 Nxc5 Qxc5:

Normally White, who has the bishop pair and no weaknesses would be assumed to stand at least somewhat better in such a position. But in these 4:3 versus 3:2 majority situations, the knights can prove effective for the side with the extra center pawn. In the game that was the case, and equality followed. From what I’ve seen, this compensation for the bishops applies to most positions in this system.

Winawer Poisoned Pawn 11 f4 dxc3 12 Nxc3 [C18]

Magnus Carlsen has played the French from time to time, but as far as I know, he’s never used the Winawer Poisoned Pawn Variation! In Giri, A - Carlsen, M, Wijk aan Zee 2018, there 3 Nc3 Bb4 4 e5 c5 5 a3 Bxc3 6 bxc3 Ne7 7 Qg4 cxd4 8 Qxg7 Rg8 9 Qxh7 Qc7 10 Ne2 Nbc6 11 f4 dxc3, and here Giri chose the messy line beginning with 12 Nxc3:

This was more popular a few years ago but is still theoretically unresolved and interesting. Quite possibly Carlsen had covered this position in his match preparation for Karjakin. At any rate, the players went down a main line and although Giri deviated a bit, neither side had anything and an early liquidation ensued. In general, I feel there’s much more to be learned about this 12 Nxc3 variation.

Winawer Armenian Variation 5 a3 Ba5 6 b4 cxd4 7 Qg4 [C18]

We haven’t looked at 3 Nc3 Bb4 4 e5 c5 5 a3 Ba5, the Armenian Variation, for some time. Black has had some recent success with positional move orders, and we are hopefully seeing a new era for interesting games in this line. First, by way of background, let me show a game from a few months back which bolsters White’s case in the traditional main line with 6 b4 cxd4 7 Qg4 Ne7 8 bxa5 dxc3 9 Qxg7 Rg8 10 Qxh7 Nbc6 11 Nf3:

Here Bodnaruk, A - Sichinava, Z, St Petersburg 2017, continued 11...Qxa5 12 Ng5! , after which Black was already close to losing. Since 11...Qc7 12 Bf4 Bd7 13 a6 is still considered promising for White, we can see why the entire ...Ne7 gambit idea is looking shaky.

The revival of 5...Ba5 therefore seems to depend upon 7...Kf8 (instead of 7...Ne7) in the variation above:

I’ll introduce this line with the game Demianjuk, A - Mirzoev, A, Baku 2017, which also has a note on the important alternative move order 6 Qg4. White plays 8 Nb5 from the diagram and faces the modern recipe 8...Bb6. This is arguably the most important position for 5..Ba5 right now, so I’ll show three recent games with the position after 9 Nf3 Nc6:

Mihok, O - Kallai, G, Hungary Teams 2017, saw 10 Qf4!?, and after 10...Nge7 White got a small edge and won a nice game. But the timely 10...f6! seems to cast this into doubt.

So the main line is 10 Bb2 Nge7, after which the Demianjuk-Mirzoev game continued 11 Nbxd4, and although White had a small edge, Black could have equalized. In the elite-level game Inarkiev, E - Vallejo Pons, F, Palma de Mallorca 2017, White preferred 11 Bd3 Ng6 12 Qg3:

This would gain a positional advantage except for that other thematic French move, 12...f6! , which fully equalized. An exciting struggle ensued, ending in a well-played draw.

Turning to White’s 8th move, it’s also possible to clear out the board a bit after 7...Kf8 with 8 bxa5 dxc3 9 Nf3 (equivalent to 7 bxa5 dxc3 8 Qg4 Kf8 9 Nf3):

This seems to be fairly easy for Black to play, as seen in Papp, P - Bivol, A, Riyadh (World Blitz) 2017 (9 a4 is the main option, discussed in the notes).

Finally, there’s the positional line 6 b4 cxd4 7 Nb5 Bc7, which I hope to get started on next month!

Winawer Exchange 4 exd5 exd5 5 Bd3 Nc6 6 a3 Ba5!? [C01]

Over the past decade the variation 3 Nc3 Bb4 4 exd5 exd5 5 Bd3 has grown in popularity. White doesn’t get any advantage in theory, but it’s a safe approach. In two games this month, Black played 5...Nc6 6 a3 and now 6...Ba5!? (6...Bxc3+ has been played far more often).

We’ve seen this only a couple of times in this column, but it has a respectable record. The following position is typical:

In Alekseev, E - Akobian, V, Riadh 2017, White came up with 9 Ng3!? Nxd4 10 Bxh7+, giving up the bishop pair in order to develop and expose Black’s king. But Black’s pawn structure was too solid and White even got the worst of things after some inaccurate moves.

White played more solidly with 9 Bg5 f6 10 Be3 in Yifan, H - Mamedyarov, S, Wijk aan Zee 2018. With 10...Bf5, Black challenged the light squares and equalized, later slowly outplaying his opponent in a complex strategic middlegame.

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.