ForumHelpSearchMy ProfileSite InfoGuests InfoRepertoireLinks
I’m happy to say that this month’s batch includes a few important variations that I’ve let slip under the radar in the past few years. First we have some fun and original ideas in the Tarrasch with 3...h6, and then some Winawer variations that might prove very useful in consolidating or simplifying ones repertoire.

Download PGN of June ’20 French games

>> Previous Update >>

Tarrasch Variation 3...h6 4 Ngf3 Nf6 5 e5 Nfd7 6 Bd3 c5 7 c3 Nc6 [C03]

My main motivation for looking at the Tarrasch Variation with 3....h6 comes from a brilliant blog article by US Master Andrew Liu on, in which he makes a strong attempt at refuting the main line upon which the variation rests. This arises after 4 Ngf3 Nf6 5 e5 Nfd7 6 Bd3 c5 7 c3 Nc3 8 0-0 g5 9 h3 h5, and here Liu recommends 10 Re1!:

That has always been considered weak due to 10...g4, but Liu noticed that instead of 11 hxg4, White can play 11 Nh2!, and it turns out that after 11...gxh3 12 Ndf3!, Black faces serious problems, because White has full control of the center and great attacking chances against Black's weakened kingside. For some background on this variation, with his and my analysis and details about this main line, see Tarrasch 3...h6 Liu Line Part 1 - 11...gxh3.

It would be remarkable if, after all these years of strong players using 3...h6 (including Eingorn, Shulman, Volkov, Short Carlsen, Nakamura, Morozevich, Vallejo Pons, and many others), the move was no longer playable. In fact, things are not so clear because Black can try 11...cxd4 12 cxd4 Nxd4. Then after 13 hxg4,

, both 13...hxg4 and 13...Bc5 keep Black alive. The game Tarrasch 3...h6 Liu Line Part 2 - 11...cxd4 has my suggested improvements over the analysis Liu offers, as well as some analysis by Robert Shylakhtenko, who first drew my attention to Liu’s work. The overall verdict is not a happy one for Black, who ends up with a slightly worse position in a couple of simplified positions; this means that he has very few winning chances. On the other hand, White can’t avoid those positions; and since they will very likely be drawn with competent play (especially if both sides come prepared), he might not be that happy either.

I also wanted to look at 3...h6 because of two games played this month by Grandmaster Gawain Jones, using a system which has netted him a perfect score. That goes 3 Nd2 h6 4 Ngf3 Nf6 5 e5 Nfd7 6 Bd3 c5 7 c3 Nc6, and now 8 Bc2:

The immediate point is to be able to play Nb3 without allowing the fork ...c4. By so doing, White prevents some combination of ...g5-g4, ...cxd4, and ...Qb6, winning the d-pawn, thus securing d4. This leaves Black the task of justifying his ...h6 move. In Jones, G - Chow, S, English Online Blitz Open 2019, 8...Qb6 9 Nb3 cxd4 10 cxd4 a5 11 a4 followed, and Black found the interesting 11...Qb4+, which should have resulted in decent counterchances (11...Nb4 isn’t bad either). The notes to this game contain two other wins by Jones, as well as Black options on move 8.

Jones, G - Saravana, Kr, Titled Tuesday 2020 went 8...Qb6 9 Nb3 a5 10 a4:

But then instead of 10...cxd4 (transposing), Black played 10...g5?!, which simply isn’t as effective after White has played Bc2 and Nb4. Now 11 h3 yielded some advantage, but 11 0-0! would have been particularly strong.

Winawer Variation 7 Qg4 Qc7 8 Bd3 c4 [C18]

I’ve neglected the closed variation of this line, in which Black plays ...c4 and ...Nf5. Theory has it that White gains a small advantage, but there are some subtle move order issues that may affect that assessment. In the main line 3 Nc3 Bb4 4 e5 c5 5 a3 bxc3+ 6 bxc3 Ne7 7 Qg4 Qc7 8 Bd3, Djukic, Ni - Ljukin, S, Titled Tuesday 2020, reached a standard position after 8...c4 9 Be2 Nf5:

Here there are some subtle modern setups beginning with 10 Bd1 and 10 Nh3. In the game, White played the traditional 10 Nf3, when Black played the flexible 10...Bd7 (I discuss two main alternatives), and after 11 Nh4, the nature of Blitz games interfered with both players ability to see that 11...Nxh4?? Should have lost outright to 12 Qxg7. Black has three instructive alternatives on move 11.

Winawer Variation 7 Qg4 Qc7 8 Bd3 Qa5 9 Bd2 c4 [C18]

In Karjakin, S - Livaic, L, PNWCC Online Blitz 2020, Black played the superficially time-wasting 8...Qa5 9 Bd2 c4 10 Be2 Nf5:

But as you will also see in the notes to Djukic - Ljukin, one of Black’s biggest mistakes in this line is to wait to force White’s bishop to d2, because often allowing White to play a4 while his bishop is still on c1 and able to come to a3 is quite bad for Black. Since there are few cases in which the insertion of ...Qa5/Bd2 has any drawbacks, it’s arguably a good idea to do it now (before, for example, White can play Bd1 and Ne2). The game itself shows one of the world’s strongest Blitz players absolutely collapsing in time pressure.

Winawer 7 Qg4 cxd4 8 Bd3 Qa5 [C18]

3 Nc3 Bb4 4 e5 c5 5 a3 Bxc3+ 6 bxc3 Ne7 7 Qg4 cxd4 8 Bd3 Qa5 is a potentially very important line, since it bypasses some complex options that arise after 7...Qc7. Traditionally, the impression has been that 8...Qa5 is favorable for White, in which case Black can’t favorably avoid the transposition back to the main lines with 8...Qc7. But the past few years have seen some very promising developments for Black in the main line 9 Ne2 0-0 10 Bg5 Ng6 11 Qg3:

I’ve used Clarke, B - Chan, K, Melbourne 2018, as a base game to introduce and analyse what has been considered a main line after 11...Nd7 12 h4 dxc3 13 f4 h5 14 Qe3 Qb6 15 Nd4. Chan played the older 15...Nb8 16 Kf2 Nc6 and we see a thorough overview of how this line has changed (mainly in Black’s favor) since Negi first recommended it for White in his repertoire book.

If that weren’t enough, Black has recently established the playability of the option 15...Nc5:

In Lagno, K - Ganguly, S, Hunan 2019, we look at some of the most critical games and ideas. The sad thing is that at several junctures the game becomes a matter of whose engine analysis and/or memory is better, because the result can be worked out to the end.

Overall, Black seems to be doing well after 9 Ne2. A supergrandmaster blitz game this month raises another important question. In Giri, A - Nepomniachtchi, I, Chessbrah Inv 2020, White played the rarely-seen 9 Rb1:

There isn’t much theoretical material to go on, and the game itself is typical Blitz back-and-forth madness; but this looks like an extremely rich area for investigation. The note on Black’s alternatives to 9...Nd7 is instructive for understanding the main themes.

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.