ForumHelpSearchMy ProfileSite InfoGuests InfoRepertoireLinks
This month I'll be looking at developments in those lines where White plays an early f2-f3 or g2-g3. It's important for Grünfeld players to be ready for these options, even if they fall outside the main Grünfeld domain. Some sources only give them a small chapter or so (perhaps because they can readily transpose to other openings) so these areas can easily get overlooked. They may have names like the Anti-Grünfeld or the Neo-Grünfeld but, whatever they are called, it's still necessary to know your stuff. So read on!

Download PGN of September ’21 Daring Defences games

>> Previous Update >>

Anti-Grünfeld 3.f3 Mainline, 7.Be3 0-0 8.Qd2 Nc6 9.0-0-0 e5 10.d5 Nd4 11.f4 [D70]

In Stockfish - Lc0 a largely abandoned line (involving 9...e5) was employed in their match.

Here Lc0 met 16.Rf1 with a novelty 16...Bh5 and was ultimately able to stave off White's initiative and save the day. However, we don't all have the same level of defensive technique and Black doesn't have an easy ride. I imagine that the whole variation will probably remain under a cloud, as there doesn't seem to be an evident path to equality.

Note that in their return encounter (with reverse colours) White varied with 16.Kb1 whereupon Black also tried 16...Bh5 and was again able to scramble a draw. A new concept, thanks to these high-powered engines, but perhaps not game changing for the popularity of this line.

Anti-Grünfeld 3.f3, 7.Be3 Nc6 8.d5 [D70]

In some of these Anti-Grünfeld lines Black has to careful not to get into hot water right in the opening, a scenario that occurred in Ding Liren - Svidler, P.

In my opinion, the easiest way to obtain a decent middlegame is by hitting at the centre with 12...c6!, as discussed in the notes.

Svidler opted for 12...e6, which I think is more risky. The game continued with 13.Qd4+ Nf6 14.0-0-0 exd5 15.e5 and now Black has to decide between 15...Ng8 and Svidler's choice of 15...Nh5. Neither of these are that comfortable, but both of them seem playable, but only with very precise play in each case.

Neo-Grünfeld 5.cxd5 Nxd5 6.e4 Nb6 7.Ne2 c5 8.d5 [D72]

In Sarana, A - Ponkratov, P although Black managed to get the better of the struggle, I have some doubts about his opening. I think that Black should delay ...exd5 to render the d5-d6 advance less dangerous.

In the game, the advance 12.d6 opened up nicely the long light-squared diagonal for the 'Neo-Grünfeld bishop'. However, after 12...Nb4 Sarana's 13.Nf4 led to no advantage following 13...Nc4. Instead, I consider 13.a3 to be more promising, even if this is pushing the knight back to a decent square on c6, my judgement being guided by the further moves 13...Nc6 14.Ne4 Bg4 14.f3 Bf5 15.Bg5 when I'd rather be White.

Neo-Grünfeld - 6.cxd5 Nxd5 7.0-0 Nc6 8.Nc3 e5 [D74]

The riposte 8...e5 of Malakhov, V - Vocaturo, D is totally in contrast with the traditional 'main line' (I mean 8...Nb6) where manoeuvring is usually the order of the day. Here play is more forcing, as well as being less well-mapped than those more established lines. It also seems fairly sound:

This is very much a modern move with the vast majority of games coming from the last few years. The adventurous brigade including Jobava and Dubov employed it back in 2017, Carlsen in 2018, and since then quite a number of grandmasters have followed. The game continuation of 9.Nxd5 e4 10.Ne5 Qxd5 11.Nxc6 Qxc6 12.Bf4 can be considered as the 'main line of 8...e5'. At this point several moves have been tried, but Vocaturo opted for the rare 12...Qb6 which certainly needs to be taken seriously, as he achieved a good game. Still, I don't know if one can prove full equality against best play, but Black certainly gets close. This also holds true for some of the other thirteenth move alternatives.

So 8...e5 is worth a try in my opinion.

Neo-Grünfeld - 6.0-0 dxc4 7.Na3 c5 8.dxc5 c3 9.Nb5 [D77]

One of those 'symmetrical, but...' variations occurred in Banusz, T - Sadhwani, R.

It seems that many elite games of late have reached the diagram position. White is nominally better but most of the time Black is able to gradually steer the game towards drawish equality. The best way to achieve this is to capture a couple of times on d4 and then drop the knight back to d6, which defends the b7-pawn thus enabling the undeveloped bishop to come into the game. Here Sadhwani tried 13...Bf5 which seems to keep more life in the struggle. He was probably doing fine until he lashed out with 15...e5?! (instead I suggest 15...Bxd4! which should lead to a balanced (but not drawn!) game) which weakened himself enough to enable Banusz to pounce on his broken kingside and eventually win a well-played rook and opposite bishop endgame.

Neo-Grünfeld - 6.0-0 dxc4 7.Na3 Na6 8.Nxc4 c5 [D77]

Romain Edouard was able to obtain equality in Van Wely, L - Edouard, R with 7...Na6 followed by ...c5, which is still something of a sideline, but it scores about the same percentage as the main lines with 7...c3 or 7...Nc6.

Here Van Wely tried 12.Re1, which is an unusual choice, as White generally places this rook on d1 in this type of position (it went there on move 16 anyway!). As there are no real threats, it probably isn't anything to write home about. Several alternatives have been tried, but I think that Black is on the verge of equality in all lines. The Dutchman found a way to complicate matters in the middlegame, but his pawn sacrifice was only good enough for a draw.

Neo-Grünfeld - 6.0-0 c6 7.Qb3 dxc4 8.Qxc4 [D78]

In Svane, R - Tomashevsky, E Black's endgame technique proved to be superior and that made all the difference. However, earlier on it was White who was in the driving seat.

This rather positional opening is something of a favourite for Svane who has quite some experience in this variation. However, Tomashevsky has also played the diagram position with White, so he was also at home in these positions.

The choice of 11...Qb6 looks natural, but I prefer the more active 11...Qa5, one key point being that the b7-pawn is taboo. Objectively, both of these queen moves look quite reasonable, but please note that after the further moves, 12.Nh4 Nxc3 13.bxc3 Be6 Svane's innovation with 14.Qa3 gave him the slightly better game. Stockfish also suggests this novelty, so I suspect that this move came about after some home preparation, especially as he had previously employed 14.Qc2 a couple of times.

In Nakamura, H - Dubov, D Black was able to demonstrate a clever way to solve any problems associated with his queenside weaknesses.

Here the surprising (well, if you haven't seen this concept before!) 10...c5! looks strange as the long diagonal is opened for White's fianchettoed bishop. However, there doesn't seem to be a clear way to exploit this fact and in the actual game Dubov equalized almost immediately. This novelty is quite an important one, as the alternatives 10...Nbd7 and 10...a5 don't look comfortable.

The idea is known from an analogous position (where Black's bishop is on f5 rather than e6), and has in fact been played twice there by Dubov, again with success.

Neo-Grünfeld - 6. 0-0 c6 7.Qb3 Qb6 [D78]

In Nakamura, H - Aronian, L Black wriggled out of his earlier difficulties only to slip up by allowing Nakamura's pieces to become too active.

In this sedate-looking position, the game soon livened up because Black was in a provocative mood. The most solid line involves 10...Na6 when the onus (that is, theoretically) is on White to try and demonstrate any advantage. Efforts so far have been unconvincing.

Instead Aronian chose 10...dxc4 11.Qxc4 Nd5 but after 12.Ng5! Nakamura invited complications which do seem to be favourable for White (especially with the precise 16.Nxg6+ rather than 16.axb3).

Neo-Grünfeld - 6. 0-0 c6 7.Nbd2 Bf5 [D78]

In Onischuk, A - Paravyan, D Black's approach in the opening was to give priority to 'quick development' whereas White decided to expand on the queenside with his pawns. This clash of ideas led to the following position:

Here White is slightly behind in development, so needs to step with care. To this end 15.Re1 followed by 16.Qc2 would keep control and so maintain the slight space advantage. Instead, Onischuk erred with 15.h3?! after which 15...Ne4! was quite dynamic, and so much so that the first player should probably have sought equality by capturing twice on e4. A further error and White was already in trouble. I think that Black's development plan makes a refreshing change from the typical one of pushing the a-pawn, but of course 10...a5 (instead of 10...Rc8) has been played more often, both here and in several similar situations.

Till next month, Glenn Flear

>> Previous Update >>

If you have any questions, either leave a message on the Daring Defences Forum, or subscribers can email me at