ForumHelpSearchMy ProfileSite InfoGuests InfoRepertoireLinks
I've put the mainstream Classical, Sämisch and Fianchetto variations to one side this month and instead delved into a number of offbeat lines which have been tried by GMs over the last few months and years. In Rapidplay and Blitz, reacting to something unusual can be testing, as there just isn't the time available that one would like. So certain ways of playing are better suited to these quicker time controls. Even if objectively a line isn't great, it can be a dangerous weapon if the opponent doesn't have a clear 'easy-to-implement' plan. Sometimes the side with the 'inferior' position (in the KID this is typically when Black has less space) is able to play coherent moves more quickly than the side with the engine-endorsed 'superior' game. It's a modern way of playing chess that has really come along in leaps and bounds in the online age.

Download PGN of September ’21 KID games

>> Previous Update >>

Modern Defence Averbakh System 4...e5 [A42]

The game Finegold, B - Xiong, J is a perfect example of how things can go wrong for White in the Averbakh variation.

Finegold's unexpected retreat to c1 is met by 9...Bh6 which ensures the exchange of White's stronger bishop, after which Black is well-placed to compete on the dark-squares. Jeffery Xiong then went on to throw his h-pawn forwards creating some further creaking in the white camp. So the moral has to be that these lines involving 5.dxe5 might offer a 'temporary initiative', but if the spark fails to ignite White can easily find himself with the inferior structure (the hole on d4 for example).

Fianchetto Variation 6...Ne4 [E60]

The position on move six in Salem, A - Firouzja, A merits a diagram:

Black seems to be 'breaking the rules' with 6...Ne4, as he moves his knight for a second time in the opening when there are other pieces that need 'improving'. There is no threat and the e4-square isn't secure, and yet this move nevertheless seems to be quite promising! It turns out that shifting Black's presence from the important e4-square isn't such a straightforward task. In the game, 7.Nc3 for example, invites 7...Nxc3 8.bxc3 when any free-flowing play that White might aspire to is hardly sufficient to make up for his broken pawns. For White, I recommend 7.Qc2 to induce 7...f5 after which the resulting Leningrad Dutch scenario is less comfortable for Black.

Irregular 5.e3 d6 6.b3 e5 7.Bb2 [E61]

A slow start doesn't prepare one for a quick finish in Nisipeanu, LD - Babula, V:

A queenside fianchetto in the KID is usually associated with various g3 systems, but it can also have bite when allied with the modest e3 and Be2. Flexibility is the order of the day as Nisipeanu delays castling long enough to give himself the choice of going either way. When it finally came, Nisipeanu went long (11.0-0-0) and combined d-file pressure with kingside play to generate strong threats. If I was pressed for advice about how to improve (in order to avoid such a fiasco!) in future, then 6...c5 is one way, but the game continuation still looks fine if 11...exd4 were to be chosen instead of the rather slow 11...a6.

Smyslov Variation 5...h6 6.Bd2 [E61]

Mamedyarov employed a pet line of his in the game Mamedyarov, S - Anton Guijarro, D. In the old days, the Smyslov was considered as a rather sedentary affair, but the following modern interpretation is anything but...

There are seven examples from the Azeri's personal experience, but no other GM has played this bishop retreat with any regularity.

With Black already having committed himself to ...h6, the idea is to put pressure on the weakened kingside (for example, with the 8.Qc1 and 9.h4 of the game). Anton Guijarro was able to stave off the first wave and entered the middlegame with an equal game, but went down later. Did all the time and effort in finding accurate defensive moves take its toll?

Smyslov Variation 5...d6 6.Be2 Nbd7 7.0-0 a6 [E61]

I quite liked the natural pragmatic way that White built up his game in Riazantsev, A - Grebnev, A. In particular how the knight on d2 (see 13.Nd2) seems to be ready to pounce if Black dares try anything active:

Grebnev may not seem to have done anything wrong and yet it doesn't seem to be evident to find a fully satisfactory way to proceed. White is ready for queenside expansion with b2-b4 and so Black tried 13...e5 after which there may be more than one way to a white edge. The 'rook and opposite bishop endgame' a pawn up was highly instructive.

Smyslov Variation 5...c5 6.d5 h6 7.Bh4 [E61]

In Dubov, D - Firouzja, A there occurred a 'Benoni response' to the Smyslov.

White seems to be ready for any attempt to steer the game towards a Modern Benoni with 8...e6, as then after 9.e4 Black won't be able to pin with ...Bg4 in the subsequent play. So the most common reaction is the forcing 8...g5 9.Bg3 Nh5 when Black seeks the elimination of White's bishop at the cost of some loosening of his kingside. This scenario is well-known here and in other analogous positions, so Firouzja (in an experimental mood!) chose something less theoretical that involves closing the centre with 8...e5. However, the reaction with 9.e3 preparing an eventual f2-f4 (it came on move thirteen) seems to give White the better chances.

Hungarian Variation 5.Nge2 a6 6.Ng3 [E70]

In Novikov, I - Fishbein, A Black played on both wings right from the start:

The diagram position is already unique as 7...b5 is a novelty. Fishbein's strategy involved gambiting the b-pawn (Novikov took it), hassling the g3-knight with his h-pawn (the steed actually moved five times in the first fourteen moves), and keeping things generally active. Castling had to wait (18...0-0!), but although all this might seem to be stretching credibility, it actually worked! Black had decent practical chances and thoroughly earned his half-point. Along the way, I reckon that it was White who had to be the most careful to avoid tripping up.

Classical Variation 6...c5 7.0-0 Bg4 [E91]

Combining ...c5 and ...Bg4 is generally considered acceptable for Black in the mainstream Classical Modern Benoni, but in Paravyan, D - Gabuzyan, H I examine how it works in a KID scenario.

If White allows ...exd5, for example with the typical 9.h3 Bxf3 10.Bxf3 exd5, then recapturing either way should yield a dynamic game for Black. The exchange of a pair of minor pieces eases any problems associated with a lack of space and Black can look forward to the middlegame with confidence. The engines favour White in the resulting positions (space + bishop pair), but in the real world it's hard to demonstrate anything clear over the board.

This brings us to the principled 9.dxe6 of the game which exposes the d6-pawn to the elements. After 9...Bxe6 10.Bf4 Qb6 White has a wide choice about how to resolve the tension concerning d6 and b2. The game move 11.Rb1 didn't work that well, so White should prefer a queen move and in particular 11.Qd2 when I don't think that Black can equalize. So Gabuzyan's opening system may have value as a surprise weapon (he did go on to win!) but wouldn't be such a good idea against someone well-prepared.

Classical Variation 6...c5 7.d5 e5 8.h3 [E91]

I've always been mystified by the type of closed position that occurred in Mishra, A - Aronian, L:

The diagram position after 9.g4 hasn't been seen very often, but the battle lines are clearly drawn up. White aims to build up and advance further, whereas Black hopes to engineer a counter with ...f5 or ...b5. The lack of space in Black's camp leads to the engines clearly preferring White. This makes me almost chuckle to myself, as I remember looking in books in my formative years where a model game (in analogous positions) would demonstrate how one can manage the advantage, but it never turned out that way in my own games! The elite are starting to play this way for Black (well at least in rapid and blitz) because they have also recognized that handling the White pieces is no easy matter. Turn your engines off and just play the position out with a friend and see what I mean! In the real world, the most creative player will often get their way.

Classical Variation 6...c5 7.d5 e5 8.0-0 [E91]

The folk who set the openings for these engine matches were perhaps intrigued as to see how things would turn out, so Ethereal - Lc0 is of interest in seeing how robust Black's position really is.

Although it's worth noting that Carlsen opted for 8...Ne8, not so long ago, 8...Nh5 looks like the most combative way because of the positional threat of ...Nf4. The engines give this as a clear advantage to White already, and yet Black scores well from here whatever response is chosen! In the featured game, 9.Re1 allowed ...Nf4 which came shortly and then, after some manoeuvring, when the time was right, Ethereal chose to capture on f4 and advance with e4-e5. In the resulting middlegame, the f4-pawn became weak and was ultimately sacrificed by Lc0 to gain time to set up his barricades. White then tried for almost 250 moves (yes, you've read it right!!) to win the pawn up endgame, but Black held firm.

Till next month, Glenn Flear

>> Previous Update >>

Don't hesitate to share your thoughts and suggestions with me. Any queries or comments to the KID Forum, or to me directly at (subscribers only) would be most welcome.