Quote:


A million thanks for this! (I was wondering how I’d missed it, till I noticed it’s only appeared recently.) Fascinating stuff, and only the last game of the four was familiar to me. Anyone know who Finegold’s opponents in the first two games were?

The plan of playing the Knight to f7 via h8 in the third game is indeed interesting. But if White plays Bd3 iso Be2, as in my posts above, I think things are a bit different. After 1. d4 Nf6 2. c4 c5 3. d5 e5 4. Nc3 d6 5. e4 Be7 6. h3 Nbd7 7. Nf3 Nf8 8. Bd3 Ng6 9. g3 O-O!? 10. h4!, I’m not sure 10 …Ne8, by analogy with Holt–Finegold from the lecture, is so good after 11 Nh2!?. Maybe 10 …h6!? is OK after (e.g.) 11 h5 Nh8 12 Nh4 Ne8 13 Nf5 Bf5 14 ef Bg5, but then why should this be better than (or why should it not be inferior to?) preparing to send the Knight on a different journey, with (e.g.) 9 …Bd7 10. h4 h6?


Quote:


At the moment I'm more tempted by the alternatives, but as yet I have no experience here ...

And how you play against the classical? Like Marin?

Quote:

Any comments on the lines I give -- do they help?

Hi Rue-64 and thanks for this, which sent me back to have another look at this main line. I noticed that in the game Brond–Agdamus, Villa Gesell 1972, Black played 12 …Kh8 iso 12 …a6; but then that Palliser, commenting on Albrecht–Neukirch, suggested that after 12 …a6 13 Nd2 Kh8 14 0-0-0(!), Black should probably defer …Ng8 in favour of …Bd7 and …b5. Naturally this set me wondering, if ...Ng8 isn’t best, why not omit …Kh8 and just get on with it on the queenside straightaway? Or am I being naïve? Of course, after 12 …a6 13 Nd2 Bd7!? White doesn’t have to play 14 0-0-0, but Black’s 13th is surely going to happen anyway ...

Meanwhile in the 7 …Nf8 8 h3 Bd7 9 g3 Ng6 line (‘V’ above, which Black can ‘force’ via 8 …Ng6 9 g3 Bd7), I’ve been wondering about 10 Kf1 a6!?, with the possible continuations 11 Kg2 b5, 11 Qe2 h5!?, or 11 a4 (best?) b6 12 Kg2 Qc8. Not massively dynamic this stuff, but White could over-press? …

I decided to start a new thread on this variation (or rather, set of variations) alone since I’m convinced it’s the biggest threat to the Czech Benoni and I’m hoping a few people might be interested!

There are, of course, quite a few move orders. After 1 d4 Nf6 2 c4 c5 3 d5 e5 4 Nc3 d6 5 e4 Be7, White plays Bd3, Nf3 and h3, but can do so in any order. Black plays 6 …Nbd7! and then has two main systems of defence: 7 …0-0, and the newer 7 …Nf8, followed usually by …Ng6 discouraging g2–g4. I’ll focus here mainly on this latter line, but personally I’m interested in both lines.

White’s strongest counter to the …Nf8/…Ng6 idea is thought to be h2–h3 and g2–g3, typically followed by h3–h4. My belief is that against Black’s ‘normal’ idea of …h7–h5 this is so advantageous for White that Black must look for alternative plans. For example, Donner–Keene went 5 …Nbd7 6 Nf3 Be7 7 Bd3 Nf8 8 h3!? h5 9 g3! Ng6 10 h4! Bd7, and now after 11 a3 a6 the key move 12 Ng5! left Black in an unpleasant position. Similarly after 10 …a6, 11 Ng5! just looks bad for Black. A key point is that the optically impressive 10 …Bg4 here leads to the undesirable exchange of Black’s ‘good’ Bishop after Be2! and Ng5 (or to Black conceding time and space if he retreats). 

So where to improve on Donner–Keene? Here I think it’s worth remembering a couple of factors that can work in Black’s favour in some related positions where he’s played …Ng6, with or without …h7–h5: (1) if White has played Qe2, Black’s …Bg4 may now make sense since if White then tries (e.g.) Qe3 and Be2 Black can meanwhile create counterplay; (2) if White has played Be3, he might find it harder to profitably play h3–h4 (since …Ng4 will hit the Bishop), so Black’s h-pawn advance may gain in force. This might suggest that Black should avoid …h7–h5 unless these conditions are met, and in fact Palliser, on p. 99 of How To Play Against 1 d4, suggests that, iso 8 …h5, 8 …Ng6 9 g3 h6!? could do with tests. This is by analogy with Lazarev–Kogan, which went 8 a3!? Ng6 9 g3 h6!? 10 h4 Bg4 11 Be2 Qc8! 12 Nh2 Bd7, and Black seemed OK. A key question here might be whether White should play h4–h5. If he does, he pushes Black’s Knight to f8 but aids Black’s dark-square strategy; if he doesn’t, Black can play …Bd8 and meet h4–h5 with …Ne7, countering in advance White’s plan to sink a Knight into f5.

But in the same spirit I wonder whether Black can start with 8 …Bd7, as played by none other than Petrosian! I’m thinking of variations like:

I  9 Qe2 Ng6 (Ghitescu–Petrosian actually went 9 …a6 10 Be3 Ng6 11 Nd2 h5 12 a3 h4) 10 g3 (10 a3 or 10 Bd2 maybe 10 …0-0) Qc8 11 h4 Bg4.

II  9 Bd2 Ng6 10 g3 a6 reaches, by transposition, Lazarev–Miladinovic, which began 8 …Ng6 9 g3 Bd7!? 10 Bd2 (10 h4!? h6!?) a6 and which Palliser (p. 99) suggests may be OK.

III  9 Be3 Ng6 10 g3 (10 a3 Qc8!?; 10 Qd2?! a6!?, Rychagov–Miladinovic) h6!? 11 a3 Qc8 12 Bf1!? Bd8 – we have reached by transposition a position that arose in the only game I can find featuring Palliser’s 8 …Ng6 9 g3 h6!?.

IV  9 a3 Ng6 10 g3 h6.

V  9 g3 Ng6 10 Kf1 (or 10 h4 Bg4 11 Be2 h6 12 Nh2 Bd7 13 Nf1 Qc8) Qc8 11 Kg2 Bd8 12 h4 (12 a4 Ba5) h6.

I hasten to add these are all just suggestions and I’d greatly welcome criticism of any silly nonsense/omission here!

In (V), 9 …a6 iso 9 …Ng6 might (?) allow 10 Nh4, but the same position without this possibility could be reached via another move order option, namely 7 …a6!? 8 h3 Nf8 (8 …Rb8!?) 9 g3 Bd7!?. This looks to me well worth exploring as well – one point being that on 8 a4 iso 8 h3 Black could, with 8 …0-0, transpose after 9 h3 to the line 6 Nf3 Nbd7 7 Bd3 0-0 8 h3 a6 9 a4, when 9 …Nh5!? (Lautier–Nisipeanu) gives interesting play. I think the biggest challenge to the 7 …0-0 plan comes with 9 Qe2! Nh5(!) 10 g3! g6 11 Bh6 Ng7 12 g4 Nf6 (Albrecht–Neukirch). Is Black really OK here? All he has gained over the line 9 Qe2! Ne8? 10 g4 g6 11 Bh6 Ng7 12 0-0-0 Nf6 (Popov–Hartston), which Palliser excoriates, is a tempo (White hasn’t castled) – is this really enough to change the assessment? The same question might be asked of the recent game Cramling–Pinho, where after 9 Qe2 Nh5 10 g3 Black tried 10 …Rb8!? and won. Anyone know of any analysis on this?

Finally, Palliser suggests 8 …Nh5!? as an alternative to 8 …a6. This might work OK after 9 Ne2 g6 10 g4 Ng7 11 Ng3 Nf6 12 Bh6 Kh8 13 Qd2 a6 14 a4 (also reachable via 8 …a6 9 a4 Nh5 10 Ne2 g6 11 g4 Ng7 12 Bh6(?!) Nf6 13 Ng3 Kh8 14 Qd2), but after 9 g3 instead I don’t see how Black can avoid Albrecht–Neukirch, e.g. 9 …g6 (9 …a6 10 Qe2 transposes directly) 10 Bh6 Ng7 11 g4 Nf6 12 Qe2, and does Black really have anything better than 12 …a6?