Training Matches - advice sought

[David Robertson]

Thursday, must be the 4th and final game of this training match. While not wishing to disturb the Prof's prep I am wondering what's been happening in the other games...I know someone who read all of Berg's 3 vols. and found other (than no 4.e5) pot holes and dead ends. But, I guess it's the same for almost every comprehensive work

Update: this week's games have been deferred to next week as my opponent nipped off to spend a few days walking in the Lakes. He picked the right week for it. So have I been using my windfall extra time vigorously prepping? Don't be silly! I've been out in the garden enjoying the sunshine. Berg's series is great. He struggled to finish Vol. 3. But no worries. Every French player should have it.

Game 1 started: 1 e4 e6; 2 d4 d5; 3 Nc3 Nf6; 4 e5 Nfd7; 5 Nf3. So, how did the next few moves go? Or is that a trade secret? Has much changed since this was written in Tim Harding's Classical French in 1991? "1.e4 e6 2.d4 d5 3.Nc3 Nf6 4.e5 Nfd7 5.Nf3 - W forgoes Pf4 for rapid dev & piece play - 5... c5 6.dc (Bb5 - Gurgenidze's move... now rarely seen) Nc6 most exact forcing the reply - 7.Bf4 Bc5 8.Bd3 main line... W sets up the classic B sac..."

No trade secrets. I'm happy to share if there's interest. The following gives some analysis and thoughts on Game 1. Most of my early comments are based on Moskalenko (source cited in an earlier post); later assessments are my own (subjective) and Houdini 4 (objective). Moskalenko provides a very helpful entry point to the Steinitz 5 Nf3 variation. He argues that the variation is dynamic in its modern treatment; under-researched; and likely to become fashionable. He could be right.

We have the opening sequence already:

1 e4 e6; 2 d4 d5; 3 Nc3 Nf6; 4 e5 Nfd7; 5 Nf3 c5; 6 dxc5 Nc6; 7 Bf4 Bxc5; 8 Bd3

Harding (1991) is right to warn of the B sac since the careless 8...OO walks into 9 Bxh7+ Kxh7; 10 Ng5+ Kg6; 11 Qd3+ f5; 12 Qg3 +/-

So 8...f6 is universally played, when 9 exf6 Nxf6; 10 Qe2 OO; 11 OOO is now effectively the modern dynamic mainline.

There are slower, quite playable, sidelines for both sides here. 9...Qxf6; 10 Bg3 OO; 11 OO is one; and 10 OO is still seen too. But the mainline here, following 10 Qe2 and 11 OOO, is the modern way to sharpen the struggle. The assymetrical character of the position has Sicilian-style features where both parties will attack on opposite flanks. To get some idea of how things can go, play through Smerdon - Lukey and Gunnarsson - Bisby. That's White doing well. But when White gets it wrong, Black has all the chances, as in Sowray - Pert from the 4NCL a while back.

Black has a number of choices now: 11...Bd7; 11...Qa5; but my opponent chose Moskalenko's preference 11...a6 when I replied with the energetic, and theoretically-approved 12 Ne5. I regard this as the signature move of the variation. White thematically occupies the key e5 square while preparing g4 and a Sicilian-style attack. But the position is really quite tricky - as I was about to discover!

Black replied 12...Bd6, and I'm embarrassed to admit that I hadn't prepped for this. I'd been transfixed by 'Sicilian' themes in a French, and had looked at 12...Bd7 and 12...Nxe5. I never expected my opponent to steer things back along devoutly French lines. Gormless me!

So now I'm playing by hand, not by book. And trying to gather my wits, I played 13 Kb1. This Sicilian-style prophyllaxis is perfectly sound anywhere but here. Moskalenko takes great delight in slapping a ? on it. Gloom. A slip like this next season in the 4NCL Div 1, and I'll likely have my backside handed to me on a plate. What's the refutation then? My opponent knew it - in part: 13...Qc7. And here, I sank into a long brood as the full nastiness of my position became apparent. Too awful to contemplate was 14 Nxc6 Bxf4; 15 Nd4 e5. Not greatly better was the move I settled on, 14 Rhe1 because, as the game shows, the easy liquidation on e5 leaves my f2 doomed. Oh, woe!

But there is worse, much worse! As Moskalenko points out, the move now is 14...g6! followed by 15...Nh5, and White is shot. Fortunately for me, my opponent had studied Moskalenko, but hadn't noticed his note on this little trick. So the game proceeded with the liquidation which, in truth, seemed good enough: 14...Nxe5; 15 Bxe5 Bxe5; 16 Qxe5 Qxe5; 17 Rxe5 Ng4; 18 Re2 Nxf2. We both thought, after the game in the light of White's next move, that 18...Rxf2 was better. It is if 19 Rxf2 Nxf2 follows. But White has 19 Na4, and Black has no more than a small edge (says Houdini).

Now White has a 'get-out-of-jail' 19 Bh7+ Kxh7; 20 Rf1 Bd7; 21 Rexf2 Rxf2; 22 Rxf2 Kg8 when I felt I'd survived one crisis, only to find myself in another. A minor piece ending, N v B, cannot long be avoided where, with play on both sides of the board, the B is almost always favoured. Time for some very accurate play then. 23 Kc1 Re8; 24 Kd2 Bc6; 25 Ne2 e5; 26 c3 Rf8; 27 Rxf8 Kxf8. And here we are: the moment of truth.

Can White save this? At the time, I concede that I was doubtful. But Engine is far more sanguine - to my surprise. 28 Ke3 Ke7; 29 b4 Kf6; 30 Nc1 d4. This was the move I'd seen way back, and feared. Black now creates play - and possible entry points - on both sides of the board. 31 cxd4 exd4; 32 Kxd4 Bxg2; 33 Nd3 Kf5; 34 Ke3 g5; 35 a4 Bc6; 36 Nc5. In playing this last move, I needed to see, and trust my assessment, of the tactics required by my highly compromising 39th move. I found this very hard-going at the board. 36...b6; 37 Nxa6 Bxa4; 38 Nc7 Bc6; 39 b5. Why "highly compromising"? Because parking my pawn on a light square risked making it a target for the B if I hadn't got my analysis right.

Fortunately I had. 39...Bb7; 40 Kd4 Kg4; 41 Ne6 Kh4; 42 Nxg5 0.5 - 0.5

[John McKenna]

I spent today in the gardens... Kew gardens. I saw a lot of trees, plants, etc. and discovered that 'Kew' comes from the Anglo-Norman word for a cook. As a consequence I am bushed so cannot say much about the game given by David, above. Except, of course, to thank him for putting the moves up and for explaining them.

I don't want to give away too many of Tim Harding's trade secrets, so all I'll say is that he gives a different main 8th move for Black in his 1991 Classical French and attributes it to Keres saying, "gives W an extra chance to go wrong." Any idea what the move might be?

[Roger de Coverly]

I
I don't want to give away too many of Tim Harding's trade secrets, so all I'll say is that he gives a different main 8th move for Black in his 1991 Classical French and attributes it to Keres saying, "gives W an extra chance to go wrong." Any idea what the move might be?

Database research indicates that 8. ..f6 is the most popular. The alternatives with at least 100 games are 8. .. a6, 8. .. Qb6, 8. .. h6 and 8. .. Be7. These are all moves you would play as Black if you didn't know the supposed theory, but did have experience in typical French positions.

Moving on, perhaps not knowing the theory can be an advantage. At move 15. Bg3 has value if you aren't sure of the analysis, but you see the tactical weakness.

[John McKenna]

I
I don't want to give away too many of Tim Harding's trade secrets, so all I'll say is that he gives a different main 8th move for Black in his 1991 Classical French and attributes it to Keres saying, "gives W an extra chance to go wrong." Any idea what the move might be?

Database research indicates that 8. ..f6 is the most popular. The alternatives with at least 100 games are 8. .. a6, 8. .. Qb6, 8. .. h6 and 8. .. Be7. These are all moves you would play as Black if you didn't know the supposed theory, but did have experience in typical French positions...

Roger, all the 8th move alternatives you gave are covered by Tim Harding.

8... f5!? is "Keres' move" (Harding) - I cannot find a Keres game for the origin of that attribution. It was played in Chekover,Vitaly-Zita,Frantisek 1946 Leningrad-Prague.

9.exf6 transposes to David's game but 9.O-O?, 9.Bxf5?! (annotations by Tim Harding) and 9.Qe2 have been tried. (Harding also analyses 9... Qxf6!? 10.Bg5/Bg3...)

9... Nxf6 10.Qe2 O-O 11.O-O-O (Alexander Zhakarov tried this thrice in 1997, so it may be his idea?) is the parting of the modern and ancient ways since Harding gives 10.O-O O-O 11.Qe2! and says that is "the main reason for taking 5.Nf3 seriously."

According to Harding (1991) the relevant games in this line were contested between the early 1960s and the late 1980s involving, amongst others, the following -

Rigo-Gulko 1988 Rome
Baker,CW-Wheeler 1987 Eng. (regional) Ch.
Johansen-Thompson 1985 Br. Ch.
Harding-Whipp 1984-5 corr.
Zaitseva,M-Aksharumova 1984 Moscow (women's) Ch.
Bellon-Speelman 1978 Amsterdam
Williams-Mestel 1975 Br. Ch. (play-off)
Ljubojevic-Petrosian 1972 Las Palmas
Larsen-Spassky 1969 Stockholm
Spassky-Petrosian 1966 World Ch. (19th game)

And, if (instead of 7... Bxc5) 7... Nxc5 - "This move may give B a playable position although it appears less consistent" (Harding) - Tim references, amongst others, the following -

Rogers-Zuger 1987 Mendrisio Op.
Rogers-Delay 1985 Martigny
Rogers-Garcia,G 1984 Biel
Trapl-Harding 1984 corr.
Binham-Karlsson 1981 Helsinki
Mariotti-Capello 1979 Rome
Bastrikov-Nezhmetdinov 1963 USSR
Kopilov-Doroshkevich 1963 USSR

After 11... a6 (Bd6 12.Bxd6 Qxd6... was Eggleston-Brown 2006 Liverpool, EU Ch.) 12.Ne5 Bd6 13 Kb1? David admits he was struggling. Instead 13.Bg3 seems the best try since, although it resulted in a loss for White in Zakharov,A-Kashtanov,R 1997, it led to a draw in Vocaturo,D-David,A 2008 Bratto Op.

I hope the training match will resume when time and the weather permit.

[Roger de Coverly]

8... f5!? is "Keres' move" (Harding) - I cannot find a Keres game for the origin of that attribution. It was played in Chekover,Vitaly-Zita,Frantisek 1946 Leningrad-Prague.

I believe it can only have been a Keres move in an annotation, as he didn't play the French after 1938. Unless it was a little known Correspondence game.

[David Robertson]

Further thoughts from Moskalenko. He names the variation: 10 Qe2 and 11 OOO, after Zakharov, referencing Zakharov-Potkin (1997) and its trademark idea, 12 Ne5. But Moskalenko seems to suggest that 12 h4 might be the better way to go.

Why? Well, in the line I played, 13 Bg3 obviously improves on my error, whereafter Moskalenko claims 13...Qc7 must be met by 14 Ng4 Nxg4; 15 Qxg4 Ne5!; 16 Qd4!? with equality (Pedraza-Moskalenko, 2011). Now it seems to me that, if Moskalenko is right, 13...Qc7 forces a hefty concession from White because 14 Ng4 isn't a move you really want to be playing.

So, what of the 'natural' alternative, 14 Rhel? Intriguingly, Moskalenko declares that Black will gain a sizeable advantage after precisely the same liquidation as in my game, ending in 18...Nxf2. Yet we know from my game that, after 19 Bxh7+, White remains in reasonable shape. Maybe I've contributed a pennyworth to theory after all. Alas, no. The seemingly minor difference in the placement of the WK (on c1 now, not b1 as in my game) means that after 19...Kxh7; 20 Rf1 falls to 20...Nd3+, and the R drops!