Sunday, 8 July 2018

038. The Chigorin Sicilian

Black: RickF - All-play-all tournament,, 2018

Actually, Mikhail Chigorin has no Sicilian system named after him. Against 1 e4 the great Russian master played 1...e5 almost exclusively, and in the king's pawn openings his name is mainly associated with a formation in the Closed Ruy Lopez: 1 e4 e5 2 Nf3 Nc6 3 Bb5 a6 4 Ba4 Nf6 5 0-0 Be7 6 Re1 b5 7 Bb3 d6 8 c3 0-0 9 h3 Na5 10 Bc2 c5.

I'm not sure why Chigorin gets the credit for that either. He seems only to have played it a couple of times; his treatment in O.Duras-M.Chigorin, Nuremberg 1906, was quite "unthematic"; and Schlechter had already ventured this set-up against Tarrasch four years earlier. If it's because Schlechter lost and Chigorin won, Schlechter had his revenge – beating Chigorin from the White side in 1907.

Well, anyway, the Chigorin Ruy Lopez it became and subsequently developed a large complex of theory. One variation, amongst many, runs 11 d4 Qc7 12 d5, whereby White hopes to keep the a5-knight out of the game; for example, after 12...Nc4 13 a4 Bd7 14 b3 Nb6 15 a5 or 12...Bd7 13 b3 c4 14 b4 Nb7 15 a4. Note that, in the latter line, 13 a4 is somewhat premature in view of 13...c4!, controlling the white b-pawn and preparing to target the a4-pawn with ...Nb7-c5.

Despite that, 13 a4 is still seen occasionally. E.Hossain-J.Hjartarson, Turin Olympiad 2006, continued 13...c4 14 Be3 bxa4?! 15 Bxa4 Bxa4 16 Qxa4 Nb3 17 Ra3 Rfc8 18 Nbd2 Nc5 19 Qc2 Nfd7 20 Rae1 a5 21 Nxc4 and White was clearly better at this point (later going very wrong in time trouble and losing). Rather than exchanging on a4 so soon, Black should maintain the tension with something like 14...Rfb8 or 14...Rfc8, or an immediate 14...Nb7-c5.

If you're wondering what any of this has to do with the Sicilian, then watch... 1 e4 c5 2 Nf3 d6 3 Bb5+ Bd7 4 a4!? (a harmless-looking move, played by GMs Bukhuti Gurgenidze, Bent Larsen and, more recently, Konstantin Chernyshov) 4...Nf6 5 d3 Nc6 6 0-0 e6 7 Re1 Be7 8 c3 a6 9 Bc4 0-0 10 Bb3 b5 11 d4 e5 (I once won after 11...Qc7 12 e5) 12 d5 Na5 13 Bc2 c4 (13...Qc7 was V.Kosenkov-M.Kletsel, 8th World Correspondence Championship 1975) 14 h3 Qc7 and somehow we've transposed to the diagram, duly making it a “Chigorin Sicilian”.

That was the surprising course of the game below. I tried Hossain's 14 Be3, my opponent replied with 14...Rfc8 and... nothing much else happened. There was one way I might have made it interesting: 29 Qf2 g6 30 Nh5!? gxh5 31 Bh6, but this fails to 31...Ne8! 32 Qg3+ Kf7 33 Nf3 Ke7 34 Nxe5 h4! 35 Qxh4 Ne6!, when Black emerges with the better chances while, most pertinently, White has no chances of more than a draw. It seemed simpler to swap things off and then just offer a draw. So that's what I did.

Tuesday, 3 July 2018

037. The king steps sideways

White: G. Crook - BCCA DJKO 35/1, 2000

The Two Knights with 4 d4 exd4 5 0-0 isn't played very much nowadays, because 5...Nxe4 6 Re1 d5 7 Bxd5 Qxd5 8 Nc3 Qa5 (or 8...Qh5) 9 Nxe4 Be6 is considered perfectly acceptable for Black. All well and good. Black has grabbed the central e-pawn; White has played a little trick (7 Bxd5, 8 Nc3) to regain it. And there the excitement ends for me.

I've mostly preferred the Max Lange Attack: 5...Bc5 6 e5 d5 7 exf6 dxc4. The main line continues 8 Re1+ Be6 9 Ng5 Qd5 10 Nc3 Qf5 11 Nce4 0-0-0 12 g4 Qe5, reaching a very messy position. That's more like it. I like messy positions. Up to a point. What I really like is messy positions where my opponent hasn't already learned what to do. Mainline Sicilians, for instance, can be extremely messy but also mapped out to move thirty and beyond. I'm not interested in all that. Max Lange variations, too, go deep into the twenties, and I'm not interested in those either. Especially not when it's me who doesn't know what to do.

So I started sliding the king sideways: 8...Kf8!?. Theory tends to dismiss this on account of 9 Bg5 gxf6 10 Bh6+ Kg8 11 Nc3.

For the price of two pawns White has shut the h8-rook in the corner and opened the black king to spectacular mating attacks. Such as 11...Bg4 12 Ne4 b6 13 c3 Ne5? 14 Nxe5! Bxd1 15 Nd7! Be7 16 Nexf6+ Bxf6 17 Re8+! Qxe8 18 Nxf6 mate, as in B.Kazic-B.Vukovic, Candidates Tournament, Yugoslavia 1940. Or 11...Bf8 12 Nxd4 Nxd4? 13 Qxd4! Qxd4? 14 Re8 Qd6 15 Nd5 and mate in four.

Except, none of that is forced. In the latter line Black can play 12...Bxh6 13 Nxc6 Qxd1 14 Ne7+ Kg7 15 Rad1 Be6 16 Ned5 “with roughly equal chances”, according to Yakov Estrin in his book on the Two Knights Defence (Batsford 1983). And I think that's being generous to White, seeing as Black still has an extra pawn and the two bishops. Or if 12 Bxf8 Kxf8 13 Ne4, then 13...f5! and again the onus is on White to justify the sacrifices.

I got to test those “equal chances” in the game below. As it happens, my opponent did manage to find compensation, tried to mate me with rook, knight and king, and eventually took a draw by perpetual check. Nonetheless, today's engines give Black the advantage all the way through, even if I was unable to convert it at the time. Certainly I'd be happy to have another go.

That is if White doesn't avoid the whole thing by throwing in 8 fxg7 first.

Tuesday, 19 June 2018

036. The king steps forth

Black: A. Gardner - BCCA DJKO 28/2, 1994

Back in 1993 (when I was reviewing books) I received a copy of Das angenommene Königsgambit mit 3.Sc3 by Volker Hergert & Alexander Bangiev (Schachverlag Reinhold Dreier). Translating for you, that means 1 e4 e5 2 f4 exf4 3 Nc3!?. I duly added this to my repertoire as White.

The obvious drawback to developing the b1-knight first is that Black can reply 3...Qh4+, forcing the white king to step forward. But it's the kind of move that amuses me – and sometimes my opponents too. 4 Ke2. Straight face. Eye contact. It often gets a laugh.

The thing about this sort of opening is that it's never quite as bad as it seems. For the inconvenience caused by the queen check, White aims to gain a nice pawn centre with d2-d4, while the king can seek sanctuary via f2 or d3-c2. Keres played 3 Nc3 regularly in his young days, and both Spassky and Bronstein won with it. More recently, IM Kamran Shirazi has brought it out occasionally; GM Richard Rapport used it to win an Austrian League game; even World Champions Gary Kasparov and Magnus Carlsen have tried it in blitz. My own OTB record (after 4 Ke2) is P15, W9, D3, L3.

Theory generally dismissed this line, not least because 4...d5 5 Nxd5 Bg4+ 6 Nf3 Nc6 7 Nxc7+ Kd8 8 Nxa8 Ne5 9 h3 Bxf3+ 10 gxf3 Qg3 11 d4 Qxf3+ 12 Ke1 Qe3+ is a draw by perpetual check. However, Hergert and Bangiev showed that 9 Qe1!? Nxf3 10 Qxh4 Nxh4+ 11 Ke1 is a serious option, after which White gets rook and pawn (at least) for two knights – even if the a8-knight is simply removed from the board.

In the following game, utilizing a trick from the Schliemann (13...Bb4+ 14 c3 Nxc3 15 a3 etc), I got rather more than that – and my gōng hé king subsequently marched down the board to win on move 23.

Tuesday, 22 May 2018

035. One for the Dustbin

White: afms - thematic tournament,, 2018

Something I've always aimed for in the opening is to take opponents away from positions they know – without thinking – how to play. In the Schliemann, for instance, a lot of players opt for 4 d3, aiming for a more "normal" game after 4...fxe4 5 dxe4 Nf6 6 0-0. In order to rob them of that, back in 1992 I began playing 4...Nf6 instead, and if 5 0-0 then 5...Bc5, refusing to exchange on e4 and offering a reversed King's Gambit should White decide (correctly) to capture on f5. GM Nigel Davies later proposed this for Black in his book Gambiteer II (Everyman 2007).

In truth, I've never done very well with it. OTB my score (1992-2018) is 54.3%: P23, W9, D7, L7. Not very impressive, even if most of those losses were to strong players: James Cobb, Fernand Gobet, Jeff Horner, and Gary Quillan (all now IMs). I've stubbornly stuck with 4...Nf6, trying to find improvements for Black as and when necessary. But I think it's time to throw it in the bin.

One critical position (which can be reached via several routes) arises at move ten.

Black has a 0% score from here in the database (with one of my own losses contributing to that). Essentially, White has a terrific reversed Schallopp (1 e4 e5 2 f4 exf4 3 Nf3 Nf6 4 e5 Nh5), where the gambit pawn is firmly defended. Yes, Black can obtain a nice-looking centre after ...d6-d5, but it's not going anywhere – and it's not even very secure since White can swipe at it with either c2-c4 or f2-f4 or both.

In the game below, my opponent went for a slow build-up before breaking on the kingside, and eventually mated me by promoting a tripled f-pawn: g2-g4-g5xf6-f7-f8Q. It's a matter of taste, I suppose. I'd have used the bishop myself.

I've posted twice recently (Games 035 and 036) about Grandmasters coming to the rescue in my openings, but I can't see any help being forthcoming in this one.

Monday, 14 May 2018

034. That'll learn me

Black: dakarportsmouth - Knockout tournament,, 2018

“If you find a good move, look for a better one” (as Emanuel Lasker apparently didn't say). It's a good maxim for correspondence chess, where you can spend all the time you like looking for better moves. Over the board, on the other hand, I usually opt for the practical “one win is as good as another” strategy. And sometimes this lazy attitude affects my Fernschach as well.

With fewer people signing up for my thematics at nowadays, I recently entered a couple of ordinary tournaments instead: a four-player round robin and a knockout with 64 players. In my own events I restrict the strength (or rather weakness) of my opponents by setting a lower limit on rating. The ordinaries are open tournaments, in other words anyone can enter. In the first round of the knockout I was drawn against someone nearly 1400 points lower rated than me. The first three moves were 1 e4 e5 2 f4 d6 3 Nc3 f6, after which I went into "simul" mode: come to the board, make a quick move, and go away again.

Now I'm regretting it because the engine shows that I failed to spot a pretty queen sacrifice leading to an unusual mate:

Having won a piece, I simply retreated the knight with check here, soon picked up more material, and eventually mated with queen and two bishops. However, Houdini pops instantly out with 16 Qxf6+! and declares a forced mate in eight: 16...Nxf6 17 Rxf6+ Kg7 18 Rf7+ Kg8 19 Rxe7+ Kf8 20 Bc5! (threatening 21 Rf1 mate) 20...Bc8 21 Rb7+ Ke8 22 Bf7+ Kd8 23 Be7 mate, this time with two bishops and rook.

Damn, that would have been really nice.

Friday, 20 April 2018

033. Another Grandmaster to the Rescue

White: AndyAndyO - thematic tournament,, 2017

From the Black side of the Vienna I have for a long time preferred 2...Nc6 to the standard 2...Nf6. Yes, the Frankenstein-Dracula variation 2...Nf6 3 Bc4 Nxe4 4 Qh5 Nd6 5 Bb3 Nc6 6 Nb5 etc can be fun, but I've never had that over the board. What you mostly get is boring equality after 3 f4 d5 or 3 g3 d5.

So I switched to 2...Nc6. Then if 3 Bc4 or 3 g3, Black still has options of a counter-attacking ...f7-f5, while 3 f4 is now far more risky for White due to 3...exf4!. On the other hand, it's very sharp, so you need to be prepared as Black. Eighteen months ago in the Notts League, I faced an unexpected Hamppe-Allgaier Gambit: 4 Nf3 g5 5 h4 g4 6 Ng5 h6 7 Nxf7 Kxf7, and I hadn't studied this in years. Fortunately, nor it seems had my opponent and we played a wholly incorrect game, ending in a draw.

Since then I've looked at the theory again. 8 d4 is considered the main move nowadays and 8...f3! the best response. In Black Weapons in the Open Games (New in Chess 2014), GM Victor Bologan gives 9 Bc4+ d5!? (“In such positions time is an extremely important factor”) 10 Bxd5+ Kg7 11 gxf3 Nf6! 12 Be3 Bb4 13 Bc4 Qe7 14 Qe2 (as arose in J.Gallager-V.Hrsec, Geneva 1991) and now the major improvement 14...Rd8!, intending 15 0-0-0 Bxc3 16 bxc3 Qa3+ 17 Kb1 Qxc3 18 Bb3 Rxd4! (“which is even better than capturing with the knight”).

All good – but there's also 9 gxf3, which I noticed scores over 85% for White in the databases. What does Black do against that? Grandmaster Bologan rides to the rescue! In the database games (dating back to I.Gunsberg-G.Mackenzie, London 1886) Black always played 9...Be7, whereas Bologan follows his theme:

“In the sub-variation 9 gxf3, no one has tried 9...d5!?, although I think this is the only way to exploit Black's material advantage. Now, if we exclude 10 exd5 because of 10...Nb4 (or even 10...Bd6!?, with the idea 11 dxc6 Bg3+ 11 Ke2 Nf6 and ...Re8), the critical line is 10 Nxd5, against which I like the active 10...Nf6 11 Bc4 Kg7, with the idea 12 Nxf6 Qxf6 13 e5 Bb4+ 14 c3 Nxe5!?, when Black has a strong initiative. I saw this theme frequently while analyzing this position. Black returns the piece to destroy White's steamroller.”

I got to try this out in a Vienna Gambit thematic. I actually commented: “not sure whether this idea of Bologan's really gives Black any serioius [sic] winning chances, but we'll see”. It turned out to be far stronger for Black than I'd thought. The forcing 15 dxe5 Qxe5+ 16 Qe2 Qg3+ 17 Qf2 Re8+ 18 Kf1 Qxf2+ 19 Kxf2 reaches a critical position.

Here it seems logical to target the white king on the open lines after 19...Bd6, but I couldn't see any conclusive way forward. So I decided to close the position instead with 19...Bc5+! 20 Kg3 Bd6+! 21 f4 h5, relying on the protected passed g-pawn as a long-term asset, and aiming eventually to get in behind and cause damage with a rook. Although White is only a couple of moves from bringing his own rooks out and challenging for the open files, it's surprisingly difficult for him to consolidate a defence.

In the end I managed to infiltrate via the pleasing manoeuvre ...Re5-c5-a5-a6-b6-b3, after which the game didn't last much longer.

Sunday, 18 March 2018

032. Grandmasters to the Rescue

White: docjan - thematic tournament,, 2018

I hate it when grandmasters start playing my pet lines. It means other grandmasters work out effective responses, which filter down to my level, so that my opponents are no longer surprised and can just learn how to reach sensible positions. The prime example is Chigorin's Defence, 1 d4 d5 2 c4 Nc6!?.

Back in 1982 when IM John Watson wrote his pioneering book on the Chigorin, it seemed nobody really much took notice. I was able to develop a whole repertoire involving ...d7-d5 and ...Nc6 against both 1 d4 and 1 Nf3, securing dynamic play for Black in positions where my opponents mostly had to think for themselves.

And then Alexander Morozevich took it up. Sorting my Chigorin database by Elo now lists a whole string of grandmasters on the black side (if only occasionally), including Carlsen, Kramnik, Ivanchuk, Mamedyarov, Ponomariov, Rapport, Short, Moiseenko, Miladinovic... Not that GM Morozevich need care about that. The Chigorin was a useful weapon for him for a time and led to a lot of interesting games, but he's hardly played it in years. Bah.

Another apparent case is the Schliemann. Sorting that database reveals a similar sort of list: Carlsen, Aronian, Radjabov, Ivanchuk, Zvjaginsev, Nisipeanu, Azarov, I.Sokolov, Nyback, Pruijssers, and so on. But this time I don't mind so much. The lines they play aren't in general the lines I play. After 4 Nc3, for instance, GMs tend nowadays to go for Tartakower's 4...fxe4 5 Nxe4 Nf6, whereas I have always favoured 5...d5. And I'm so well versed in the theory that when grandmaster games do venture into my territory I am better able to appraise their value.

The 5...d5 main line continues 6 Nxe5 dxe4 7 Nxc6 Qg5 8 Qe2 Nf6 9 f4 Qxf4, which GM Ventzislav Inkiov does in fact play. My OTB score with this is 66% as Black: P22, W10, D9, L3 (and two of those losses were to GMs Michael Adams and James Howell, who would probably beat me whatever I played). However, early on, I identified a potential problem in 9 Nxa7+!?, the crucial point being that 9...Bd7 10 Bxd7+ Nxd7 11 f4! Qxf4? sees Black being move-ordered into a bad 9 f4 variation; while after 11...Qc5 12 Nb5 Qxc2 13 d4!, Black ends up having to defend a pawn-down endgame, which is no fun at all. I note that Inkiov has lost this twice as Black.

But here elite grandmasters have come to the rescue. In his book The Ruy Lopez Revisited (New in Chess 2009), GM Ivan Sokolov directed attention to 11...Qf5!?.

Other players have since refined and strengthened Sokolov's analysis; and in a critical line, 12 Nb5 0-0-0 13 a4, GM Liviu-Dieter Nisipeanu brought out 13...Bb4! (improving on Sokolov's 13...Bc5). The game A.Nekhaev-Nisipeanu, correspondence 2012, is annotated in detail in Junior Tay's new book, The Schliemann Defence: Move By Move (Everyman 2018), which I worked on as editor.

One continuation Junior doesn't mention is 12 Nb5 0-0-0 13 Nc3. No one had ever played this, and Sokolov hardly mentions it either, writing only that “13...Bc5, followed by ...Rhe8, ...Nf6 gives Black nice compensation.” So the game below has some small theoretical value. I did manage to find some nice compensation, just as Sokolov had indicated – and some dubious knight moves by my opponent (moves 23-25) even allowed me to go on and win. Hurrah for the grandmasters!