Κυριακή, 7 Αυγούστου 2016

Corr Lines I - Giri-MVL, Norway 2016

 In this series, we will be examining topical opening lines. We single out ideas tested in Correspondence play that have influence over what the super-elite plays over the board. When possible we will refer to analysing techniques that seem to work good and produce results in competitive Correspondence Chess, at least according to my experience. Posts will be written in English.

In this first installment we will be looking at a Najdorf line that was played in the game Anish Giri - Maxime Vachier-Lagrave at the Norway 2016 tournament earlier this year.

Our story, though, starts much earlier. On the 15th of September 2015, my game for the LSS World Championship 2018 Quarter-Final against the American ICCF SIM David Enrique Hernández Molina started and we both blitzed out - spending only minutes - 1.e4 c5 2.Nf3 d6 3.d4 cd4 4.Nxd4 Nf6 5.Nc3 a6 6.Bg5 e6 7.f4 h6!? 8.Bh4 Qb6. The !? designation on 7...h6 probably needs some explanation. In Correspondence chess nowadays, at least at the level of this game and higher, the Poisoned Pawn Variation of the Najdorf is generally considered to be drawn with perfect play. There are some really long computer lines that are mapped out and exhausted even up to the 60th move. There is no memory contest here. Everybody plays with books open and engines on. Occasionally there are some new ideas in certain lines -  like the ones that GM Negi presented in his repertoire book against the Najdorf, but always these ideas only work for a couple of games and the way to disarm White is found with astounding speed. It's impossible to see something like Radjabov-Grischuk, Tbilisi FIDE GP 2015 or Najer-Vachier Lagrave, Dortmund 2016

Therefore, the 7...h6 move-order that my opponent chose is equivalent to a direct provocation NOT to enter the Poisoned Pawn proper at all and a clear declaration that Black is playing for a win even accepting a slightly sub-optimal Najdorf position to achieve it. To cut a long story short, it allows 9.a3!? a move that I instantly played that same day and then put that game to rest. My opponent took one hour to decide on 9...Be7 but I had already stopped paying attention by then. I was focusing on my ICCF games at the time trying to cross the 2300 barrier which I eventually managed before the deadline of the 30th of November and so I was at last eligible to join an IM-norm tournament beginning in January of the following year.

During the next month I occasionally played a move and gradually exactly one month after the beginning of the game we reached the position after 10.Bf2 (the whole White's idea behind 9.a3 is based on the fact that he can play this move) Qc7 11.Qf3 Nbd7 12.0-0-0 b5 13.g4 g5. Here I took another 20-day sabbatical from this game focusing on other positions.

I returned to check this at the start of November. Basically there are two ideas that people usually try in this position. The first is to play 14.Bg2 Rb8 15.Qh3 which is considered harmless after the super-level Corr game Grigoryev-Zhak Champion's League 12/A email, 2012. Boris Mikhailovich Zhak is a world-class player and showed the way for Black. 15...b4 16.ab4 Rxb4 17,Rhf1 h5!= Efforts to deviate on the 17th move with 17.Na2 didn't lead to any success. Actually, White has only won two (2!) games after 14.Bg2 in Corr, the one against a much lower-rated opponent and the second against someone who for some reason deviated from Zhak's plan.

The second plan is to play the sharp 14.h4 gf4 15.g5!?.  Bear in mind that after 15.Bg2 in this line the simple 15...Ne5 attacks the queen and the pawn on g4 at the same time. It may seem childish to note something like that, but actually this simple fact led me to the idea that introduces the critical line that by now is considered to put the whole line under a new light. But before all that I tried to check the position after 15.g5!? Ne5 16.Qxf4 Nfg4 (again, two black forces go to g4, one white force protects it) 17.Be2 Nxf2 18.Qxf2 Rb8 19.gh6 Rxh6 20.h5 b4 21.ab4 Rxb4 which was already known from the games Kunzelmann-Batrakov, Gold-2012(RUS) and Brookes-Burne, MT-Hollis (ENG), 2014.

At this point I started a routine procedure I follow when I reach critical positions in my analysis. I started an IDEA project with this position, then fed the tree with various continuations I created by hand, checking all moves that had been played before as well as the top recommendations of the engines and finally moves that seem logical to me. All  in all, a spaghetti of variations gets generated and fed to the IDEA project which I then proceed to leave crunching for some days.When I return to it, I check all IDEA results by hand and do the same thing all over again. The whole thing is comprised of 3-day cycles of IDEA analysic aided by a couple of hours of manual labour between each 3-day cycle. I stop when I decide that I can't get anything more out of it or if I find a winning combination (or sometimes, alas, a losing one). This one didn't even get past the first 3 days. White has nothing here as expected - since much higher rated Corr players than myself weren't able to break up Black's position.

And then, at long last, I put 14.h4 gf4, did a database reference check there and saw Nowak-Khan, LSS 2014 and for the first time a continuation with the bishop going to the e2 square! From the moment you put the bishop there on the board, you understand that it's a strong idea because in the majority of lines that Black ends up OK, he makes use of the g4 square. This idea breaks the 2-to-1 force ratio on that square. Moreove, the bishop is doing absolutely nothing on g2 to start with. After this epiphany, my focus was turned to variations occuring with the bishop on e2. Convincing myself that I have at least practical chances there, I have decided to change the move order and play 14.Be2 first, if not for other reasons, to keep my opponent under the illusion that this was a novelty. It turned out that it wasn't a novelty even on the 14th move. There was a game played in a Russian server Nekhaev-Yamaliev, RCCA Gold email, 2013 where White also put the bishop on e2 before playing h4. But I found this game much later.

My opponent played the expected 14...gf4 15.h4 (transposing to Nowak-Khan) Ne5 16.Qxf4 and now proceeded with the dubious novelty 16...Ng6?!N

 I will not bore the readers with providing all the details of the game after this. It is suffice to say that this position offers a slight but stable and undisputed advantage to White. It took me 4 months to convert to a winning a position and an additional 2 for my opponent to come to terms with that. I got a very precious point that might give me a ticket to a semi-final - to understand the importance of this Be2 idea, a +2 score will probably be enough to achieve the promotion.

Game proceeded as follows: 17. Qe3 Nd7 18. Kb1 Bb7 19. Qd2 Nf6 20. Bg3 Ne5 21. Qf4 Qc5 22. Bf2 Ng6 23. Qf3 Ne5 24. Qg2 Qc7 25. g5 hxg5 26. hxg5 Rg8 27. Rh3 Nfd7 28. Qh2 Ng6 29. Bg3 Nde5 

30. Ncxb5! +- (1-0 in 50 moves), Sarakenidis-Hernandez Molina, WC-2018-Q-00008

All this would not be worthy of a blog post. However, in Norway Chess 2016, on the 20th April this year and while casually watching the games I watched MVL going for the 7...h6 move order. His Najdorf score with Black is unbelievably good and he goes to provoke opponents into playing those Najdorf positions because it fits his style so much. Giri dully went into the 9.a3 line and by now I was focusing on this game full time. So you can imagine my excitement when 15.Be2 was played!  Of course the commentators labeled this move as a novelty and I've become used to them not having correspondence games in their databases. I don't scream at the laptop anymore when such innacuracies are mouthed. But I did scream after 15...Rg8N? was played. "THIS LOSES!". Anish strangely enough didn't find the way and ended up on the receiving end of a brutal beating, but Maxime himself notes in New In Chess 2016/4 that White does indeed win. He also offers his take on the whole line, after noting that "15.Be2 is a great idea, already seen in correspondence chess which I wasn't aware of!". Let's see what he has to say - and what he misses even in the analysis. First of all, let's get 15...Rg8?, the game continuation out of the way:

Anish Giri-Maxime Vachier-Lagrave, Norway Chess 2016, after 15...Rg8

MVL gives two different winning lines for White in his notes (italics are mine):

A. 16.g5! is the most logical and natural continuation

- 16...Ne5 17.Qxf4 hg5 (Giri was afraid of 17...Nfg4, but then there is 18.Bxg4! hg5 19.Qg3 gh4 20.Rxh4 Bxh4 21.Qxh4 Nxg4 22.Rg1 +-) 18.hg5 Nfd7 19.Nxe6! is the move that MVL missed during the game 

19...fe6 20.Rh7! Bb7 (20...Kd8 21.Bh5 Bb7 transposes) 21.Bh5+ Kd8 22.g6 +-

- 16...hg5 17.hg5 Rxg5 (?)

17...Ne5 seems more tenacious. 18.Qxf4 Nfd7 19.Nxe6! fe6 20.Rh7 +/-

18.Rh8+ Rg8 19.Rxg8+ Nxg8 20.e5 Bb7

After 20...de5, 21.Ndxb5! Qb7 22.Qh3! +- 

 21.Nxe6 +-

B. 16.Qxf4! works as well, 16...e5 17.Qxh6. Now MVL mentions only 17...exd4 18.Bxd4 +/- and 17...Rg6 18.Ne6!! +- that lose easily. I did try to defend the position:

Analysis Diagram after 17.Qxh6

17...Bf8 18.Qd2 Nc5 seem like a good practical try but White has 19.Bxb5+ ab5 20.Ndb5 Qc6 21.Bxc5 Qxc5 22.b4! Qc6 23.g5 with an overwhelming advantage

Now let's return to the critical position after Giri's 15.Be2:

Let's start with

A. 15...h5. 16.g5 Ng4 (16...Ne5 17.Qxf4 Nfg4 18.Bxg4 Nxg4 19.Qd2 Bb7 20.Bg3 +/-) 17.Qxf4! e5 18.Nd5 Qb7 19.Qg3  +- is as MVL notes totally hopeless for Black

B. 15...d5 16.g5 Ne5 17.Qg2! Nxe4 (17...Rg8 18.ed5 Bb7 19.Qh3! Nxd5 20.Nxe6 +- MVL) 18.Nxe4 de4 19.Qxe4 Bb7 20.Nxe6!

20...fe6 21.Bh5+ Kf8 22.Qxf4+ Kg8 23.Bf7+! Kh7 24.g6+ Kg7 25.Bd4 Bf6 26.Rhf1 Nd3+ (everything else loses instantly) 27.Rxd3 Qxf4+ 28.Rxf4 e5 29.Rxf6! Kxf6 30.Bc3 +/-

C. Maxime in New In Chess thinks that 15...Bb7 is the best try. He gives:

16.g5 Ne5 17.Qxf4 Ng6 18.Qf3 hg5 19.hg5 Rxh1

20.Rxh1(?!) Nd7 21.Be3 Nde5 22.Qf2 0-0-0 +=

But in the diagram position 20.Qxh1! seems an improvement over MVL's line and after 20...Nd7 21.Bh5 White has a clear advantage. Still 15...Bb7 keeps the game going

D. 15...b4 doesn't even get a mention in the NIC analysis. It is the sister variation to 15...Ne5 that we will focus on next. The idea of 15...b4 is that after 16.ab4 Ne5 17.Qxf4 Neg4 18.Bxg4 e5 White doesn't have 19.Nd5?? Nxd5 20.Qf3 as he has in the 15...Ne5 line because now 20...Nxb4 comes with a mate threat on c2 and Black has turned the tables! What White has instead is 19.Qxf6!!

19;..Bxf6 20.Nd5! Qd8 21.Nf5 Bxf5 22.Bxf5 (all this is practically forced) which is a position that the engines do not understand at all. They say 22...Rb8 23.c4 Be7 is equal and when you show them 24.Kb1! they see the light! Or 23...a5 24.c5! Rxb4 25.Kb1 Rc4 26.Nxf6! +/-

E. 15...Ne5

MVL gives 16.Qxf4 Nexg4 17.Bxg4 e5 18.Nd5! (now this works and not 18.Qxf6? Bxf6 19.Nd5 Qd8 20.Nf5 Bxf5 21.Bxf5 Rb8=, nor 20.Nc6? Bxg4 -/+)

18...Nxd5 19.Qf3 Bxg4 (19...Nf4 20.Bxc8 Rxc8 21.Rd2 followed by Nf5)

20.Qxg4 Nf6 21.Qf3 ed4 22.Bxd4 Rg8 23.Bxf6 Bxf6 24.Qxf6 Rc8 25.Rh2 Rg6 26.Qf4 +/-

However, he has missed something very important after 19.Qf3:

19...Nf6! is the move that saves Black. I believe MVL was misled by the engines.

20.Bxc8 Rxc8 21.c3 Rg8! 22.Nf5 a5!! and only now the engines wake up and understand what's going on! 23.Rhg1 b4! with only a symbolic advantage for White. The position is not winable anymore.

The conclusion is that Black holds with difficulty. The line will dissapear though. Because its purpose was to provoke White into a complicated position to play for all three results. And the 14./15.Be2 idea has shut down this possibility once and for all.

I hope you found that interesting.

Δεν υπάρχουν σχόλια: