After my first 3 games (described here), I was sitting on 3 points and my reward was the black pieces against GM Anton Kovalyov (2601 FIDE). This was our third game in 5 months – I drew the first one as black in June before losing as white in August.
I had a feeling he would play 5.b3 against the Semi-Slav (it was his first time doing so), partly because it suits his positional style and also because in the 2 games I have in the database against it, I lost to GMs Granda Zuniga and Vescovi. Neither loss was due to the opening – in fact, I was worse out of the opening and Granda but then outplayed him in the middlegame to reach a completely winning position, while against Vescovi, I equalized in the opening only to be outplayed in the middlegame. However, I had expected somebody at the Montreal International in August/September to try it out against me, so before that tournament, I had done some work on the line and I got to use that preparation here.
Here is the position we reached after 15.Ra2-c2:
I had met his 5.b3 system with a Stonewall setup and he decided to force an exchange of bishops on a3 which led to a queen exchange there as well. Black shouldn’t have too many troubles in this endgame, but he still should be a little careful to avoid drifting into a worse position.
I played 15…Ba6 16.Ke2 c5, fighting for the center. White can’t play 17.cxd5 yet because of 17…Bxd3+ 18.Kxd3 Nxf2+, picking up the rook on h1. After his 17.Rhc1 move, though, White can take on d5. Pretending that Black passes, White will play 18.cxd5 Bxd3+ 19.Kxd3 exd5 (19…Nxf2+ 20.Ke2 isn’t much better) 20.Ke2, when Black has to worry about his pawns on d5 and c5 all the time.
After a long think, I came up with the correct solution – after 17.Rhc1, I played 17…Nef6!. It might seem a bit odd to retreat Black’s nice knight, but now Black is ready to capture on d5 or c5 with a knight, keeping the files closed and Black’s pawns out of the White rooks’ line of fire. Funnily enough, it was this same knight move that helped me equalized against Vescovi, although the situation was quite different (I wanted to bring it back to d7 to fight a white knight on e5). After this, it wasn’t too difficult to hold the endgame even in time pressure (I ended up with 5 minutes to his 58 minutes!).
In round 5, I had the white pieces against GM Tamaz Gelashvili (2610 FIDE, Georgia). He played an offbeat opening that I wasn’t so well prepared for (The Two Knight’s Tango: 1.d4 Nf6 2.c4 e6 3.Nf3 Nc6), and while I should have been better after the opening, I accidentally let him go after my dark-squared bishop with his knight. In order to save it, I had to do some funny stuff that cost me some time and he gradually outplayed me in the middlegame.
In the following position, I was nearing some time pressure and had just played 32.Re1-e4:
I had lost/sacrificed the b4-pawn, thinking that with his bishop making the trip from g7 to take on b4, his kingside would be weakened and I’d be able to take advantage of that. The game continued 32…Bd2 33.Rd1 Bf4 34.g3 Bd6 35.hxg5 hxg5, and it was pretty much only now that I realized that on my planned 36.Rg4, he could play 36…Ra3 and grab my f3-pawn. After the check on g5, his king steps over to f8, as it can’t be hurt there. For some reason, I had only been counting on something like 36…Qg7, which fails to 37.f4.
After losing the pawn, he really should have found a way to finish me off – instead, he played some natural, but not quite best, moves and I managed to hang around. In the following diagram, I was down to a couple minutes against his nine, and decided that my best chances for a draw were to sacrifice an exchange and play with the bishop pair against his exposed king:
With that in mind, I played 47.Qa3+ b4 (47…Qd6 48.Qa7+ and it isn’t obvious how Black is going to make progress, while on 47…Kf7, I was planning 48.Qd3) 48.Rxb4!? Nxb4 49.Qxb4+ Kf7 50.Be4. With Black’s king being so exposed, I was hoping that I would be able to drum up enough counterplay for a perpetual. In the end, I did win the c6-pawn, but he got my d4-pawn and brought about an exchange of bishops. After a lot of maneuvering and posturing to try and gain time on the clock, Gelashvili finally decided that he wasn’t making any serious progress and went for the endgame.
He played 70…Qxf4+ here, to which I replied 71.gxf4 and managed to hold a draw by means of a fortress. Black’s problem is that his king has a couple routes into the position (via c5 and h5, for example), but both of them take his king too far away from the e6-pawn. If the pawns get exchanged, it’s a simple draw for me, so he can’t stray too far. He first tried with his king coming around to c5, but with my king on e4 and d3 (when it was checked away by …Re1+), he didn’t find a way in. He then brought it around to h5, but that allowed a nice drawing finale:
Black has just played 96…Rc3-c1, thinking he’s stopped 97.Bd7 (which would have won the e6-pawn). However, I played 97.Be2+ Kh4 (97…Kg6 would have admitted that the Kh5 journey was not making any progress) 98.f5! Re1 99.fxe6! Rxe2+ 100.Kf5. White only has a pawn for the rook, but the king on f5 shoulders Black’s king on h4, and Black can’t avoid a draw. The game ended 100…Kh5 101.Kf6 Kh6 102.e7 Kh7 103.Kf7 Rf2+ 104.Ke6 Re2+ 105.Kd7 Rxe7+ 106.Kxe7 with a draw.
Phew! That was my longest game of the event and we finished at about 1:30 AM.
In round 6, I was black against the French-Israeli GM Thal Abergel (2533 FIDE). This wasn’t a particularly exciting game, as he played a Scotch against me and I sacrificed a pawn in the opening to get a lead in development (and his king stuck in the center). He managed to bring about a trade of queens, after which I played 20…f7-f6 to reach the following position:
White’s problem here is his lack of development and exposed pawns on f4 and c4. Even though he’s up a pawn here, those problems will make sure that he can’t hang onto it. In fact, he might well end up down a pawn here!
He didn’t play it quite correctly in my view, as he missed a resource of mine after 21.e6. Instead of that, I think 21.exf6 was correct, acquiescing to an equal position after 21…Bxf6 22.Rb1 Rd4!? (both 22…Rfe8 and 22…Rd3 are also interesting) 23.Be3 Rxc4 24.Rhc1 Rxc1 25.Rxc1 Be5 26.g3 Bd6 with equality. He can’t take the pawn on a7 because of …Ra8xa2+, but material is equal and he’s caught up in development.
Instead of that, he played 21.e6, trying to hang onto his pawn. After 21…Rfe8 22.Re1 Rd4!, though, he realized that 23.c5 is met by 23…Bf8! when Black is on top! If the pawn advances on to c6, then Black can try either 24…Bc5 or 24…Rc4 with advantage. He decided to cut his losses with 23.Be3 Rxc4 24.f5 (the f-pawn was likely to be lost anyways, but this way he ruins my pawn structure) gxf5 25.Rec1 Rxc1 26.Rxc1 Rxe6 27.Rxc7 a6 and a draw was agreed in short order. The a-pawns got liquidated when Black’s extra kingside pawn is good for nothing.
So, after 3 straight wins, I had 3 straight draws (although none were without an interesting moment or two) and had 4.5/6. I’ll recap the last third of the event in a later post.