It was a last round and my opponent was a new, Russian-speaking guy, unrated, but his rating after the tournament should be about 2000. He replied to my Semi-Slav by 4. g3 and it was a line I didn’t know well.

His 12. Ng5 looked like a premature attack. The idea of getting my bishop to f7 was to play e5 of course, but I didn’t play it, having second thoughts about possible weakness of the pawn on e5. His 31. d5 looked strong to me and I spent a lot of time considering my replies and lost my half an hour advantage, but computer says it was equal.

After his 37. Rd3 I got worried about Re3 and possible attacks on 7th horizontal and got an idea about perpetual, it explains the title. So I played 37… Qe1+ which was a mistake and then played 38… Re8, which was a fatal mistake. The problem was I didn’t see move Rf3, completely defeating my idea. I could play instead 37… Rc8 perfectly holding the position.

So after a few moves it became clear that the game is lost and after I found myself in a hopeless rook endgame I resigned.


My opponent was an old guy, our score in the past was =1, -2. I got Black and played Semi-Slav. I didn’t quite like his 6. b3 though he would be OK after 6… Bb4 7. Bd2, but he played Bb2. I saw of course Qa5, but didn’t realize I could win a pawn after 8… Qa5 9. Rc1 Qxa2. Then I missed another opportunity by playing pre-move 10… Nxc3 instead of Ng4 winning “f” pawn. I saw Ng4 before playing Ndf6, but thought that he had Bg2 and O-O and after a3 forgot about it.

After 26. Bxd5 I spent quite some time, because I wanted of course to play exd5, but thought that he will occupy “e” line, so played cxd5 which I knew was worse. Actually I could play exd5, because he has to lose a tempo to play 27. f4, otherwise if 27. Rde1 I play 27.. f4 and if Re7 then Qh3 with advantage.

On move 32 I considered putting my rook on g3 with a purpose of sacrificing on g3, but didn’t see a forced win or even draw, so decided to play h5. Computer thinks that 36. Rg5 was better than Qg5 and was equal.

Then he suddenly blundered playing 39. Rg5. He had 45 minutes vs. my 25 and the only reason for the blunder, I think, was feeling under pressure. Though computer evaluates the position as equal after 39. Kf1, I would prefer to play Black here. Then after 40. Qxd4 he played Qc3.  I saw the check before that and certainly it looked very bad for him. After 42 … Rh3+ it was a mate in 2 and he resigned.

These are the words from the song in one good Russian movie. I think I can relate that to what happened in my game on Monday.

I played with an expert to whom I lost 3 times before. But in the last game I missed a simple deflection that was winning the game. This game was somewhat similar, missing a chance due to a time pressure.

He had White, I played Slav again. I think it suits my style, but I definitely should study it more. I think I was OK after the opening, though was feeling some pressure. I spent quite some time thinking how to defend against his 21. Nd6 and computer confirms that f6 was a good idea. Maybe I got tired defending, on the move 31 I blundered a pawn. Then 37… Nh3 was another blunder because I was losing this knight after 38. Re5 Qh6 39. Kg2 and if 39… Ng5 then 40. Qd5. He didn’t see it. After 41. hxg3 I decided to play Nxf4. It was more of an intuitive sacrifice and I had very little, 20-30 seconds, time left. Computer suggests the same move. So after gxf4 I automatically play Rxf4 and it blinked a bit later in my mind that Qh4+ was possible. Of course I had to play Qh4+, winning a rook,  computer says it is still a draw. Though, I am sure how he would react on that strike, also having not much time.

After his 43. Qe3 I missed another strike  – 43… Rxe4 and if  44. Qxe4 then 44… Rf1+, so he has to reply with 44. Qh3+ and I am a pawn up. Looking at this now I am just shaking my head, but at least you can’t require somebody to see this having just seconds left. Instead I made that stupid check on f1, because it seemed to me that I can mate him or win his queen. Of course, nothing like that happened, he forced queens exchange and after some meaningless resistance in the endgame I resigned.

Unlike my favorite Magnus Carlsen who gets a win by grinding his opponent, in the last two rounds of the tournament after grinding my opponents and having a winning position in the end I got nothing (a draw with a much lower rated opponent).

After I got only 1 point (2 draws and a loss) in the first three rounds, I decided just try to play a good game. I had a +3, =1 score with my opponent, but decided not to take him lightly. He had White and played Queen’s gambit. I wasn’t intimidated by his aggressive g4, but  decided to be careful.  The intention of 10… Nd5 was to close the diagonal, I also considered the knight sacrifice on f7 and thought that Bf6 should be enough to defend. After long thinking he played Nxe6. I am really proud of my 14… N7b6, as Fritz plays the same. Then I found that there are more threats than I thought and his bishops are very dangerous. My exchange sacrifice was planned right after 16… e5, but after my king got on g8 it actually wasn’t necessary, because after 18. Bh5 computer recommends Rh7  with ~-2 evaluation.

I still considered the position after move 21  dangerous and went for the queens exchange. Computer thinks it was a big mistake with my advantage dropping by almost 2 points. I still had some, but my Be7 made the position equal. Soon after rooks exchange he offered a draw, I refused. I was feeling the power of my bishop and knight playing together and decided to try to win. The funniest part started when he lost his pawns. I realized that if he gives up his rook for the pawn I don’t know how to mate with B+N. So I tried not to let him do that. He also had essentially less time, so I could just ignore this idea of rook sacrifice and even with B+N try to pretend that I want to mate him. I didn’t write down my last moves because of the time left being less than 5 minutes. What happened was that with him having a few seconds left I put my bishop, king and knight on one line and after taking the bishop he sacrificed the rook for the knight and declared a draw. He had 3 seconds remaining.

I just realized that he still had to take my pawn, though his king was nearby and probably was able to do it in time. I will add the next game to this post later.

It is exactly how I felt during this game of the first round of the big tournament. The guy got White and we played Queen’s gambit Slav, Exchange variation. I misplayed it and had to go back with my bishop. Still I was OK until I played these bad moves Nh5 and f6. Right after f6 I saw Bc7. He played it and I got into trouble.

Then after Nh4 I thought that I lost the game, that’s how bad it looked. Amazingly he played 20. Bg6 and I saw f5 right away. I couldn’t believe my eyes! Then I  probably played it too safe, computer doesn’t like at all my queens exchange. For the price of the pawn I eventually got all my pieces in play and after 45… Rc4 I felt that I am winning. But the guy was not in  a hurry to agree… My Na3 was the right move, but 49… fxe4 was of course a mistake, g6 was keeping the fort. The same g6 that I played one move later could cost me a game after f5, as it is losing by force due to a two passed pawns and remote knight. He didn’t see it.

The play continued and then he played 54. h4, to which computer puts ?? comment because of Nxe5. Would you believe me if I will say that I considered Nxe5 at least a few times? But it seemed to me that his king gets to the pawns in time. What I didn’t see was that my king after eliminating “b” pawn succeeds in taking “e” pawns too. And the last thing I didn’t see with 8 minutes remaining (no increment) is that I could play 56… h5 and get my king to c6. Still, natural Kxb6 allowed him to get a draw after f5 and my king is too far. So, only careful maneuvering with  a knight and a king wins here. After 58. Kb5 he said it was a 3-fold repetition. I didn’t see a win, so agreed to a draw without even checking the score sheet.

I can only add that the guy played really well later at the tournament and shared a second place with 4/5.