Hanging around against a Master

Final Round

In this last and final round for the month, I was set to face Life Master Brian Wall.  At first, they thought that Alexander was going to show up, and in that case I would have only faced a 900 player, so this was a pleasant surprise for me.

Brian said that his strategy with openings is “to be unpredictable”, and thus his reasoning for choosing Petrosian’s variation of the Petroff Defense, which he said that Petrosian nearly always could draw from; besides, Brian said he wasn’t so pleased about the opening against me last time he faced my Scotch.  I, on the other hand, remember what happened the last time I tried to pull out my unprepared King’s Gambit against a Master (got slaughtered), and so here went for the safer bet.

I played 7.c4, but Stockfish first suggested 7.Nc3, and then changed it’s mind to 0-0, Re1, and c3.  Needless to say, I played more aggressively than that.

11.g4.  I wasn’t sure whether or not 11.Qd2 might be a better try, but Stockfish surprisingly rates the position as 0.0 after 11…BxNf3, 12.gxB d5, 13.cxd5 Nb6, and this is presumably because Black has no kingside pawn-weaknesses, whereas White has one, even though it is an opposite-sides castling position after 14.0-0-0.

13.Qd3?!  This is where I missed my “big chance” to go up +.60, according to Stockfish, with 13.Qb3!  Actually, I never gave it any thought, didn’t see it, and did feel a bit stuck for what to do next.  The line given is 13.Qb3 Nb6, 14.0-0-0 c6.

15.b4?! Stockfish gives 15.0-0 as best, and evaluates it as 0.0.  As tempting as 15.0-0-0 looks, the computer puts black over a pawn up after 15…b6, 16.b4 a5.  The same problem lies with 16.cxb6 because after …axb6, Black will have 17…b5, followed by 18…Qa5, which is an even more winning line for him.

So after 15.b4?!, Stockfish is giving this as =+ for Black after 15…Re8.  However, he returns the favor with 15…b6?!, and after 16.0-0 Re8 (best), as played, now White has the +=.

17.Bf4?!=  It’s hard to knock this move, but Stockfish gives the nod to 17.Re1 Qc8, 18.Bd2 a5, 19.a3+= which is the sort of position that only a computer could seemingly love, but I’d much prefer being on the White side of it.

I spent a lot of time on move 17, and against a weaker opponent, I may have been more tempted to play this strange 17.Nf3-d3 with f4 idea, but I felt it had too many weaknesses.  One line I sparred with against Stockfish is this one: 17. Nd2 Qc7 18. f4 a5 19. a3 Bf8 20. Rae1 axb4 21. axb4 Ra3 22. Ra1 Rea8 23. Rxa3 Rxa3 24. Qc2 bxc5 25. bxc5 Be7 26. Ndb1 Ra5 27. Bd2 Ra7 28. Kg2 Nf8 29. f5 gxf5 30. Qxf5 Bd8 31. Bf4 Qb7 32. Be5 Ng6, but White is down more than -1 here.

Brian Wall was more concerned about if 17.Re1 Qc7, then White can spring an 18.Nc3-e2, 19.Bf4 idea.  I did see this idea OTB, but the main thing that I missed regards timing; commit to the building move first, and thwart opponents plan only after that.

The funny thing about B.W.’s 17…Ne4, is that I thought he would play a pawn-sac in the line 17.Ne2 Ne4?!, but then I sort of slapped myself on the head, proverbially speaking, when he played this move anyway.

When I played 18.Re1, I saw the line 18.NxN exN, 19.Qxe4 Bxc5, 20.Qxc6 Bxb4 (both of these lines are 0.0), but all things being equal, I wanted to keep the tension and keep material on the board.

18…f5.  I thought he would play this, but the computer prefers 18…NxNc3 (best) or 18…g5 (2nd) or 18…bxc5 and calls them all equal.

Stockfish says that 19.b5 is +=, and this would have been difficult to find OTB, given my (typical) clock situation.  In fact, I should look for safe advantages like this one when I play a Master.  Ultimately, after 19…Qc8, that line would be equal too, but at least it would have given him more to think about, defensively.

19.Ne5?  The first bad move of the game, and played quickly (time-pressure).  19.b5 may have been best, but even 19.Rac1 would have been more believable than this move.

21…Bxc5 winning a pawn.  Naturally, I hadn’t forseen this continuation when I played 19.Ne5?

22.f3?  22.NxN fxN, 23.Qc3 Bd6, 24.Qxc6 BxB, 25.exB Rxe seemed unappetizing to me, and B.W. should win this against me, even if it slightly less than -1 advantage.  So, I went for the great unknown with 22.f3?

22…Rxe5!! I thought that he may play this a move down the road or so, but not straight-away.  He’s -2 here.

23.dxBc5?  Relatively best was 23.fxN, but after …fxe, 23.Qc2 Bb6, 24.Kg2 Re8, he’s extricated his pieces in time to remain up those two pawns.  Still, this would be a total loss for White.

When I had 23.dxBc5? I had forseen ….Nxc5 24.Qd4 forking, but had missed …RxR, 25.RxR Qh4, or rather that it is attacking the unprotected rook, when Black remains up the two pawns.  B.W. had seen this line, but then somehow was also contemplating or worried about 25…Nd7, when he is two pawns up but thought I might have some attack out of there, which I thought was rather silly to be concerned about , but okay, it probably would have lead him to the 25….Qh4 line anyway.

One final note is that if 23.Kg2, not only does Black have simply 23…Bb6, once again retaining his two pawn advantage, but also has the lights-out move of 23…Nf2 (24.KxN?? Qh4+ skewer) 24.Qd2 RxR, 25.RxR Qh4, 26.dxB (26.QxN?? QxQ, 27.KxQ Bd4+ picking up the Nc3 as 28.Re3 walks into the pin) …Nxh3, 29.Qe3 (say), then …Nf4+ sets up unavoidable mates, so that 30.QxNf4 Qh4xRxe1+ is -10 for Black.

23….Qa5?  B.W. said that he had seen this move at the last second and decided to play it instead, which is rather strange, and I felt it was a blunder OTB, but I realized that I didn’t really have the time to deal with it properly and at this point needed to pace myself toward an endgame, for clock and energy reasons.

I could see by the look on his face that he was none too pleased with his 23…Qa5?, but I still felt that he had probably done some analyzing of it, as he had spent quite a long period of clock time on this move.  He thought that he “could simply resign” after 24.Qe5, but these are rather harsh terms he has given himself, even if he were to go down an exchange for a pawn after 24…Qc3 (24…Qc5 is same problem), 25.QxQ NxQ, 26.RxR  and this is what we both missed, that the rook is hanging on e5.  It’s tough to say whether or not I would win from this position, with my endgame skills against his, but it’s all rather a moot point as he has instead 24…Ng5!, and if 25.QxR, then …Nxf+ followed by …NxQe5 wins, so he was still up -2.7 after 24…Ng5!  Actually, my move of 24.fxN was quite attractive in it’s own right, as I am getting a knight for three pawns, and Stockfish says it’s equal (0.0).

26.Na4?, it’s interesting how Stockfish at first thinks this is equal, then says it’s down -1.37 after …Qe7, and this is mostly because the other move that I was considering 26.Qg3 -1, gets out of the way of the pawn-roller.  Interesting note is that at this point, I initially thought I’d probably lose after the queen trade, and then …Rb8, because B.W. is AKA “The pawn-wave guy”, as he has won a lot of games this way.  Ironically, he thought I had made a grandmaster move, forcing him to trade queens, and that he thought he was probably losing.  Stockfish finally settles on 28….Rae8 =+, after the queen trade.

32.Rec1.  By the time I played this move, I knew that this whole maneuver wasn’t best, but it was easy-cheesy to play on the clock.  In fact, the computer was suggesting 31.Rc3 instead of 31.Rc5, but that is a very computer-like move, and difficult for a human to handle in time-pressure.

33.RxRe5  Again, I didn’t feel that this was the best continuation, so much as it was the easiest one to play on the clock.  I had looked at 33.Rc4, but then thought after 33…d3, 34.Nc3 that he could simply start pushing the pawns instead of merely bringing his other rook into play.  However, when I showed this line to B.W.,  we looked at …d2, 35.Rd4+ Rd5, 36.Nxe+! Ke5 (we are following Stockfish) whereupon B.W. said “Is this was you were worried about?”, but after …Rxd2, 37.KxNe4 Rf1 it is =+, although should be a draw with best play.  Well, believe it or not, the line that I chose is a 0.0 draw.

45.Kg3??  My endgame inexperience finally shows.  45.Nxd2! was 0.0.  The big difference between the two lines is that now 47.Rxc+ comes with check.  I felt his strongest move now was 48…Rc4+, and I was happy that he gave me some time here to think about my response, as he left for the restroom, and the reason he didn’t play it was because of 49.Kg3, also best but around -6.7 according to Stockfish.  The reason 49.Kg3 loses is that because Black can then “shoulder” off the White king with ….Kd2, and …Ke3, whereupon Black will eventually gobble up those kingside pawns.  Also, 49.Kf5 will lose to 49…Rh5, and 49.Kh4 will lose like in the game.

48….Ra3?  Throwing away the win!  When he played this move, I felt relieved, but the mental catharsis left me lazy, and prone to making the devil-may-care sort of move that was played later in the game.  At first glance, it’s hard to understand why it is necessary to play 48…Rc4+ first.  Well, if 49.Kg3, White gets shouldered by Black’s king.  If 49.Kf5, then …Rh4, and if 49.Kh5, then the White king is curiously blocking his h-pawn, which he can only promote by getting to g7 and playing h8, but then a1(Q)+! will foil that and win for Black.

Even should White play correctly in the line 48…Rc4+, 49.Kh5 Ra4, 50.h3 Rxa2, 51.Kg4 it is interesting to know why it loses and that is because when the pawns get to the 7th rank, pushing to h8 will lose because rook on 8th rank and Qa1 will cover that square, and after g8(Q), which B.W. and Daniel H. both thought was a draw, Black is still winning with …Qe5+ when the White king will either get skewered along the g-file, after Kh4…Qh2+ or will get mated (Note:  A Rc2-c8 idea/line is never necessary, as …Rg2 works in all of the lines).  The difference between the winning 48…Rc4+ variation, and the 48….Ra3 drawing variation is that while, yes, it would seem the 48…Rc4+ loses a tempo toward the goal of taking the a2 pawn, White loses TWO tempos moving the king back to g4, and so that net difference of one tempo is the balance between winning and losing.  Trust me when I say that it took me a long time, too, starting at the computer, to figure out what the difference was.

50.Kf5?? Needless to say, it was this pointless move, which B.W. spotted instantly, which was the culprit.  I’ve done a really poor job with pawn-races, in general, and was incredulous that he could play 50….a5!! for the win, but just as surprisingly understood that I was losing right after I saw it and calculated it.

There were two times during this game where I forgot to push the clock, and when I actually flagged I noticed it right after it hit zero.  I asked him how we would have continued and he immediately said that if nothing else, he has QxQ followed by Rh2.  He actually played it for me, and Earle said he was looking at that, too.  So, all in all, that is how I lost to a Master!  🙂

Another interesting, yet disconcerting aspect of B.W. post-mortem style is that he will look at the game once, in a half-interested way at most about the game as a whole, but then he will be willing to go over the endgame portion of almost any interesting endgame, endlessly.  We went over this game at least five times, and 90% of that was going over different endgame tries at the very end of the game.  Lest you wonder where Master’s look for rating points (the endgame), now you know!

My new rating is 1728 – and hitting the floor at Mach speed!  This has mostly been because with my lowered rating, I am now the top of the bottom-half, and therefore face the top of the top-half, in round one, and even after that my pairings seem to make me a marked man.  Nevertheless, this is the first game I can remember where I never got nervous during a game, never sweated before/during/after the game, was cool as a cucumber.  I think between playing up and playing a lot of online chess, it’s started to feel as if playing a game of chess were just the most natural thing.  I even played quite a bit of online chess before the game yesterday.