…usually don’t end well, and tonight was one of those nights for me, as you can see.
I spent a long time on the move 7.c4, and in fact I saw the entire line that was played all the way until 13…Nxa1, where I didn’t count the Na1 to be attacking my Qb3. I figured that I was getting two active knights for the rook in the corner.
As do most games at G/90, this game came down to psychology and emotions. I know that Isaac plays active, so instead of choosing any of the other plans I had considered such as c3 or sacking a pawn with 0-0 or Rc1 for powerful play, I instead decided to try and catch my “over-active” opponent.
What I failed to consider is how well this kid plays in concrete situations; he plays like a computer. Isaac is basically playing “chess engine moves” from the time I actually give him something tactical which is concrete. I used to pride myself on my calculating ability, but now I realize that I need to rely more on my feel for the game, and give opponents more nebulous positions to look at, which is what a lot of the super-GMs actually do to their opponents – they understand how risky such lines are.
I had realized that move 16.Na3?? was the critical moment of the game, and cringed when I played this move, but G/90 is a real beast for sharp positions. 16.Nd4 is the move which keeps the game alive, and I had spent quite a bit of time looking at 16.Nxa7+ followed by 17.c6, but it was simply taking too much time and I wasn’t finding all of the follow-up ideas, so decided to overprotect c4, but then realized that I can’t now play 18.Bf3 because 18..g5 wins my knight, and in any case I don’t even have that extra move because 18..g6 simply wins straight-away for Black. Even ideas like g3..Bh3+, Ng2 lock my h1 rook into the corner. A truly ugly and hopeless position. I would have been far better off playing the c3 or pawn-sac lines, and was relying too much on bad psychology for move-selection.
My opponent had also realized that after 10.Nh4?, he could also win with 10..Nxd4 -+. I guess I was right that he went after the most trappy move himself, but I simply failed to defend with either 10.Qb1 =+ or 10.Qc1 =. My defensive skills are not as good as RollingPawns are, so it’s a wonder that I try to play this way. I find it much easier to attack than defend at G/90 and faster time-controls. I now think it is easier to see the board 2D, such as online, and it is much more difficult to see 3-D unless standing up and looking down at the board of course, or away from the table where it’s easier to see everything going on at once, so that 90 minutes isn’t as much time OTB as it is online.
Oh, I looked at Isaac’s clock after I had dropped my rook and it said that he had 58 minutes left. How depressing is that? Isaac describes himself as “the most inconsistent chess player in Colorado”, but now I understand why. When Isaac plays all the time (and takes his lessons from Lee Simmons who is a “retired Expert” player) I’d say he plays around 600 points stronger than when he is only occasionally playing chess and doing more of making videos and writing music (he can play many instruments). Also, Isaac said he has been practising on ChessTempo and his rating there is 1744! Well, I just went there and solved 6/6 and my rating there went from low 1300’s to 1760. Yay! I bested his rating and only took me 20 minutes. hehe.
I understand what Botvinnik meant by “forgetting” that something was en-prise or such. If you don’t think quickly at the board and feel flustered in time-pressure, you can miss really obvious stuff. In this case I had both seen and forgotten that after 17.Bf3, the reply ..g5 is not possible because of 18.Bxg5, but here is the “sick” continuation that Black has at his disposal:
17.Bf3 Bg7, 18.Ke2 Bf6, 19.g3 g5, 20.Ng2 Bf5, 21.RxNa1 Bxb2 – wins the Na3 and Rb1 isn’t possible because of Bf5xRb1, and then even after Bxg5, Black will respond with Bxc5.
When it comes to trapped pieces, I used to think that the burden of proof lied in the hands of the trap-ee, but now I realize that it lies in the hands of the trapp-er, even more so. No wonder Fischer appeared to take Spassky’s h2 pawn with such a seemingly glib move-gesture in their game 1 at Reykjavik. Of course, I was able to show Paul Anderson and Anthea their missed wins (sometimes multiple) from their lost games because we all suffer from different chess pathologies and abilities. Anthea was funny when she said “I don’t need a chess-coach, I need a psychologist!” hehe.
There’s something more to this trapping, though. Anand deserves a Tony for his “Broadway” performance of walking away from the board, looking at Gelfand as if waiting for him to resign. But really, didn’t the trapper have more to prove?!!
GM Daniel King, add another GM to the list, doesn’t present the move 17…Nc6! (Chess Improver says that Black’s position is hopeless after 18.dxNc6 Qxc6, 19.Bg2 – but what happens after ..Qc7? Black has a rook and two pawns for the two knights, but so where is White’s mate? I’d like to at least see it) Besides, what a computer sees and a human sees can be two different things. Here is the variation I am seeing without an engine, just me calculating (if White doesn’t take on c6):
17…Nc6!, 18.Be2 (else …Qf3 when Black escapes) Nd4+, 19.Kd2 RxBd2+! giving back the exchange, 20.QxR QxRb1 (winning the exchange once again), 21.NxQb1 NxQe2, 22.KxN and Black’s position is easily won.
Also losing for White is 19.QxNd4 cxd, 20.RxQh1 cxNf3. But even more decisive for Black in this line is 19..QxRb1+ followed by 20…c5xQ and Black is up a full rook, plus a pawn.
I did play a game on FICS the other night and lost, but immediately after the game, with no computer analysis required, I calculated that 20.Nxf7 is crushing, whereas 20.Nc6 was really a time-pressure move where I was fishing. I felt happy about that, but it’s not that complex compared to Nezhmetdinov’s games. 😉
Here is the original king-walk line which I was going to play, had spent most of my time on, and had thought that it all calculates out for me. My opponent said after the game that he still gets the Nc2+ in. Yes, but the refutation was tricky and I didn’t see Black’s final move.
11.Qa4 QxQ, 12.NxQ Nc2+, 13.Kd1 and now NxRa1? would be a mistake. Unfortunately I had not seen 13…NxBe3+, 14.fxNe3 Bd7!! and now both my Na4 and Nf2+ (winning the exchange!) are hanging. In the end, I gave 11.Qb3 even less of a look, but it felt more right positionally, which is why I had only looked cursorily at his 11.Nc5 sac reply, didn’t take it as seriously or give it as much of a deep look, and had sorted of blended the lines together in my head.
I still hadn’t calculated this _second_ king-walk (two king-walks in one go!) out correctly in any case, thinking I had an advantage if he had played 14…Nf2+ in the line above, but either continuation doesn’t work.
15.Kd2 NxRh1, 16.NxBf5 Nf2, 17.Ng3 h5!, 18.h3 h4!. So 18.Bxh4 Ne4+, 19.NxN RxB, 20.h3 and Black is +1 because he has the exchange for only a pawn. Line 2, the line I was planning on playing is:
15.Ke1 NxRa1, 16.NxBf5 (why is White not winning here) e6, 17.Nh4 Bb4+, 18.Kf1 Bd2!!, 19.Bd3! Bxe3, 20.Ke2 Bxd4, 21.Ra1xNh1 g6 and Black is up -+ .78, even though two pawns and a rook for two knights is more or less even material.
These king walk lines are too tricky to calculate at G/90, but even worse is that it’s easier to find the bust, 14.Bd7!! (-1.45) -+, which I am rather sure that Isaac would have found OTB, than it is to walk the tight-rope as the trapper.
Last update: I’ve gotten 8 out of 8 correct and my rating on ChessTempo is now 1811 (‘ChessAnalyst’ is my handle there). These are easy, simply need to calmly find the answer. This “Aarrgh!, I’ve got to look at 100 tactics an hour until I am blind and know the answer without thinking!” stuff is nonsense. All this, and my game, tells me is that is much easier to be destructive, by finding a tactical refutation to a plan, than it is to play good chess!
In regards to getting a “good” ChessTempo rating, I did not spend “an hour” on these problems, probably spent under 4 minutes on each one of them. Isaac seemed to think our game was about tactics, but I told him that it was more about calculation. I shouldn’t have put myself into that situation however.
9/9 – 1836, but now I spent 12:29 on this last problem. It wasn’t the two move cheapo, that someone could have blurted out thinking they are special for doing so, it’s the long sideline which I had to calculate, that trading pawns opens up a rook check once you force the king to step in front of it’s rook. 10/10 – 1856. I see the answer in 1 minute, but spend 4:48 because I was scared of an obvious, easily solved sideline. 11/11 – 1875, 2:29. Haha, this one you could win queen for rook straight away, but a knight check wins it for free. This problem is rated 1727. I can only imagine people taking a quick guess and then wanting to pull their hair out. Tactics books puzzles are much, much harder than this.
This is why I don’t agree with school of thought that it is better to see easy problems quickly. Yes, that may be true in time-pressure, but if chess in the modern world weren’t so crack-addicted to fast time-controls in order to placate the lower attention-spans of kids, then this wouldn’t be as much of an issue. Perhaps the part that should scare me more of what Isaac told me is that he is now spending a lot more time (with his coach, no doubt) studying endgames.
12/12 – 1895, 22:30. A five-mover, this one scared me because it was a long variation. Still, I think it is better to get tactics right than to be thinking “I am okay, missed them all, but at least I got my patterns in.” That is such a defeatist attitude. One thing realistic about this is that they give you problems rated close to 150 pts. lower than your current rating. I’d say that that’s about right that nearly 150 rating points can come just from consistency – look at what’s happened to my OTB rating, it’s exactly that, loss of consistency! The reason it’s important to get it right, IMHO, is because calculations are more important than tactics, thus we just crossed over from MDLM to Stoyko!
13/13 – 1910. Went to sleep, woke back up and solved it. Had to find two mate in threes, and realize you had three moves since moving first – hardest one yet.
14/14 – 1934. Eight move king-walk, while avoiding the opponent’s king and cheapo mate. I still had to spend a lot of time after the end of the problem because it could have gone on for yet another eight moves.
BTW, I am going to do a whole series on Nezhmetdinov’s games, perhaps even setup another blog for it.