Never too late to lose
The following is a section from a recent issue of White Belt, a weekly training e-mail newsletter from ChessNinja.com. Annotated reader games and questions are in just about every issue. Subscribers get a free six months at the best place on the internet to play chess, Playchess.com.
A rich and interesting reader game this week, and even better it comes with excellent and entertaining notes from our hero with the white pieces. With Andrew on the case we feel like taking the week off, but we hope we managed to contribute something!
The opening is a fine example of why studying openings can get you into more trouble than it's worth. Even if Andrew had known the next "book" move on move nine, he probably would still have gotten into some trouble in the opening. You can never know enough just from memorization (or his innovative method of using book reviews!). When you start making moves you don't really understand you are just begging for a sock in the jaw.
Mig's comments are marked, all others are by Andrew. Replay the annotated game online.
Here's an extremely frustrating game which teaches a lot of good lessons: 1) Tactics win and lose games. 2) Don't give up until it's over. 3) Never relax; stay focused on the board until you shake hands. Also, it repeatedly raises the question: how should you play when you've got a lost position? What kinds of goals and plans can give you a chance to save it? I offer the game and my analysis as penance for the howling blunders I made. [Time control: 40/100, 25/60 thereafter, no sudden death.]
1.d4 Nf6 2.c4 e6 [2...c5 3.d5 e6 4.Nc3 exd5 5.cxd5
d6 6.e4 g6 7.f4 Bg7 8.Bb5+ Nfd7 9.Nf3 a6 10.Ba4 b5 11.Bc2 0-0 12.0-0 Re8
13.f5 Ne5 14.fxg6 fxg6 15.Nxe5 Bxe5 16.Qf3 Ra7 17.Bd2 Schmitz,M-Wolf,S/Germany
1993/EXT 97-B/0-1 (56)] 3.Nc3 c5 (D1)
Okay, starting around here I began to review what I knew about the Benoni, all of which came from reading a book review a few months before! I knew the main line was the Taimanov Benoni, a feared attacking line with an early f4 and Bb5+, which black is supposed to avoid by only playing the Benoni when white's already played Nf3 (thus killing the f4 possibility).
Also, I remembered that black's best move after Bb5+ is supposed to be ...Nfd7, but I hadn't a clue why. And if ...Nfd7 gives black hope, the attack must be "feared" but not automatically winning, at least not at my level. Lastly, I had a vague sense that the attack would have to feature either e5 or f5 to break through and get at black's king.
That's it for my mental opening book (at the time). Needless to say, I couldn't remember a single game in this line, so I'm on my own. Let's see what I can learn over the board...
Mig: A little knowledge is a dangerous thing, as is shown very clearly in the result of your foray into Benoni theory. You know just enough to get yourself into a sharp line that your opponent knows better and that you don't understand!
We always advocate taking the bull by the horns and playing the best moves you can see. In the opening it can mean playing the best moves you are sure you understand. For example, here you could play 4.Nf3 and avoid all the sharp stuff after d5.
4.d5 exd5 5.cxd5 d6 6.e4 g6 7.f4 Bg7 8.Bb5+ Nfd7 (D2)
And here we are. (I've used about 15 minutes thinking about the Taimanov and looking at possible deviations, while my opponent is blitzing his moves out.) I "know" that ...Nfd7 is the book move, but why? Well, it gets the knight off f6, where it could be hit by the e5 advance, and opens up the Bg7 to protect the e5 square. It allows ...f6 which would clamp down on e5, though that looks like an ugly move to play.
On the down side, ... Nfd7 means black's queenside pieces don't look like they're going anywhere soon. Okay, my plan will be to castle and marshal my pieces to support pushing the e- and/or f-pawns. Psychological problem: if my opponent weren't higher rated and blitzing out his opening moves I'd feel a lot better about this plan...
Mig: Very good explanation for why this odd-looking move is best. Blocking the check with other moves allows e5 and Black can get into trouble very quickly. Now Black is covering the key e5 square.
9.Nf3?! Mig: This is what we meant about knowing enough to get into trouble. If you have some experience in the Benoni you know that preventing/delaying Black's queenside expansion is crucial. Without playing a4 Black is allowed to gain time with ..a6 to expand with ..b5.
The other possibility for White is simply to retreat the bishop with Bd3. It's a gain of a tempo since Black will have to move the d7 knight again to unwind his pieces.
[After the game, my opponent suggested 9.a4 to prevent the queenside pawn expansion ...a6 and ...b5. Logical, but it never occurred to me; I was too fixated on trying to push the e- and f-pawns. 9.a4! ]
9...a6 Hmm. Now ...Nfd7 is looking pretty good. If I exchange on d7, he can recapture with the knight and have a nice grip on e5. But if I don't exchange, where should my bishop go? ...b5 and maybe ...c4 are coming, so I picked c2 via a4, to defend e4 and support an f5 advance.
Maybe c2 via d3 would be better, since if he doesn't play ...c4 my bishop is on the same diagonal as c2 and I've gained a tempo. Should I expect ...c4? It's committal, making a later ...b4 more difficult perhaps. So maybe ...c4 is a bad idea for black and I should have played Bd3 instead of Ba4.
Mig: Bad moves follow bad moves. Now Black gets TWO free moves for queenside expansion. Black has basically equalized in tempo now. [10.Bd3]
10...b5 11.Bc2 0-0 12.0-0 Re8 13.f5? A trade-off: he owns e5, but my pawn advance has made contact and my pieces look poised to move against his kingside. His queenside pieces are still in bed.
Mig: Positionally dubious, but it's not an easy position for White. How are your pieces going to move against his kingside after he drops a knight on e5 for all eternity? Your e4 pawn is going to be stuck there blocking in your c3 knight and c2 bishop. White might get some play on the newly opening f-file but even f7 will be well protected by an e5 knight. White needs to focus on the usual plan of e5, or at least using his pawn center to limit the black forces.
The problem is that now that Black has so many extra tempi, he is threatening to play ..b4 and then ..f5 himself to break up the center. This is a very sharp opening and giving up two moves puts White at serious risk. It's interesting to see that computers like the White position because of the space advantage, but there is no simple strategic plan for White to make progress after losing so much time.
13...Ne5 14.Bf4 I wasn't sure how to proceed, and spent 12 minutes pondering what to do with my pieces. I finally decided to aim for an exchange of the dark-square bishops, hopefully creating weak squares around his king in the process. My first thought was to play Bg5, to gain a tempo on his queen, or make him play Bf6 Bxf6 Qxf6 with possible tactics due to my rook facing his queen, or ...f6 which I figured weakened his kingside (though it could actually help by enabling Ra7-g7 bringing pieces to the kingside fast), but then I spotted that Bg5 just loses a piece to Nxf3+, so I played Bf4, which has the benefit of also threatening to remove his well-posted knight at some point.
[14.fxg6 Mig: This
is a better attempt to take advantage of Black's poor development for
a quick kingside attack. Letting a knight stay on e5 doesn't work. 14...hxg6
15.Bg5 Qb6 16.Nxe5 Bxe5 17.Qf3] 14...Nbd7 (D4)
15.Qc1?! Too slow. White is in serious trouble strategically so he had better get moving with an attack before Black continues on the queenside with his pawn majority.
15...Bb7?! Mig: Development for development's sake. The bishop doesn't have much of a future on this diagonal and the c8-h3 diagonal is where the action is.
[15...b4 Mig: I have no idea why Black never plays this typical move. Queenside expansion is Black's entire concept in these positions.; 15...Nxf3+ 16.Rxf3 Ne5]
16.fxg6 Why now, instead of following my plan of Bh6? I honestly can't remember. I think I might have been concerned about him playing ...b4, Nc3 moves, then ...gxf5 with the idea that exf5 Bxd5 wins a pawn and if Nc3-e2, Bxd5 would be possible even without exf5 due to the pin on the e-pawn. I certainly didn't calculate these lines carefully, I just figured I intended to capture on g6 at some point, it might as well be now so I don't have to worry about these sorts of problems.
Mig: If Black has anything to be certain of it's not capturing on f5 and freeing both your c3 knight and c2 bishop! Capturing on g6 isn't necessarily bad, but that sort of "when in doubt, release the tension" is bad thought processes. Don't be afraid to keep the tension in a position, as long as breaking it is bad for the other guy. [16.Bh6]
16...hxg6 17.Bg5 Bf6?! (D5)
Mig: Too helpful. This is a mighty bishop, why trade it off? Now White uses the time to build an initiative on the f-file. [17...Qb6]
18.Bxf6 Qxf6 19.Ng5 Qg7 20.Qf4 Re7 21.Rf2 Rae8 22.Raf1 Bc8 Mig: Just when things were finally going White's way he throws it all away in one move.
23.Bd1?? Lesson #1. Tactics, tactics, tactics... Dan Heisman over at ChessCafe's Novice Nook talks about 'Hope Chess', where you don't bother to think about your opponent's possible responses, but just hope you can meet whatever he throws at you. Here's a good example. Bizarrely enough, my goal was to play Bg4, exchange one of the knights, and then... I have no idea what! I had been bringing my pieces to the kingside in a nice, threatening build-up, but I was at a loss as to how to continue. I wanted to get my Bc2 or Nc3 over to the kingside (maybe Nc3-b1-e2-f3xe5 or Nc3-d1-e3-g4, but the knight's journey seems very long), so I mindlessly shifted the bishop. Sigh. [>=23.Qh4+/=]
23...Nd3-+ 24.Qxd6 Nxf2 25.Rxf2 Qe5 26.Qxe5 Nxe5 (D6)
Time to regroup, but it's tough to orient myself here. I'm (only!) down the exchange for a pawn, but psychologically I feel totally lost. We each have about 20 minutes to get to move 40, so the clock's not going to help me, and he's still higher rated than I am. His knight is on a dominant square, his two rooks are eyeing my e4 pawn (both of my knights can be kicked away from its defense), he has a potential passer on the queenside, and most of all, I've just played a horrible blunder to wreck what I thought was a fairly nice position.
So what do I have? I have a passed d-pawn, but I don't see it going anywhere soon, especially since the e-pawn will probably be gone in a few moves. I have four pieces left, so I might be able to create some confusion, but I don't really see how because there aren't any clear targets (accessible, unguarded pieces or pawns, an open king, anything would help). Okay, fine, my plan will be to try to protect the e4 pawn, avoid piece exchanges, and try to keep my pieces active to stir up trouble while I wait for him to give me some targets.
Mig: A fine breakdown. White wants to avoid piece exchanges and try to liquidate pawns as much as possible. If Black gets organized he will quickly start pushing his queenside pawns and win rather easily. [26...Rxe5? 27.Nxf7 R5e7 28.Nd6+/-]
27.Bb3?! Sets up potential tactics along the a2-g8 diagonal (if he kicks my knight with f6 sometime down the road).
Mig: It looks more like it encourages the ..c4 advance with a gain of tempo, losing a move and space.
27...c4 28.Bc2 Rd8 Mig: Keeping an eye on the d6 pawn.
29.Rf4 [29.Ne2 Striving to activate all his pieces. Being the exchange down isn't the end of the world, especially when the rooks don't have any open lines.]
29...Kg7 30.Rh4 Rh8 31.Rf4? (D7)
Mig: White can't keep giving up squares and hope to survive. This loses an exchange directly, but the Rh4 and back again idea was going to be bad no matter what.
31...f6 Mig: Embarrassing the knight.
32.Nf3 Mig: Boxing in his own rook, but there weren't better alternatives. [32.Nh3 Bxh3 33.gxh3 Rxh3; 32.Ne6+ Bxe6 33.dxe6 Rxe6]
32...g5 Another missed tactic, another exchange gone! Oddly, I felt less bad about this one, since the tactic wasn't blindingly obvious. I actually don't know which move to blame, since I had quickly rejected the other knight retreats, followed by BxN and another pawn lost, as being hopeless. I didn't spot the fact that my rook had no retreat ('Hope Chess' again) but at least it isn't a stupid knight fork! Okay, assess the position: I'll be down two exchanges, he'll have two weak pawns on the f-file, I'll still have a d-pawn passer, I'm totally lost...
Options: a) resign now and be home in bed by midnight, or b) play it out to the time control hoping for a blunder and then resign with a clear conscience. Oh, hell, I only have 7 minutes left, he only has 17, I'll still be home by 12: 30, so I'll go with option b). Oddly enough, it was psychologically easier to play on being convinced I was going to lose soon, as opposed to when I'm convinced I'm going to lose way down the road. Lesson #2: don't give up until it's over. Secret option c) assume I can fight my way back into the game, focus like a laser, and find the toughest defense. I don't think this occurred to me at the time.
Mig: Fatalism is nothing to be ashamed of, and you did a good job of fighting through it. You just have to keep making good moves and looking for chances and hoping for a mistake. And mutual time trouble never hurts either! And since the rooks don't have open lines or clear targets yet, it's not so simple. And knights are typically dangerous in time trouble because of the difficulty human brains have with visualizing knight moves.
33.Nxe5 gxf4 34.Nc6 Ree8 35.Nd4 Kf7 (Now we each have 6 minutes left...) 36.Nce2 f5!
Mig: The rooks need open lines. 37.Nxf4? Mig: Why not grab a pawn? [>=37.exf5 f3 38.gxf3-+]
37...fxe4 38.h3 e3? (D8)
How about that - he can miss tactics, too! (He thought for 2 minutes before playing this move, leaving each of us with 1 minute left.)
39.Bg6+ Kf6 40.Bxe8 Rxe8 And now we each get another hour to look at the position. Augghh!!! Worst of both worlds: I'm still lost, of course, but I can't possibly resign this position, so I'm going to be here all night! And then I'm going to lose! What a messed up head I've got! I'm (only!) down one exchange, I have three passed pawns, two centrally placed knights, and most important, he just made a tactical mistake.
This gives me a slight psychological boost, since it demonstrates that he's tired too, and vulnerable to the kind of goofs I'll need him to make if I'm going to survive this. Also, perhaps he's now kicking himself for letting his dead-won position evaporate into just a won position. Okay, time for a grand plan. How can I possibly get to a draw from here?
A) it might be possible to set up a blockade/ fortress across the entire board, keep his king and pieces out and give him no targets - well, probably not, but trying to might slow him down; B) I can exchange off bishop and knight, and all the pawns (hard to imagine, but not impossible, since he might be distracted by my two kingside passers), and then R vs N is drawn; and C) offer a draw and see if he's more tired than he looks! I'll try all three.
Mig: All right, a plan! And having two passed pawns can't hurt either. Anything that MAKES YOUR OPPONENT WORK is great. Don't make it easy, make him think and that means he can make mistakes.
41.g3?! (Played with a draw offer, which he declined in a friendly, and slightly chagrined, fashion.)
Mig: And hanging a pawn. You have to keep a closer eye on the black e-pawn!
41...Ke5 [41...Bxh3! 42.Nxh3? e2] 42.Nde2 [42.Nf3+ Ke4 43.Ng5+ Kf5 44.Nf3-+]
42...Bb7 43.Kg2 Bxd5+ 44.Nxd5 Kxd5 45.Kf3 b4 Mig: Now if Black had played this 20 moves ago...
Mig: Yes! Get him worried, make him think. Passive defense isn't going to work against that army of pawns. White should try anything he can to make Black make decisions.
46...a5 [>=46...c3 47.bxc3 bxc3 48.Nxc3+ Kd4 49.Ne2+ Kd3 50.Nc1+ Kc2 51.Nb3-+]
47.g4 a4 48.h5 I spent 13 minutes evaluating ...a3, ...b3, and ...c3, and checking to see if 48.a3 or 48.b3 would hold him up at all. I came to the conclusion that he's winning, since my knight is terribly over-worked trying to hold back the e-pawn and the queenside pawns, but he has some options that make his life harder. 48.a3 and 48.b3 are pointless, so I might as well push my passers.
[48.a3? c3 49.bxc3 b3 is hopeless; 48.b3? cxb3 49.axb3 axb3 is also hopeless; 48.g5 a3 49.Nf4+ Kc6 50.bxa3-+]
48...c3 He spent only 1 minute thinking about this (he had 48 minutes left), and didn't choose the line I was most scared of.
Mig: ..a3 is definitely quicker because the knight cannot easily control the b1 square, unlike the c1 square.
[48...a3! I figured this was best since 49.bxa3 (49.b3 cxb3 is even worse) 49...b3 is hopeless 50.axb3 cxb3 51.g5 b2 52.h6 Kc4; 48...b3?! 49.a3 and I've got things covered for the moment]
49.b3 [49.Nxc3+ bxc3] 49...axb3 [>=49...c2 50.h6 Rf8+ 51.Kg3 axb3 52.axb3-+]
50.axb3 c2 and again I've got things covered for the moment. I'd seen this far earlier, making me think 48...a3 was the better choice. I wasn't sure how he was going to try to proceed.
[51.h6 Rf8+ 52.Kg3-+]
51...Ke5?! Aha! This revealed a key tactical idea to me. I can't play Kxe3 because of Kf5+ Kd2 RxN KxR c2=Q. But he can't play Kf5 due to Nd4+, so to get his king over to my pawns he has to play Kd6-Ke7-Kf7.
[51...Rf8+ Fritz pointed out this win; I still can't play Kxe3 obviously, and black's king (or rook) can invade faster than my pawns can advance.
Mig: Black is too cautious and decides to try and "win without work" by taking the white pawns before trying to break through with his own. This is sound philosophically, but unsound in calculation! Black seems oblivious to the fact that a knight can draw against a rook if the pawns are off the board. 52.Kg3 (52.Kxe3? Re8+ 53.Kd2 Rxe2+! 54.Kxe2 c1Q) 52...Ke4 53.g6 Kd3 54.g7 Rg8 55.Nc1+ Kd2]
52.h6 Kf5? Yessss!!!!!! I have the draw! I get his two pawns, he collects mine and his king is too far away to stop me from snagging the b4 pawn! The last two hours of work were worth it!!! I have only 9 minutes to reach move 65, but that shouldn't be too hard.
Mig: Yes! This is a good illustration of how losing can focus the mind and winning can cloud it. When you are in trouble you concentrate on finding any desperate try, any saving material imbalance or theoretical draw. When you are winning you only look at your own plans and ignore those of your opponent. [>=52...Rd8 53.h7 Rh8-+ 54.g6 Kf6 55.Kxe3 Rd8]
53.Nd4+ Mig: Now White wins the c-pawn with tempo.
53...Kxg5 54.Nxc2 (D11)
54...Kxh6?? Mig: Only this is the blunder that loses the win. Why give up the e-pawn so quickly? Black can reach the same material balance with a much more active rook and still have a winning position. With work, but holding the b-pawn easily.
[>=54...e2 55.h7 Kf6 56.h8Q+ Rxh8 57.Kxe2 Rh4-+ Mig: Now Black holds the b-pawn and can work for a win. The black pawn is far enough advanced that the rook and king can push in and eventually win, although it is far from trivial.
58.Ne1 Ke6 59.Kd2 Kd5 60.Nd3 Kc6 61.Ke3 Kb5 62.Nf2 Rh6 63.Kd4 Rd6+ 64.Ke3 Rd7 65.Nd3 Rc7 66.Kd2 Rc3 67.Nc1 Rc7 68.Ne2 Rf7 69.Nd4+ Kc5 70.Ne2 Rf3 71.Nc1 Kd4 72.Ne2+ Ke4 73.Nc1 Rf2+ 74.Kd1 Ke3 75.Ke1 Rc2 76.Kd1 Rh2!]
55.Nxe3= And there it is - I was absolutely positive that this was drawn. I gave a sigh of relief, relaxed, and ...
In the second between releasing the piece and hitting the clock I knew this lost; my mind had flickered over Nf5+ then Nd4 (but why bring his king closer?) and Nc4, and had even flickered over RxN. I just hallucinated that my king was close enough to stop the pawn.
This is the most aggravating blunder I can recall for three reasons: 1) even if it didn't lose instantly, it doesn't get me any closer to capturing his pawn with either my knight or king, so what's the point of Nc4?, 2) after struggling for a 2 hours in the hope that he'd blunder away the win, to have him do so and then blunder it right back to him is awful, and most importantly, 3) this is exactly the sort of simplified ending where I usually do well. I enjoy studying and playing pawn endgames, and I've gotten a lot of satisfaction from heading into seemingly drawish endgames and coming out on top.
To miss the knight fork on move 23 was pathetic, but to hallucinate that I'm okay after RxN was just a horrendously miserable blow to my pride. Sigh.
Mig: We can sum that up with AAAAAAUUUUUGGGHHHH! It is not an entirely illogical move because it blocks the threat of ..Rc3+, winning the b-pawn.
[56.Nd5 and the draw is obvious 56...Kg5 57.Ke4 Rb8 58.Kd4 Kf5 59.Kc5; 56.Ke4 Rc3 57.Nd5 Rxb3 58.Kd4 Rb1 59.Kc4 b3 60.Nc3 Rb2 61.Na4 Rb1 62.Nc3 Rb2 63.Na4 Rd2 64.Kxb3=
Mig: I can think of one thing maybe worse than what happened in the game. That would be getting to this theoretical draw and contriving to lose it! It IS possible to lose this endgame if you let your knight wander too far from the king or if you let your king get stuck in the corner. Next week we'll look at a few tricks for both sides to be aware of.]
56...Rxc4! 0-1 and White resigns.
Lesson #3: never relax until you're shaking hands. So to conclude: I need to study tactics and visualization more, work on my game-time thought processes (no more "hope chess") and emotional control (don't despair, don't relax), and then, maybe, go learn something about the Taimanov Benoni.
Mig: Take the time you are thinking about working on the Benoni and spend it on more tactics and visualization and you'll be on the right track! The 15 minutes you spend worrying about the opening, which didn't turn out well anyway, would have come in very handy during the tactical complications, no? Anyway, a great battle and thanks for sharing it and your instructive thoughts.
[56...Rxc4 57.Ke3 Rc3+ 58.Kd4 Rxb3 59.Kc4 Rb1 60.Kc5 b3
61.Kc4 Kg5 62.Kc3 Kf4 63.Kb4 b2 64.Ka3 Ke4 65.Ka2 Kd3 66.Kb3 Rc1 67.Kxb2
Rc3 68.Kb1 Rc2 69.Ka1 Kc3 70.Kb1 Kb3 71.Ka1 Rc1#]
The preceding was a section from a recent issue of White Belt, a training e-mail newsletter from Mig's ChessNinja.com.