Page 1 of 2
Three swindles
Posted: Sun Jun 03, 2012 1:36 am
by Christopher Kreuzer
Posting three samples here of swindles where I or my opponent managed to salvage a game after being lots of material down. Two are posted without comment to illustrate how to draw games despite being a rook (or more) up in material (there are annotations, but I'm not posting them yet, as I want to see what others think with just the bare moves provided). The first game is a London League match (from the 2011-12 season), and isn't that interesting despite the way I nearly stumble into a loss. The second game is from the 4NCL (also this season) and has a few more fireworks (starting from move 16) and a position where Black's pieces cozy up to the White king and his pawns in a way that someone described as 'intimate' (of the three pawns, two are blockaded, and one is pinned and can only move to allow mate next move). Others just described the position as 'crazy'. The third game is included to try out the annotation and variations feature available here, and features an amazingly lucky swindle. All three games end in a draw, which is probably quite appropriate, really.
1) Drawing a rook up - part 1
EDIT: This game has been re-posted here:
http://www.ecforum.org.uk/viewtopic.php ... 365#p90016
2) Drawing a rook up - part 2
3) Drawing by allowing mate in 7
EDIT: This game has been re-posted here:
http://www.ecforum.org.uk/viewtopic.php?f=2&t=4366
Re: Three swindles
Posted: Sun Jun 03, 2012 8:49 am
by John Hickman
I was going to see if I could find 3 of mine, and had to make a selection. So to cut it down, I've limited it to the month of October 2011.
To contrast' the nouvelle cuisine of the World championship, here's some amateur food fights.
For the entrée, direct your mouse to move 61. The delight of this morsel is that the swindle doesn't actually work, as pointed out to me by one of the amused spectators after the game. Can you find the flaw?
The main course follows swiftly. Direct your electronic cutlery to move 42. A much stronger player than me when shown the game also thought I was completely winning in this position.
An so for dessert, where you can watch me chase the nicest man in chess across the board, miss many wins, and then with
my eye in the prize, whoops
Hmm, maybe I needed some cathartic treatment.
Re: Three swindles
Posted: Sun Jun 03, 2012 9:19 am
by Roger de Coverly
Christopher Kreuzer wrote:
2) Drawing a rook up - part 2
1.e4 c5 2.Nf3 d6 3.d4 cxd4 4.Nxd4 Nf6 5.Nc3 a6 6.Bg5 e6 7.f4 Be7 8.Qf3 h6 9.Bh4 Qc7 10.0-0-0 Nbd7
Bg5 Najdorfs don't really count. Someone being a rook up is part of normal play. Witness the opening of this recent game where Black could go a Rook up at move 21 with Bxg7, but chooses to sacrifice the exchange instead by Rxd4.
At the end 35 .. Ne5 should draw.
Re: Three swindles
Posted: Sun Jun 03, 2012 1:20 pm
by Christopher Kreuzer
Roger de Coverly wrote:Christopher Kreuzer wrote:
2) Drawing a rook up - part 2
1.e4 c5 2.Nf3 d6 3.d4 cxd4 4.Nxd4 Nf6 5.Nc3 a6 6.Bg5 e6 7.f4 Be7 8.Qf3 h6 9.Bh4 Qc7 10.0-0-0 Nbd7
Bg5 Najdorfs don't really count. Someone being a rook up is part of normal play.
But to be fair, the position I got wasn't normal play, it was just completely winning for me. You cut the opening short at move 10. What I really want to find out is whether I went wrong on the next move: 11.Be2 b5. This early b5 (before castling, for instance) allowed my opponent to take on f6 with the bishop, and all the recaptures have downsides. I still haven't managed to find someone able to tell me whether this line is covered in theory or not.
The recapture I chose (with the knight) led (eventually) to the rather strange position in the game. My opponent should have captured on g7 with the knight and either taken the perpetual with Ne6+ and Ng7+, or after returning the knight to e6 with check, carried on with the capture on c3 and been a pawn up (though the position is probably still drawn). The mistake was my opponent missing 17...Ba3, after which White is busted.
This is the position, with White to play:
There are only two moves that keep White in the game. If you haven't played through the game yet, it is worth trying to see if you can work out what they are. Sadly, having the Black pieces, I then proceeded to play like a patzer for the rest of the game, declining the chance to go a rook and two pieces up, neglecting to recapture a piece and so being 'only' a rook up at the point where I threw away a piece to a cheapo, and then gave back the exchange to end up only a pawn up. Very annoying. If I'd been less material up, I'd probably have managed to lose somehow...
Re: Three swindles
Posted: Sun Jun 03, 2012 1:59 pm
by Roger de Coverly
Christopher Kreuzer wrote: I still haven't managed to find someone able to tell me whether this line is covered in theory or not.
It depends on what you mean by theory, but in the sense of "has it been played before", there are several examples, all with Nxg7+. Earlier Be2 had been seen before and the idea of the e5 prod has been part of Najdorf theory and background since the 1950s. For what it's worth, here's how Tal handled it.
The idea was also seen in this famous example, which would be known to all readers of 60 Memorable Games. Your game is the same, except Black has played h6 as well. In pre database days, GMs recommended that to have a well versed knowledge, you needed to know around a thousand games. All the Fischer games from MSMG would have been among them. The annotations are useful as well, if you're ever stuck for a comment. If I recall correctly move 13 had the comment "So I chopped it off".
(edit) A more recent example
Nerverov's Qc4 may well be a strong novelty. I think I was present when Jack showed the game in the Commentary Room. (/edit)
Re: Three swindles
Posted: Sun Jun 03, 2012 2:37 pm
by AustinElliott
Yes, remember the Keres-Fischer game from MSMG. I think Fischer says that he reeckoned that in his pre-game analysis Keres had possibly overlooked the strength of the 20.. h5 and 21 ..Rh6 idea to get the Black R into play, and just looked at ...g6 and ..Kg7. Interesting since ...h5 strikes me as the sort of move a computer would find easily.
Re: Three swindles
Posted: Sun Jun 03, 2012 2:48 pm
by Geoff Chandler
Hi Chris.
Three games!
Posting one at a time I think would have been better.
So now we are going to have to say 'in game one you could have played....'
Everyone has to scroll back to see what game what game we are talking about.
Someone will forget to add in the game number and we will all wonder
on earth he is talking about.
Also, one (or possibly even two) games will be ignored as people get into
the juicy Najdorf game. (as is happening.)
Roger is correct. The Najdorf has a brilliant habit of throwing up positions
where all normal assumptions about the game are of no good at all.
If you are a Rook up in the Najdorf and the game is still going on then
there is strong chance you are losing. One can be forgiven for thinking
some of the Najdorf positions came from a Fishcer Random Chess game.
I'll have a look at all three but slight hangover and other things
(7 sold e-bay china items to wrap up.) should see me adding in my 2p worth mid-week.
Re: Three swindles
Posted: Sun Jun 03, 2012 6:40 pm
by Christopher Kreuzer
Geoff Chandler wrote:Hi Chris.
Three games!
Posting one at a time I think would have been better.
Possibly, but the first one is just an appetiser, and the third one is more something to do with chess psychology and less to do with actual chess variations (though speculating as to how such things happen on a chess board is interesting in itself).
Geoff Chandler wrote:So now we are going to have to say 'in game one you could have played....'
Everyone has to scroll back to see what game what game we are talking about.
Someone will forget to add in the game number and we will all wonder
on earth he is talking about.
Maybe, though I was hoping people would learn how to add annotations to the moves and would add their own annotation to the previous lots (adding their initials or something), and then this whole group annotation thing might work, but I don't think it will work that well.
Geoff Chandler wrote:
Also, one (or possibly even two) games will be ignored as people get into
the juicy Najdorf game. (as is happening.)
As I said, it was that game that I was really looking for commentary on.
Geoff Chandler wrote:Roger is correct. The Najdorf has a brilliant habit of throwing up positions
where all normal assumptions about the game are of no good at all.
If you are a Rook up in the Najdorf and the game is still going on then
there is strong chance you are losing. One can be forgiven for thinking
some of the Najdorf positions came from a Fishcer Random Chess game.
I still disagree about that in the position from my game, as I was winning that one (it wasn't a case of White having any counterplay or anything, just me messing things up), but I agree that some of those games Roger has posted do show how random things can get.
Geoff Chandler wrote:I'll have a look at all three but slight hangover and other things
(7 sold e-bay china items to wrap up.) should see me adding in my 2p worth mid-week.
Thanks. What Roger has posted is really good so far, and I'm going to add a bit to what he said there. Do have a look at John Hickman's games as well, some nice tactics there.
Re: Three swindles
Posted: Sun Jun 03, 2012 6:59 pm
by Christopher Kreuzer
Roger de Coverly wrote:Christopher Kreuzer wrote: I still haven't managed to find someone able to tell me whether this line is covered in theory or not.
It depends on what you mean by theory, but in the sense of "has it been played before", there are several examples, all with Nxg7+.
Many thanks for these example games, Roger, much appreciated. I had looked up the Keres-Fischer game (a very famous game, sometimes called 'Bled to Death'!), prompted by an offhand comment from my opponent in the post-mortem that he had considered this line in our game, but realised that the fact that Black had played h6 in my game was making a difference, and it was something like the Kliavinsh-Tal game that I was looking for. That is an absolutely crazy game, and my head is still spinning after playing through that.
It is an interesting comment you make, saying: "It depends on what you mean by theory". What I mean by theory is anything that has been played before, even if refuted by later theory. I suppose what you mean by theory is something that is still playable and hasn't been discarded or been superseded by later lines. When you say there are several examples, all with Nxg7+, would you be able to list those and/or say what the result of the games was? I guess you are also saying that the position after White's Nxe6 and Black's Qxc3 has been seen before, but that in previous games White has always avoided Qxg7, presumably because the White players see that Ba3 is the refutation? So are you saying that the position after Qxg7 and Ba3 is completely new? If it is, that would be nice to know (though I guess you can never be 100% about this sort of thing).
Being able to mention the Tal game from 1959, and some of the other games that went down these lines, is what I was looking for to add to my annotations, as I don't have databases to be able to do that sort of annotation. Would it be OK if I included reference to those three games in my annotations? I could credit you with pointing them out if that is the right way to do it.
One thing I am puzzled about though: in all three games, 13.exf6 was played. What happens in all three games if 13.Qg3 is played instead, and why wasn't it? Is the queen sacrifice the natural continuation, more natural than 13.Qg3, or is there a variation that the White players want to avoid in the 13.Qg3 lines?
Jack's game is interesting as well. I wonder if he would be willing to say how much of the theory he was aware of and/or remembered at the time of that game, and whether he was aware of both the Tal and Fischer games you mention?
Re: Three swindles
Posted: Sun Jun 03, 2012 8:18 pm
by Roger de Coverly
Christopher Kreuzer wrote: What I mean by theory is anything that has been played before, even if refuted by later theory.
One more stringent definition of theory is that it's a game that's been published somewhere, probably in a book on openings. Depending on the author, the recommendation may or may not be correct. So Be2 in the Najdorf is theory from having been discussed in the Keres game. I wouldn't know whether any authors got as far as Nxg7.
You can follow theory without the game having been played before, by just using the usual ideas from the tabiyas.
To return to the actual game, here's one of the Nxg7 games
White declines to offer his queen as in the Keres-Fischer or Rudd-Khenkin games and play looks plausibly forced provided you spot Qxg7 as an error.
Here's the same thing without h6. Black doesn't have .. Bg5+, so there is a minor difference. Although whether it matters , I don't know because Nxc3 was preferred in the other games.
and for good measure, a game without h6 but with the Qxg7 blunder.
Perhaps we can say then, that the .. Ba3 combination has been known, if not widely, for nearly fifty years. There may have been more games, it would have been only four years since the high profile example of Keres-Fischer, so players would have been willing to explore what happened if you don't sacrifice the Queen.
So assuming White avoids the Qxg7 blunder, whether players are prepared to play to go into the line will depend on what they think of the resulting ending after Nxe6 and Qxc3. Black is slightly worse perhaps.
Re: Three swindles
Posted: Sun Jun 03, 2012 8:25 pm
by Dan O'Dowd
Roger de Coverly wrote:In pre database days, GMs recommended that to have a well versed knowledge, you needed to know around a thousand games.
Is that meant to mean one thousand in total, or one thousand from games played in that opening?
I'm just in the process of looking at 2400+ games in various openings I play.
Re: Three swindles
Posted: Sun Jun 03, 2012 8:25 pm
by Daniel Young
Here's a Rook-up disaster I managed to craft today in Gatwick - as I said in another thread, I have a horrible habit of choking in positions against stronger players (white is graded 32 higher than me) where it ought to be impossible to lose:
I had 20 minutes left when I played my 39th move.
Re: Three swindles
Posted: Sun Jun 03, 2012 8:32 pm
by Roger de Coverly
Dan O'Dowd wrote:
Is that meant to mean one thousand in total,
It was one thousand games in total, so you know a handful of key games in just about every opening. This was "GM advice to amateurs though", so may not be the whole story.
Re: Three swindles
Posted: Sun Jun 03, 2012 8:44 pm
by Roger de Coverly
Pragmatically you should trust your technique and play 37 .. Nxd6 with Re6 available if he takes with the Bishop. Earlier activating the Knight by 33 .. Nb7 would have activated the Rook as well. 39 .. Nxd6 would be good also. So he plays 40 Bxd6 Rxc7 41 Bxc7 . The ending the exchange ahead is simple enough, you just stop his one trick of b5 by parking the Rook on d5 and then win with the King and passed f pawn.
Re: Three swindles
Posted: Sun Jun 03, 2012 8:55 pm
by Daniel Young
Roger de Coverly wrote:
Pragmatically you should trust your technique and play 37 .. Nxd6 with Re6 available if he takes with the Bishop. Earlier activating the Knight by 33 .. Nb7 would have activated the Rook as well. 39 .. Nxd6 would be good also. So he plays 40 Bxd6 Rxc7 41 Bxc7 . The ending the exchange ahead is simple enough, you just stop his one trick of b5 by parking the Rook on d5 and then win with the King and passed f pawn.
Of course, you're quite right, and thanks for the comments. The tragedy is amplified in my view by the fact that I saw how straightforward 37/39... Nxd6 would be, but I wanted to be as exact as possible, with spectacular consequences; I just missed 41. Bc5, and the earlier mate threat was too tempting. I was wary of the psychological boost white might get from me giving back even a little material, given that he's even decided to play on.