The Swindler
June 15, 2017 11:35 PM   Subscribe

Any patzer can blunder a winning position into a loss. But turning what looks like certain defeat into chess victory? That takes a swindler.

The Fine Art Of Swindling

Outrageous Swindles

Probably the most famous swindler was US champion Frank Marshall - who apparently didn't like the term "swindle." Levitsky vs. Marshall, the Gold Coin game.

an annotated swindle

As long as mistakes are possible, victory is possible.
posted by the man of twists and turns (18 comments total) 12 users marked this as a favorite
 
Been on both sides of this......I won first place, and $900, in an amateur division once, by swindling some poor kid; it was glorious (except for the part where I got a lost game in 15 moves and had to look for cheap shots).....it really is agonizing to be on the losing side of that kind of thing. I sacrificed an Exchange to eliminate his dark squared Bishop, which gave me an aggressive, but objectively worthless, check from my Queen. He had 2 squares to move his King out of check. One lost, and one left me with nothing. Here is why this was a swindle. The correct move for him was to move his King to a square that felt more dangerous and exposed, and the losing move was to tuck it away in the corner where it seemed more protected, but where he would be checkmated by force. He only thought for a minute, and played the losing move. The whole thing was based on jarring him into playing an emotional move, instead of doing a really small amount of calculating and finding the correct path.

I was so proud of that tournament! 5-0 baby! Then I showed my games to a chess master, and he demonstrated that I was completely lost in at least three of them, and won only because my opponents were even worse than I was. Oh well, the check cleared.

I don't play much tournament chess anymore, but in the late 90s I was super into it. I'd meet people in social settings and they'd be almost audibly gearing up to make fun of that activity. Mentioning the $900 win would always shut them down. Isn't that sad? Since I won money, they were compelled to respect me, but, if I had, say, played in a higher division, done much better than expected, but won no cash, they would feel vindicated in looking at the whole thing as a waste of time. Sometimes I could also get that kind of person to consider the abyss of time and money that, oh, golf represents, and see that it's not particularly extravagant, by those standards, to get obsessed with competitive chess.
posted by thelonius at 3:46 AM on June 16 [20 favorites]


I think the focus is blurry on the definition of the word "swindle"; how it's used generally and, in this case, specifically. But since I'm a new devotee to chess (6 years in), I dig the links. Chess is an incredibly deep "game".
posted by rmmcclay at 5:13 AM on June 16


The difficulty is that in order to play to the best of our abilities, it is important to mind desperately, to love winning as much as life itself, and to hate losing as if it were a form of death: to care much less would detract from the necessary motivation.
This here? This is just flat wrong.

My own game improved by leaps and bounds once I stopped caring about who won and started caring whether or not the game was beautiful. A beautiful defeat inflicted by a skilled opponent is a song to be sung to others.
posted by flabdablet at 5:20 AM on June 16 [9 favorites]


This here? This is just flat wrong.

Agreed. Wanting it too bad makes any kind of performance suffer. You can't keep the emotional distance to see the game clearly if you're more invested in winning than appreciating and feeling passionate about the beauty and art of the technique of the game play. I always play worse if I let myself care too much about winning.
posted by saulgoodman at 5:31 AM on June 16 [2 favorites]


My own game improved by leaps and bounds once I stopped caring about who won and started caring whether or not the game was beautiful. A beautiful defeat inflicted by a skilled opponent is a song to be sung to others.


Can't we check our egos at the door? Chess we can!
posted by neustile at 6:35 AM on June 16 [1 favorite]


I've spent most of my life dancing at the border between being a good hobbyist and a bad "serious" chess player. I still remember one of the very first pieces of advice that I received from a much better player than myself (when I was about 14 and still liked to try to sneak Scholar's Mate variants into games as often as I could).

He said: "A plan that depends on your opponent making a mistake is no plan at all."

These days, I'd generally rather lose a blunder-free game than win one on the back of a blunder. I think. I feel a sense of loss (losing the beauty of the game) when my opponent blunders, but if I'm honest I do still make moves, when I am completely lost, designed to provoke errors.
posted by 256 at 7:21 AM on June 16 [1 favorite]


neustile: Can't we check our egos at the door? Chess we can!

Is that... does that have something to do with anything outside itself, or is it a one-off bit of weirdness?
posted by clawsoon at 7:53 AM on June 16


Metafilter: dancing at the border between being a good hobbyist and a bad "serious" chess player.
posted by sammyo at 7:57 AM on June 16 [1 favorite]


(hmm are these "Metafilter:" collected somewhere?

So what 256 says about using swindler strategy to trick an opponent into a blunder, seems totally valid. As a game of perfect information it's between two people (unless Watson and it'd be deeply cool to trick a computer!)

There's a disturbingly subtle issue, is there a perfect strategy in every game but "playing" the opponent to fool him into making a less optimal move invalid? Don't have the attention span to work through famous games that are "swindles".
posted by sammyo at 8:13 AM on June 16


Is that... does that have something to do with anything outside itself, or is it a one-off bit of weirdness?


It's from the very Canadian movie "Ivory Tower", about "Jazz Chess" -- starring Tiga, Gonzales, Feist and Peaches. trailer.
posted by neustile at 8:18 AM on June 16 [2 favorites]


it'd be deeply cool to trick a computer!

If that's what you want to do, your best bet will be to play positions that leave as many pieces on the board as possible for as long as possible, and play a plan that involves an unstoppable attack hinging on an absurdly expensive sacrifice.

You need the robot to fail to see it coming, and the way to achieve that is to take advantage of the fact that robot strength rests squarely on the sheer speed with which it can analyze the future paths of play, while your own strength depends more on a practised intuitive ability to judge a position's assets and liabilities at a glance without heavy analysis. You want your plan to hide behind a move that the robot will prune from its search tree on the grounds that initial evaluation makes it look too expensive for you to contemplate.

The better the robot, the less of the search tree it will need to prune in order to remain within its time controls, and the more expensive your attack's initial sac will need to be to compensate. Shitty robots can be beaten by sacking a bishop for a pawn. I've only ever beaten one reasonably good robot, and that required a rook-for-pawn swap.

Finding attacks that survive that kind of material loss is hard. But it is indeed oddly satisfying when you do.
posted by flabdablet at 8:40 AM on June 16 [1 favorite]


Better find an old computer.
The original IBM PC was pretty good for that. You could trick it by playing 1.e4.
These days, trying to get past its ply depth is not easy, unless you change it.
Best thing I can think of is a situation where you lure a piece into a position where it can't do anything.

I'd much rather trick a person than a computer. Computers don't get mad.
(Yeah, I still have issues about competition)
posted by MtDewd at 10:02 AM on June 16 [2 favorites]


it'd be deeply cool to trick a computer!

It used to be do-able; it really isn't any longer. They used to exploit the "event horizon" - slowly building an attack whose payoff was beyond the computer's calculations, while probably also feeding it some material so it thinks it's better. A high-level example is Kramnik vs. Junior. This was a humilitaing defeat for the Junior programmers; most top programs were already well immune to this kind of strategy. Kramnik had screwed around at home until he found that he could sneak it past this program, though.

I myself found that you could, playing White against Chessmaster 3000 for Mac, a very weak program by today's standards, let it take a pawn on e5, like in a French structure, and it would then continue capturing pawns all the way down the long diagonal, while you prepared a mating attack. It would fall for this every time; you just had to get the timing right. It's really satisfying the first five or so times!

As for detachment, the first thing I learned when I was hanging out with the above mentioned master was that I just simply did not care about a game as much as he and his peers. If they lost an important game, it ate at them for days - they'd be up all night analyzing it, trying to find what went wrong.
posted by thelonius at 10:06 AM on June 16 [2 favorites]


One of the two highlights of my chess life was beating Chessmaster (Game Boy, 1990) on its highest difficulty exactly one time. The game documentation ranks the engine at ELO 2000, though these things are highly inflated by marketing.

I was down a couple of pawns coming into the late midgame when I saw that I had on opportunity to make a sacrifice that would lead to a series of checks from which I could probably eventually force mate. Sure enough, the sacrifice worked and, about nine moves later, I put the computer into checkmate.

It responded by moving another piece to put me in check. So I took its king. Then it made another move, putting me in "checkmate" and declaring victory. It was rather surreal.

(The other highlight was when I noticed a better solution than the official one to a Toronto Star chess problem, sent in a letter, and got a correction printed in the next Sunday issue.)
posted by 256 at 10:27 AM on June 16 [7 favorites]


Metafilter: does that have something to do with anything outside itself, or is it a one-off bit of weirdness?
posted by hanov3r at 11:45 AM on June 16 [2 favorites]


Then it made another move, putting me in "checkmate" and declaring victory

Classic swindle: just declare victory.
posted by dmh at 11:57 AM on June 16 [2 favorites]


I was down 4 pawns, but caused a fork that won a rook for knight AND allowed me to eat up all but one of his pawns. My King was in opposition, but he slammed his king two squares away then dragged the piece back as I pushed my King into what I thought was the correct square only to find out as I hit the clock that he was movin to where I thought he had been.
A small crowd had gathered (I was 1500 and drawing a 1900), and they all groaned. One old man slapped me on the back and said the same thing happened to him once in 1967.
posted by whatgorilla at 10:24 PM on June 16 [1 favorite]


sammyo: "(hmm are these "Metafilter:" collected somewhere?"

From time to time, yeah.
posted by Chrysostom at 10:31 PM on June 24


« Older Canada: Gender Identity And Gender Expression...   |   Friday night's alright for wrestling (or lots of... Newer »


You are not currently logged in. Log in or create a new account to post comments.