Re: Two articles pondering the issues with Chess: Stale openings, declining prize pool, and issues with draws.
- From: Quadibloc <jsavard@xxxxxxxxx>
- Date: Fri, 18 Apr 2008 02:51:22 -0700 (PDT)
On Apr 7, 4:44 pm, "David Kane" <davidek...@xxxxxxxxxxx> wrote:
For example, instead of 29% pseudo chess, and 71% White
wins/Black draw, which we get from 1867-scoring (draw=.5)
, what if we had a tournament format that produced 0% pseudo-chess,
and 100% of contests starting with the objective of
White win/Black draw? Or what about, God forbid,
0% pseudo-chess, and 50% White win/Black draw, 50%
White win/Black win objectives?
That we don't know exactly what tournament format/alternate
scoring/prize fund distribution would produce the above situations
is really just a detail. The key thing is to understand that chess is
killing itself by rewarding strategies that devalue the individual
game, and that create an anti-competitive environment. Those
are just fundamentally inconsistent with creating a dramatic
event, and Sophia rules do nothing to address the problem.
This is a good statement of the problem.
I favor using an alternate scoring, rather than a different tournament
format or a different prize fund distribution as a suitable measure.
1) It is simpler to describe.
2) It can be made applicable to matches as well as tournaments.
3) It does not create a situation where players have two conflicting
objectives to deal with.
What kind of alternate scoring, though, would change this -
particularly if one of the fundamental goals is to affect all chess
games equally, so that 1/3 - 1/3 for a draw, which makes the total
point value of a game nonconstant, therefore affecting tournaments,
but not matches, is excluded as an alternative?
I emphasize applicability to matches for two reasons:
Players shifting gears between two habits is difficult for them, and
likely will mean they will stick with the older habits if they work in
one of the contexts they face, slowing the results.
The World Championship will create more interest than any tournament,
and so if we are to revitalize public/spectator interest in Chess, we
must have more interesting Chess *there*.
Changing the rules of Chess trivially gets rid of the opening book.
How to do so in a way that promotes more aggressive play, though, is
less obvious.
How to change the scoring in Chess to promote aggressive play isn't
obvious either, though. If aggressive play means sticking your neck
out, making yourself more likely to lose, then you will get both
players playing defensively, accumulating, at best, small positional
advantages, and going for the win only if they have the rare good
fortune of having enough advantage for that to be sound.
So, for a long time, I was stumped. I didn't have a good idea for
this, and it didn't seem that anyone else did either. But then I heard
of how _komidashi_ in Go solved a similar problem.
The problem is that komidashi works because of how Go is scored. You
have a board with 361 spaces, and you count up how many are controlled
by each player to determine who wins. So it's easy to say the first
player (Black in Go) has to win by 7 or more stones to win, and
otherwise he loses.
Checkmate is all or nothing. And the problem in Go was different than
the one in Chess: Chess is usually a draw; in Go, the first player
wins (by three spaces or so) when both players play defensively.
So I put on my thinking cap, and tried to figure out how one could
create an analogue of komidashi that would apply to Chess. And I came
up with a scoring scheme that at least *possibly* could reward
aggressive play, taking _reasonable_ risks to win, even in a match
instead of just in tournaments.
(Since tournaments lend themselves to certain abuses of draws that
don't arise in matches, i.e. "The Russians Have Fixed World Chess",
Fischer's famous Sports Illustrated article, perhaps additional
measures of the 1/3 - 1/3 nature are also warranted. But that is a
_separate issue_; the issue of making Chess more like that of the pre-
Steinitz era in excitement is one that has to be addressed all the way
up to the World Championship.)
Since komidashi worked by cancelling out the first player's advantage,
my scheme would need to somehow reduce White's advantage in Chess, so
that each game is balanced on a knife-edge. This is an apt metaphor,
because one of the other things in komidashi is that the score offset
always has an odd 1/2 stone, so as to make OTB draws impossible.
So my thought was this:
To reduce the vast expanse of advantage over which the result is a
draw, give scoring points for things less than checkmate. Partial
credit for stalemate, bare king, and even perpetual check!
To encourage aggressive play by both players, and to give Black an
advantage, start by giving Black more credit than White if a smaller
victory is achieved, but as the victories become more decisive, and
the credit to the winner increases, make the difference smaller.
So if the game remains in the territory where someone *wins* by
perpetual check, having only a tiny advantage, Black, who is
disadvantaged by moving second, is at an advantage due to the scoring.
So even if Black, by playing aggressively, only wins 1/3 of the time,
and loses 2/3 of the time, the scoring means Black ends up the winner!
But as the game moves *out* of that territory, up to where victory is
obtained the old-fashioned way, by checkmate, the scoring no longer
favors Black - full credit for a win is the same for both players. So
White can't just play for a win by perpetual check, he has to play for
checkmate to pull the game out of the region where the scoring favors
the other player.
This is the handwaving argument - I made up some arbitrary percentages
for what happens if players choose agressive/defensive strategies, and
applied a simplified version of this type of scoring to it, and then,
from the payoff matrix obtained, found out what game theory would
recommend. In the example, the result was that White was still best
advised to play defensively, but Black finally was encouraged to play
aggressively. This is somewhat like the situation in Go *before*
komidashi; it's still an improvement, and it just happens to be the
result because of the particular percentages I made up, too.
This is on the page
http://www.quadibloc.com/chess/ch0101.htm
So the scheme might work better than exhibited in the example, or not
as well.
John Savard
.
- Follow-Ups:
- References:
- Prev by Date: Re: Are they playing using Computer
- Next by Date: Re: What is a "Category 20" tournament?
- Previous by thread: Re: Two articles pondering the issues with Chess: Stale openings, declining prize pool, and issues with draws.
- Next by thread: Re: Two articles pondering the issues with Chess: Stale openings, declining prize pool, and issues with draws.
- Index(es):
Relevant Pages
|
Loading
