<marksteere@[EMAIL PROTECTED]
> wrote in message
news:d70c874d-b809-493f-8673-30591d024f32@[EMAIL PROTECTED]
> On Mar 12, 6:24 am, David Richerby <dav...@[EMAIL PROTECTED]
>
> wrote:
>>
>> Even assuming the players don't collude, it seems more likely than
>> anything else that, with perfect play, chess is a draw. It seems
>> wrong to penalize players for achieving what may well be the natural
>> result of a well-played game.
>>
>
> Well put.
Do we say that scoring 1 point for a free throw and 2 for a field goal in
basketball
"penalizes" free throws? Do we say that scoring 3 points for a field goal
in
American football, while 6 points for a touchdown, "penalizes" field
goals? Of
course not.
We are simply trying to design a scoring system that produces the
"best" competitions, where "best" is debatable but should include things
like
fun to play, fun to watch etc.
Chess' existing scoring (draw = one half of a win) does produce
collusion, big time. The last rounds of many tournaments are a
flurry of quick, uncontested draws. Why? Because the leaders
calculate that the return on playing for a win is low,
so they play the famous "grandmaster draw" (essentially collude not to
play and
split the point) and go home with a guaranteed prize. This behavior
is a function of the scoring system. If we changed the scoring system so
that
the return of playing for a win were higher, we get more players playing
for
wins. So it produces collusion to play chess vs. the existing system which
produces
collusion to not play chess. A straightforward improvement, if you ask
me.
|
