On Thu, 22 Mar 2007 15:02:01 -0700, "David Kane"
wrote:


Larry Tapper hit on the answer exactly. He noted
that "high-level players are canny utility maximizers".
They aren't maximizing good chess, they are maximizing
"highest tournament score". They aren't really
playing chess, they are engaging in a competition
which has chess as a component. They've
concluded that playing the best move is a
sub-optimal strategy and that it is better to
play a number of uncontested or partially
contested games.

Here's a nit: It might be better to say "striving for the best move
is a sub-optimal strategy...", since if they could count on *playing*
it, they probably would. The effort to always find the best move too
often leads to exhaustion driven blunders.

You've established nothing. To put it another way, if playing
the best move (aka "playing chess") at the GM level
produces about 25% draws (as evidenced by computers),
then why do GM's not play the best moves? (as
evidenced by their 60+% draw rates)

Larry Tapper hit on the answer exactly. He noted
that "high-level players are canny utility maximizers".
They aren't maximizing good chess, they are maximizing
"highest tournament score". They aren't really
playing chess, they are engaging in a competition
which has chess as a component. They've
concluded that playing the best move is a
sub-optimal strategy and that it is better to
play a number of uncontested or partially
contested games.

You are free to like that situation, but I am free
to dislike it. I think it is ridiculous to dissuade
the world's best players from playing the best
chess possible. To distort the game to the degree
where the draw rate is double or greater the
natural draw rate is lunacy.


Chess Canada November 2006 P.21

IM Zugic vs GM Charbonneau
Canadian Zonal (9) 26.08.2006

Quoting Igor Zugic
2006 Canadian Champion

"Just moments before I played this
move, the game Gerzhoy-Bluvshtein
ended in a draw. It meant I did not
have to win this game anymore. Following
my last move I immediately ofered
a draw to Pascal. He had no better
choice but to accept"

Yet his notes after 47.Bh5

"This wins too but I missed a pretty
checkmate 47.Rh5 Rh6 48.Rh6 Kh6
49.Rc2!! and black does not have a
good way to preventing checkmate"

So In a Zonal Final an IM admits to
having a won game but offers a draw
because

"However, I did not want to wait
a second to become the new
Canadian Champion."

On 22 Mar, 22:02, "David Kane" wrote:
"Mark Houlsby" wrote in message

oups.com...

On 22 Mar, 18:47, "David Kane" wrote:

The hotly contested, perfectly
played draw is not as common as you think.

In fact, it has never been played, so it is *exactly* as common as I
think.

You are very confused.

Not about this, I'm not.

There *are* many games
without obvious errors

Right. Without *obvious* errors. We don't know how many errors there
are even in the shortest GM draws. Nevertheless 1.d4 may prove, one
day to be a significant error. If it does, it will be beside the
point.

where the result
is a draw and the games appear fully contested.
(Admittedly, we haven't defined exactly what is
meant by "fully contested" but most thinking people
can accept that it is a meaningful concept.) But these
are a minority.


What has their being a minority got to do with anything? You're
writing about periphery.

In the 2006 IPCCC computer event, 8 of 35 (23%)
games were draws. That would be a reasonable
first guess for the natural drawing rate in contested
games at the GM level. You've yet to explain where
the extra 40+% of draws comes from.

As I just stated, we've *already* established this. We have, haven't
we? I mean, you're not going to troll this one *again*, I hope?

You've established nothing.

Oh, but I have, and if you could read, you'd understand that I have.

To put it another way, if playing
the best move (aka "playing chess") at the GM level
produces about 25% draws (as evidenced by computers),
then why do GM's not play the best moves?

That would be because it's not necessary to play the best moves in
order to score GM norms and have one's title ratified by FIDE.

DUH!


(as
evidenced by their 60+% draw rates)

Larry Tapper hit on the answer exactly. He noted
that "high-level players are canny utility maximizers".

Uh huh. Larry's a smart guy. I've never had to take issue with him for
reading badly or putting words in my mouth or anything like that.

They aren't maximizing good chess, they are maximizing
"highest tournament score".

Correct. We've already been over this, I think.

They aren't really
playing chess,

Oh, but they are!

The short draws are a necessary adjunct of the contested games
(regardless of the results of the contested games). READ NIMZOWITSCH'S
ESSAY and you'll find out why.

they are engaging in a competition
which has chess as a component. They've
concluded that playing the best move is a
sub-optimal strategy and that it is better to
play a number of uncontested or partially
contested games.


Yup. Standard boilerplate tournament technique. Understood by GMs.
Almost universally misunderstood by illiterate morons.

You are free to like that situation, but I am free
to dislike it.

Sure you are.

I think it is ridiculous to dissuade
the world's best players from playing the best
chess possible.

Newsflash: requiring the world's best players to avoid GM draws exerts
*exactly the same effect* as dissuading them from playing the best
chess possible. In order to play the best chess possible, it's
necessary to make short draws, quite often.


To distort the game to the degree
where the draw rate is double or greater the
natural draw rate is lunacy.

The natural draw rate is 100% of games' being drawn. Decisive games
result from decisive mistakes. It's impossible to produce double or
greater than double 100% of drawn games.

You really don't understand these very basic concepts, do you?

On 22 Mar, 22:27, Mike Murray wrote:
On Thu, 22 Mar 2007 15:02:01 -0700, "David Kane"

wrote:
Larry Tapper hit on the answer exactly. He noted
that "high-level players are canny utility maximizers".
They aren't maximizing good chess, they are maximizing
"highest tournament score". They aren't really
playing chess, they are engaging in a competition
which has chess as a component. They've
concluded that playing the best move is a
sub-optimal strategy and that it is better to
play a number of uncontested or partially
contested games.

Here's a nit: It might be better to say "striving for the best move
is a sub-optimal strategy...", since if they could count on *playing*
it, they probably would. The effort to always find the best move too
often leads to exhaustion driven blunders.

An excellent point well made, Mike. The above seems beyond doubt.

On 22 Mar, 22:42, "Inconnux" wrote:
You've established nothing. To put it another way, if playing
the best move (aka "playing chess") at the GM level
produces about 25% draws (as evidenced by computers),
then why do GM's not play the best moves? (as
evidenced by their 60+% draw rates)

Larry Tapper hit on the answer exactly. He noted
that "high-level players are canny utility maximizers".
They aren't maximizing good chess, they are maximizing
"highest tournament score". They aren't really
playing chess, they are engaging in a competition
which has chess as a component. They've
concluded that playing the best move is a
sub-optimal strategy and that it is better to
play a number of uncontested or partially
contested games.

You are free to like that situation, but I am free
to dislike it. I think it is ridiculous to dissuade
the world's best players from playing the best
chess possible. To distort the game to the degree
where the draw rate is double or greater the
natural draw rate is lunacy.

Chess Canada November 2006 P.21

IM Zugic vs GM Charbonneau
Canadian Zonal (9) 26.08.2006

Quoting Igor Zugic
2006 Canadian Champion

"Just moments before I played this
move, the game Gerzhoy-Bluvshtein
ended in a draw. It meant I did not
have to win this game anymore. Following
my last move I immediately ofered
a draw to Pascal. He had no better
choice but to accept"

Yet his notes after 47.Bh5

"This wins too but I missed a pretty
checkmate 47.Rh5 Rh6 48.Rh6 Kh6
49.Rc2!! and black does not have a
good way to preventing checkmate"

So In a Zonal Final an IM admits to
having a won game but offers a draw
because

"However, I did not want to wait
a second to become the new
Canadian Champion."

Yes, it's very common in GM (and, in this case, IM) praxis, no
question. A good contribution, thanks.

"Mike Murray" wrote in message
news
On Thu, 22 Mar 2007 15:02:01 -0700, "David Kane"
wrote:


Larry Tapper hit on the answer exactly. He noted
that "high-level players are canny utility maximizers".
They aren't maximizing good chess, they are maximizing
"highest tournament score". They aren't really
playing chess, they are engaging in a competition
which has chess as a component. They've
concluded that playing the best move is a
sub-optimal strategy and that it is better to
play a number of uncontested or partially
contested games.

Here's a nit: It might be better to say "striving for the best move
is a sub-optimal strategy...", since if they could count on *playing*
it, they probably would. The effort to always find the best move too
often leads to exhaustion driven blunders.

I stand corrected. After the usual bleating
from Mr. Houlsby, it's nice to be reminded that
some have the intelligence to understand the
written word.

"Inconnux" wrote in message
news:EDDMh.423$__3.220@edtnps90...

You've established nothing. To put it another way, if playing
the best move (aka "playing chess") at the GM level
produces about 25% draws (as evidenced by computers),
then why do GM's not play the best moves? (as
evidenced by their 60+% draw rates)

Larry Tapper hit on the answer exactly. He noted
that "high-level players are canny utility maximizers".
They aren't maximizing good chess, they are maximizing
"highest tournament score". They aren't really
playing chess, they are engaging in a competition
which has chess as a component. They've
concluded that playing the best move is a
sub-optimal strategy and that it is better to
play a number of uncontested or partially
contested games.

You are free to like that situation, but I am free
to dislike it. I think it is ridiculous to dissuade
the world's best players from playing the best
chess possible. To distort the game to the degree
where the draw rate is double or greater the
natural draw rate is lunacy.


Chess Canada November 2006 P.21

IM Zugic vs GM Charbonneau
Canadian Zonal (9) 26.08.2006

Quoting Igor Zugic
2006 Canadian Champion

"Just moments before I played this
move, the game Gerzhoy-Bluvshtein
ended in a draw. It meant I did not
have to win this game anymore. Following
my last move I immediately ofered
a draw to Pascal. He had no better
choice but to accept"

Yet his notes after 47.Bh5

"This wins too but I missed a pretty
checkmate 47.Rh5 Rh6 48.Rh6 Kh6
49.Rc2!! and black does not have a
good way to preventing checkmate"

So In a Zonal Final an IM admits to
having a won game but offers a draw
because

"However, I did not want to wait
a second to become the new
Canadian Champion."


Another clearcut case that
the alleged drawishness of
chess does not explain the
actual draws that occur.

The real question is to what
quantitative degree are draw
rates inflated by these external
factors, and what can be done
about them.

In some sense I think it is a mistake
to focus on special cases like this
one, or the usual target of the
obviously uncontested GM draw,
because it leads people to overlook
that the same factors behind
those events have an insidious
influence throughout chess.

Larry Tapper hit on the answer exactly. He noted
that "high-level players are canny utility maximizers".
They aren't maximizing good chess, they are maximizing
"highest tournament score". They aren't really
playing chess, they are engaging in a competition
which has chess as a component. They've
concluded that playing the best move is a
sub-optimal strategy and that it is better to
play a number of uncontested or partially
contested games.

"Just moments before I played this
move, the game Gerzhoy-Bluvshtein
ended in a draw. It meant I did not
have to win this game anymore. Following
my last move I immediately ofered
a draw to Pascal. He had no better
choice but to accept"

Yet his notes after 47.Bh5

"This wins too but I missed a pretty
checkmate 47.Rh5 Rh6 48.Rh6 Kh6
49.Rc2!! and black does not have a
good way to preventing checkmate"

So In a Zonal Final an IM admits to
having a won game but offers a draw

The real question is to what
quantitative degree are draw
rates inflated by these external
factors, and what can be done
about them.

In some sense I think it is a mistake
to focus on special cases like this
one, or the usual target of the
obviously uncontested GM draw,
because it leads people to overlook
that the same factors behind
those events have an insidious
influence throughout chess.

I posted this because it was an example
of an IM admitting publicly that he
offered a draw prematurely. How does
this affect the overall draw rate? I have
no idea. There are many examples
of fighting chess that end in a draw.
I doubt anyone has a problem with these
games.

How does this affect the local chess
tournaments? In the last tournament
I was in, in the last round the leader
found out that all he needed was a draw
and proceeded to offer his opponent a
draw before the game was started.
The TD immediately told him that that
was illegal to do. The game did end
up as a draw, but one has to wonder if
all he was playing for was a draw.
Since he was a young junior player, he
wasn't penalized for his draw offer.

I think giving the BAP system a try
would be an interesting experiment.
This might reduce the amount of draws
but then again it might produce its own
set of problems.

J.Lohner

Mark Houlsby wrote:
David Richerby wrote:
This isn't science: it's mathematics. I have seen the evidence and
I, like most people, believe that the evidence points towards chess
being a draw. But that is not proof.

Playing chess is *not* like building mathematical models.

Agreed. But we are not discussing playing chess: we are discussing
whether or not chess is a theoretical draw.


Informator is (its creators asserted) an attempt *scientifically* to
classify chess, principally by opening variation. Chess is more
science than mathematics. Scientific proof is fundamentally
different from mathematical proof.

Yes. In particular, mathematical proof exists and scientific proof
does not. A good scientist will never claim to have proven anything:
they will claim only that the results of their experiment are
consistent with one theory and inconsistent with some other theory.

The results of the `experiment' that is several hundred years of
high-level chess playing is consistent with the theory that chess is a
theoretical draw and inconsistent with the theories that it is a win
for white or black.


If you know of *any* credible evidence which suggests that chess is
not inherently a draw, would you be so good as to share it?

I never claimed there was any such evidence. As I have said several
times, I believe that chess is a draw with perfect play. I see no
evidence to the contrary. My point is that this is not proof; it is
merely an accumulation of evidence. This is why I am careful to say
that I *believe* that chess is a draw, rather than that I *know* it.


Dave.

--
David Richerby Disgusting Flammable Monk (TM): it's
www.chiark.greenend.org.uk/~davidr/ like a man of God but it burns really
easily and it'll turn your stomach!

On 23 Mar 2007 13:12:46 +0000 (GMT), David Richerby
wrote:

The results of the `experiment' that is several hundred years of
high-level chess playing is consistent with the theory that chess is a
theoretical draw and inconsistent with the theories that it is a win
for white or black.
...
I never claimed there was any such evidence. As I have said several
times, I believe that chess is a draw with perfect play. I see no
evidence to the contrary. My point is that this is not proof; it is
merely an accumulation of evidence. This is why I am careful to say
that I *believe* that chess is a draw, rather than that I *know* it.

In chess, evidence doesn't accumulate in quite the same way as it does
in experimental science. In chess, a bust to a variation renders
irrelevant all prior experience with that variation. It doesn't
matter that GM praxis is 40-1 for White in a given line, if you and
Fritz discover a forced win for Black. In this sense, it's more akin
to mathematics where, if someone proves a theorem, it's merely of
interest to biographers that several top mathematicians have spent
months and years failing to prove it.

In these cases, the accumulation of evidence functions more like
heuristics for how to spend one's finite amount of earthly time.