| |
Main
Date: 01 Jun 2008 18:45:44
From: Guy Macon
Subject: Does this so-called "draw problem" actually exist?
|
Ray Gordon wrote:
>Tablebases are up to seven pieces now. I'd call that a threat.
And you imagine that if the game was changed with a new promotion
rule -- and the new rules became popular -- that nobody would
create tablebases using the new rules?
I have yet to see a convincing argument that this so-called
"draw problem" in the rules of chess actually exists.
I have seen plenty of people who have a personal preference that
there be fewer draws, but that does not prove that there is a
"draw problem" in the current rules. Why is their preference
more important than mine?
I have seen plenty of claims that fewer draws will bring in more
players or more paying spectators, but I haven't seen any evidence
that these predictions will actually come to pass.
I have seen evidence of top-level players drawing early without
even trying to win (which is a distinct and separate issue from
there supposedly being too many draws), but no evidence that this
is caused by the present rules rather than, say, the reward system.
Nobody who has created a rule change to address the supposed
"draw problem" actually thinks that their new rule would prevail
over the current rules in any sort of election/vote/poll.
Nobody who has created a rule change to address the supposed
"draw problem" actually thinks that their new rule would prevail
over the current rules in any sort of free market such as
announcing a tournament using the new rules and seeing how many
players show up.
I like new chess rules and variants on the game. I like reading
about them, I like thinking about them, I like playtesting them,
and I hope that someone can come up with something useful that
becomes the new standard, as happened with "In Passing" pawn
captures (and no, I am not going to start speaking french just
so I can describe that rule...). But just because new rules
appeal to me personally. that doesn't mean that there is anything
wrong with the current rules.
In my opinion, there is nothing wrong with the current rules.
Chess is not broken, and does not require fixing.
In my opinion, the "solution" for those who prefer fewer draws
at the top level is simple, straightforward, and far from being
a new idea; simply don't invite players who draw "too often"
(whatever that means to you), and do invite players who you
consider to play "exciting" chess. "Problem" (if one actually
exists) solved.
--
Guy Macon
<http://www.guymacon.com/ >
|
|
| |
Date: 11 Jun 2008 18:30:20
From: Guy Macon
Subject: Re: Does this so-called "draw problem" actually exist?
|
Here is another interesting idea that occurred to me.
Right now, a win is one point, a loss is 0 points, a draw
is 0.5 points, and draws can be agreed upon.
What if we make it so that instead of offering a draw
("shall we end the game now at 0.5 to 0.5?), you can offer
some other division of points ("shall we change it so that
whoever wins gets 0.55 points and whoever loses gets 0.45
points?") and play on.
The opponent can them reject the offer and play on at the
usual 1.0/0.5/0.0, make a counter-offer, accept and play
on, or accept and resign.
This would raise the cost of accepting a draw and lower the
risk (both the upside risk and the downside risk) of playing
on in a more aggressive manner.
If both players really think that there is no chance of
either side winning even with aggressive play, they can
offer and counter-offer until the offer is 0.5 to 0.5,
then resign, thus making the outcome the same as our
current draw. Or they could come close (0.51 to 0.49,
for example) and play on with both sides trying risky
strategies now that the stakes are so much lower.
What do you think?
--
Guy Macon
<http://www.guymacon.com/ >
|
| | |
Date: 12 Jun 2008 02:44:25
From: Quadibloc
Subject: Re: Does this so-called "draw problem" actually exist?
|
On Jun 11, 12:30 pm, Guy Macon <http://www.guymacon.com/ > wrote:
> This would raise the cost of accepting a draw
I was wondering how it was supposed to do that, but _now_ I finally
grasp the intent of your proposal.
If one player has a slight advantage, you are seeking a scheme whereby
that player would have to _surrender_ that advantage if he accepted a
draw, because he has another option available by means of which he can
retain the advantage.
The option as stated doesn't seem to give the player with the slight
advantage any bargaining power with which to obtain anything for his
advantage. If slight advantages lead to draws OTB, then the opponent's
rational strategy is to play defensively against that advantage, not
concede partial credit for it.
It might, however, be the appropriate rule for agreeing to draws if
the OTB result included - as my Dynamic Scoring proposal does -
partial scores for varying results. Then players would have the option
of agreeing to the score they both expect to realize OTB -
intermediate values would still, usually, play little role.
John Savard
|
| | |
Date: 11 Jun 2008 16:55:43
From: Quadibloc
Subject: Re: Does this so-called "draw problem" actually exist?
|
On Jun 11, 12:30 pm, Guy Macon <http://www.guymacon.com/ > wrote:
> This would raise the cost of accepting a draw and lower the
> risk (both the upside risk and the downside risk) of playing
> on in a more aggressive manner.
It seems to be just a different way of offering a draw in effect, so
I'm afraid I have to concur with the critics on this one. Since draws
are usually offered to avoid working hard at the chessboard, it would
seem that once the point spread is agreed downwards, instead of
attacking play, one would get careless woodpushing.
But the idea might give good results if tweaked somehow; the doubling
cube, after all, put new life in Backgammon.
John Savard
|
| | |
Date: 11 Jun 2008 13:13:25
From: Mike Murray
Subject: Re: Does this so-called "draw problem" actually exist?
|
On Wed, 11 Jun 2008 18:30:20 +0000, Guy Macon
<http://www.guymacon.com/ > wrote:
>What do you think?
Not a winner, IMO.
Seems like this would offer more creative opportunities for short
draws and point selling when class prizes, titles, etc., are at stake.
And the bargaining might lead to cacophony in large Swiss
tournaments.
|
| | | |
Date: 11 Jun 2008 22:35:28
From: David Richerby
Subject: Re: Does this so-called "draw problem" actually exist?
|
Mike Murray <[email protected] > wrote:
> Guy Macon wrote:
>> [A point-bargaining scheme.]
>
> Seems like this would offer more creative opportunities for short
> draws and point selling when class prizes, titles, etc., are at
> stake.
You mean that a player who needed, say, only a quarter of a point to
win a tournament might immediately offer to split a win 0.75-0.25?
That could be a down side but remember that, in the last round of a
Swiss, the leader is likely to be playing his closest rival who would,
of course, decline any such offer.
> And the bargaining might lead to cacophony in large Swiss
> tournaments.
Seems unlikely. How often have you even heard a draw offer from more
than a couple of boards away?
Dave.
--
David Richerby Poisonous Tongs (TM): it's like a
www.chiark.greenend.org.uk/~davidr/ pair of tongs but it'll kill you
in seconds!
|
| | | | |
Date: 11 Jun 2008 17:24:54
From: Mike Murray
Subject: Re: Does this so-called "draw problem" actually exist?
|
On 11 Jun 2008 22:35:28 +0100 (BST), David Richerby
<[email protected] > wrote:
>Mike Murray <[email protected]> wrote:
>> Guy Macon wrote:
>>> [A point-bargaining scheme.]
>> Seems like this would offer more creative opportunities for short
>> draws and point selling when class prizes, titles, etc., are at
>> stake.
>You mean that a player who needed, say, only a quarter of a point to
>win a tournament might immediately offer to split a win 0.75-0.25?
>That could be a down side but remember that, in the last round of a
>Swiss, the leader is likely to be playing his closest rival who would,
>of course, decline any such offer.
Not necessarily. The leader may well be playing a weaker player for
whom a quarter point scrap would give place money. Happens quite often
in Swiss events. The leader probably wouldn't agree to a quick draw
with such an opponent, but three-quarter point (perhaps sweetened by
some kickback) might let him win the event outright. And, at the
higher levels, you'd have norm considerations, in addition to
tournament placing.
>> And the bargaining might lead to cacophony in large Swiss
>> tournaments.
>Seems unlikely. How often have you even heard a draw offer from more
>than a couple of boards away?
But that's under current rules "Draw?" "Declined". (maybe not even
that -- just silence if declined or a handshake if accepted). Versus
the proposed point haggling which could go on for some time, with the
voices getting louder as the two negotiators get into it.
|
| |
Date: 09 Jun 2008 17:48:25
From: Guy Macon
Subject: Re: Does this so-called "draw problem" actually exist?
|
Let's review, shall we?
After having considerable success using the attack-at-all-costs
style that was popular in 1860s, William Steinitz developed a
new, more positional style of play and went on to demonstrate
that it was superior to the old style. By the time Steinitz
died in the year 1900, his new approach was widely accepted.
Although I haven't seen any actual statistics, it is widely
believed that (and makes sense that) the new style leads to
more draws and fewer decisive victories.
Chess in the United States hit a peak around 1972 -- around
the time of because of the Fischer vs. Spassky match. That
match, BTW, had more than 50% of the games end in draws,
including a run of seven drawn games in a row. By the end
of the 1970's, US Chess was declining, and has never again
reached the 1972 level.
Strong chess computers came later; in 1976 the best that
a computer could do was to win the Class B section of a
tournament, and in 1978 a computer drew an international
master in a tournament for the first time.
The Soviet Union collapse started in the mid 1980, and
was complete by 1990.
Given the above facts, I find it difficult to assign the blame
for the decline in US Chess to Steinitz (much too early) or to
the rise of strong computers or the fall of the USSR (too late).
--
Guy Macon
<http://www.guymacon.com/ >
|
| | |
Date: 10 Jun 2008 02:27:16
From: Quadibloc
Subject: Re: Does this so-called "draw problem" actually exist?
|
On Jun 9, 7:58 pm, Quadibloc <[email protected] > wrote:
> Rather, it's dull, defensive play that comes about
> all too often because of the same characteristic of Chess that leads
> to draws
What is the name of that characteristic?
Go, before _komidashi_, had an expected outcome of a three-point win
by the first player (Black). Since that was a win, Black had no
incentive to play more aggressively, and if Black played well, White
had no real way of changing the result.
In Chess, the expected outcome is a draw; if White plays for one,
Black finds it difficult (though not impossible, but with the risk of
losing more often than winning) to change the result.
If it is also true that when Black plays defensively, White, despite
his advantage, will only win very rarely as long as he plays
cautiously enough so that his score, when playing the white pieces, is
a plus score, then Chess has *almost* the same problem as Go before
komidashi. It would have the same problem if "very rarely" became
"virtually never".
This is despite the fact that in one case what happened was that one
player won, and in the other case what happened was a draw.
When players are evenly matched, and are playing above a certain
standard, not only do games have an expected outcome, but there is
little that either player can do effectively to have a chance of
varying from that outcome except through play sufficiently wild as to
improve the result for his opponent, then we have the problem for
which I don't have a simple name.
In Chess, the symptom is draws, while in Go, the symptom is three-
stone victories for Black.
John Savard
|
| | | |
Date: 11 Jun 2008 16:59:25
From: Quadibloc
Subject: Re: Does this so-called "draw problem" actually exist?
|
On Jun 11, 2:48 pm, "David Kane" <[email protected] > wrote:
> Your idea has an additional element - that of adding
> complexity to the game by changing the number of
> end states (and has no counterpart in Go).
No, but that's because Go had lots of end states to begin with.
Although the result is just win/lose/draw, one can evaluate the score
in points of territory.
A more exact counterpart to komidashi might be, for example, to decree
that White wins with bare king or better, and everything else -
including a dead draw - is a win for Black. But _that_ requires having
in hand a precise value for White's advantage. Since I didn't have
one, that's why I came up with the point scale I did - people getting
experience with that system would eventually come to an equilibrium
which would show what White's advantage is, and the simpler scheme
could be used.
John Savard
|
| | | | |
Date: 11 Jun 2008 17:30:14
From: David Kane
Subject: Re: Does this so-called "draw problem" actually exist?
|
"Quadibloc" <[email protected] > wrote in message
news:[email protected]...
> On Jun 11, 2:48 pm, "David Kane" <[email protected]> wrote:
>
>> Your idea has an additional element - that of adding
>> complexity to the game by changing the number of
>> end states (and has no counterpart in Go).
>
> No, but that's because Go had lots of end states to begin with.
> Although the result is just win/lose/draw, one can evaluate the score
> in points of territory.
That's not really relevant. The sole motivation for komi
is to balance the game. Chess can also be balanced simply
by changing the score for White and Black wins, so that
each color has the same expectation.
Many have proposed changing the rules of chess so that
stalemate is a win. The original element of your proposal
is counting it as something in between a draw and a win.
It's unknown whether that would have any practical effect on
play, but I see that at least in principle it could. Perhaps the
simplest test would be to find people willing to play with
stalemate = win and see whether there is any change.
|
| | | |
Date: 11 Jun 2008 12:48:45
From: Quadibloc
Subject: Re: Does this so-called "draw problem" actually exist?
|
On Jun 11, 12:35=A0pm, "David Kane" <[email protected] > wrote:
> Actually, I think one of the difficulties you are having is
> that you are *incorrectly* equating your idea with komidashi
> in Go. Go would not have many draws even with an integer
> handicap.
This is true, but then, as I've noted, the problem isn't really about
draws.
> The real benefit of your idea is that you are taking a
> game with a large winning margin (leading to draws)
> and turning it into one with a somewhat smaller
> margin (by having different end states).
Yes, but that in itself isn't original - as it has been noted, bare
king and stalemate were inferior wins before. Because chess is very
different from Go, I couldn't steal komidashi without a lot of
adaptation; and the idea I came up with was to start off with giving
Black a huge proportional advantage when it came to the relatively
small reward for the smallest victory (perpetual check) - so that
Black was encouraged to try to win - and then make the point system
become more equal as the margin of victory was higher - so that White,
with the first move advantage, is discouraged from trying to coast on
that advantage to the smaller victories.
> BTW, I don't think you want machine tests. You need
> to get people to play (even two) and then describe
> what happened and how it felt.
Well, the idea is out there; it's on my web site. If anyone's
interested, they can go right ahead and investigate it.
John Savard
|
| | | | |
Date: 11 Jun 2008 13:48:49
From: David Kane
Subject: Re: Does this so-called "draw problem" actually exist?
|
>"Quadibloc" <[email protected]> wrote in message
>news:eac5fb0d-658a-41ef-885c-dbe7f4cdd7b2@y38g2000hsy.googlegroups.com...
>On Jun 11, 12:35 pm, "David Kane" <[email protected]> wrote:
>
>> Actually, I think one of the difficulties you are having is
>> that you are *incorrectly* equating your idea with komidashi
>> in Go. Go would not have many draws even with an integer
>> handicap.
>This is true, but then, as I've noted, the problem isn't really about
>draws.
No. But it is about *playing for* draws. That just doesn't
happen in Go, and komi had nothing to do with it.
>> The real benefit of your idea is that you are taking a
>> game with a large winning margin (leading to draws)
>> and turning it into one with a somewhat smaller
>> margin (by having different end states).
>Yes, but that in itself isn't original - as it has been noted, bare
>king and stalemate were inferior wins before.
When? Chess has had just 3 end states for a long time.
> Because chess is very
>different from Go, I couldn't steal komidashi without a lot of
>adaptation; and the idea I came up with was to start off with giving
>Black a huge proportional advantage when it came to the relatively
>small reward for the smallest victory (perpetual check) - so that
>Black was encouraged to try to win - and then make the point system
>become more equal as the margin of victory was higher - so that White,
>with the first move advantage, is discouraged from trying to coast on
>that advantage to the smaller victories.
The purpose of komi was to change the White/Black
balance. This can be easily done in chess with alternate
scoring, by simply giving Black a bigger score for draws
and wins. I believe this, alone, has enormous promise
and warrants further investigation for the reason that it
can devalue draws, thereby reducing the strategies
based on playing for them.
Your idea has an additional element - that of adding
complexity to the game by changing the number of
end states (and has no counterpart in Go). This, in
theory, could increase the number of positions
where there is something to play for. (Practical
evidence that it would have any effect remains
lacking).
>> BTW, I don't think you want machine tests. You need
>> to get people to play (even two) and then describe
>> what happened and how it felt.
>Well, the idea is out there; it's on my web site. If anyone's
>interested, they can go right ahead and investigate it.
Even a 10-game match between two equal
master level players would give useful results.
My hunch is that the details of your intriguing
proposal will not stand up to practical tests,
but it is interesting enough to warrant them.
|
| | | |
Date: 10 Jun 2008 22:12:18
From: Quadibloc
Subject: Re: Does this so-called "draw problem" actually exist?
|
On Jun 10, 10:23 pm, "David Kane" <[email protected] > wrote:
> I think I'm starting to appreciate
> the merits of your proposal more - am still not sure whether the
> valuing the various end conditions differently would make a
> practical difference. Have you ever tried it?
Not yet; I am beginning to look into how I might try to set up a
modified chess engine to get some numbers.
Based on simple knowledge about Chess, it seemed reasonable to view
stalemate as almost as difficult to achieve as checkmate, bare king as
a bit easier, and perpetual check as much easier.
What I didn't know, though, and wouldn't know without getting a lot of
people - not just myself - or at least a lot of chess engines in some
type of computer run as a first cut - is where the "value" of Chess
is. If even perpetual check counted as a win, then Chess might still
usually be a draw - or it might always end up as a win (by perpetual
check) for white.
So this is why my proposal is a bit complicated. I designed one that
ought to "work" even though I didn't quite know where to aim.
Black still wins _some_ of the time by checkmate. So - following the
way komidashi worked for Go - I figured that since Black is at a
disadvantage, only giving a large advantage for the smallest of
victories would Black have a reason to struggle for a win.
White has the advantage - so to encourage White not to play
defensively, while Black has a big advantage in the way the scoring is
set up if the wins are small ones, the advantage disappears as the
wins get bigger.
So the hope is that in practice, it will lead to battles, even though
I can't predict what these battles will be over.
It could well be that bare King and stalemate are close enough to the
real victory of checkmate that they will happen only occasionally. But
perpetual check is much easier to obtain an opportunity for; whether
that will mean that there is a real fight over it, so that sometimes
Black, and sometimes White, wins this way, or whether White almost
always wins this way except when his advantage leads him to spurn it
and go to a greater victory - only experience will tell.
And in Go, the real benefits of komidashi were only realized when Go
Seigen achieved impressive results and deepened the understanding of
the game. A change that is as much like komidashi as I could come up
with might allow a Grandmaster to forge a new style of play, built on
the extended understanding Steinitz gave us instead of rejecting it,
but with additional attacking elements that the new scoring system
permits. I'm no Kramnik or Anand, so I can't do that.
What I've tried to do - and I'm repeating myself, since I've said this
before - is to collect information on what has been tried and worked
for Checkers and Go, compare attitudes among players of those games to
the changes made there, look at what has been tried in Chess - and try
to come up with something at least a little different than what's
already been tried.
First, I came up with a modest proposal - if some people don't like
Fischerrandom because it looks messy - and if Capablanca Chess could
grow a body of opening theory as impressive as ordinary Chess given a
few decades of people playing it - and if Chess players would not
stand for imposed openings, which even Checkers players barely
tolerate - how about Chess with some added pieces, but chosen from a
number of possibilities?
But my Random Variant Chess, even if it was a little original, didn't
address any really important problem. Sure, memorizing lots of
openings is discouraging, and a game that concentrates on tactical
skill is perhaps better pedagogically.
I didn't think there was a way to get Chess players to fight harder.
Partial points for stalemate might help a little, because that would
chip away at draws from insufficient material. But then I found out
how successful komidashi had been for Go. Chess, though, isn't won on
points, so translating that into a Chess equivalent couldn't be exact.
John Savard
|
| | | | |
Date: 11 Jun 2008 11:35:03
From: David Kane
Subject: Re: Does this so-called "draw problem" actually exist?
|
"Quadibloc" <[email protected] > wrote in message
news:3610e3d9-b05d-4c87-b78a-e06f7b08acfd@k13g2000hse.googlegroups.com...
> On Jun 10, 10:23 pm, "David Kane" <[email protected]> wrote:
>> I think I'm starting to appreciate
>> the merits of your proposal more - am still not sure whether the
>> valuing the various end conditions differently would make a
>> practical difference. Have you ever tried it?
>
> Not yet; I am beginning to look into how I might try to set up a
> modified chess engine to get some numbers.
>
> Based on simple knowledge about Chess, it seemed reasonable to view
> stalemate as almost as difficult to achieve as checkmate, bare king as
> a bit easier, and perpetual check as much easier.
>
> What I didn't know, though, and wouldn't know without getting a lot of
> people - not just myself - or at least a lot of chess engines in some
> type of computer run as a first cut - is where the "value" of Chess
> is. If even perpetual check counted as a win, then Chess might still
> usually be a draw - or it might always end up as a win (by perpetual
> check) for white.
>
> So this is why my proposal is a bit complicated. I designed one that
> ought to "work" even though I didn't quite know where to aim.
>
> Black still wins _some_ of the time by checkmate. So - following the
> way komidashi worked for Go - I figured that since Black is at a
> disadvantage, only giving a large advantage for the smallest of
> victories would Black have a reason to struggle for a win.
>
> White has the advantage - so to encourage White not to play
> defensively, while Black has a big advantage in the way the scoring is
> set up if the wins are small ones, the advantage disappears as the
> wins get bigger.
>
> So the hope is that in practice, it will lead to battles, even though
> I can't predict what these battles will be over.
>
> It could well be that bare King and stalemate are close enough to the
> real victory of checkmate that they will happen only occasionally. But
> perpetual check is much easier to obtain an opportunity for; whether
> that will mean that there is a real fight over it, so that sometimes
> Black, and sometimes White, wins this way, or whether White almost
> always wins this way except when his advantage leads him to spurn it
> and go to a greater victory - only experience will tell.
>
> And in Go, the real benefits of komidashi were only realized when Go
> Seigen achieved impressive results and deepened the understanding of
> the game. A change that is as much like komidashi as I could come up
> with might allow a Grandmaster to forge a new style of play, built on
> the extended understanding Steinitz gave us instead of rejecting it,
> but with additional attacking elements that the new scoring system
> permits. I'm no Kramnik or Anand, so I can't do that.
>
> What I've tried to do - and I'm repeating myself, since I've said this
> before - is to collect information on what has been tried and worked
> for Checkers and Go, compare attitudes among players of those games to
> the changes made there, look at what has been tried in Chess - and try
> to come up with something at least a little different than what's
> already been tried.
>
> First, I came up with a modest proposal - if some people don't like
> Fischerrandom because it looks messy - and if Capablanca Chess could
> grow a body of opening theory as impressive as ordinary Chess given a
> few decades of people playing it - and if Chess players would not
> stand for imposed openings, which even Checkers players barely
> tolerate - how about Chess with some added pieces, but chosen from a
> number of possibilities?
>
> But my Random Variant Chess, even if it was a little original, didn't
> address any really important problem. Sure, memorizing lots of
> openings is discouraging, and a game that concentrates on tactical
> skill is perhaps better pedagogically.
>
> I didn't think there was a way to get Chess players to fight harder.
> Partial points for stalemate might help a little, because that would
> chip away at draws from insufficient material. But then I found out
> how successful komidashi had been for Go. Chess, though, isn't won on
> points, so translating that into a Chess equivalent couldn't be exact.
Actually, I think one of the difficulties you are having is
that you are *incorrectly* equating your idea with komidashi
in Go. Go would not have many draws even with an integer
handicap.
The real benefit of your idea is that you are taking a
game with a large winning margin (leading to draws)
and turning it into one with a somewhat smaller
margin (by having different end states).
BTW, I don't think you want machine tests. You need
to get people to play (even two) and then describe
what happened and how it felt.
>
> John Savard
|
| | | | |
Date: 11 Jun 2008 07:49:20
From: Quadibloc
Subject: Re: Does this so-called "draw problem" actually exist?
|
On Jun 11, 8:42=A0am, David Richerby <[email protected] >
wrote:
> (This last
> part can be fixed easily by giving the bared king one more move to
> complete the exchange and declaring the game drawn in such cases.)
Yes, that may be a good idea; on the other hand, the ability to
control whether or not the exchange takes place may be worth some
credit.
John Savard
|
| | | | | |
Date: 11 Jun 2008 16:48:37
From: David Richerby
Subject: Re: Does this so-called "draw problem" actually exist?
|
Quadibloc <[email protected] > wrote:
> David Richerby <[email protected]> wrote:
>> (This last part can be fixed easily by giving the bared king one
>> more move to complete the exchange and declaring the game drawn in
>> such cases.)
>
> Yes, that may be a good idea; on the other hand, the ability to
> control whether or not the exchange takes place may be worth some
> credit.
I doubt it but there's nothing that says that the two-bare-kings draw
has to be scored equally.
Dave.
--
David Richerby Sadistic Monk (TM): it's like a man
www.chiark.greenend.org.uk/~davidr/ of God but it wants to hurt you!
|
| | | | |
Date: 11 Jun 2008 13:17:15
From: David Richerby
Subject: Re: Does this so-called "draw problem" actually exist?
|
Quadibloc <[email protected] > wrote:
> "David Kane" <[email protected]> wrote:
>> I think I'm starting to appreciate the merits of your proposal more
>> - am still not sure whether the valuing the various end conditions
>> differently would make a practical difference. Have you ever tried
>> it?
>
> Not yet; I am beginning to look into how I might try to set up a
> modified chess engine to get some numbers.
That could be quite difficult, since the win-draw-loss assumption is
probably quite deeply ingrained in the code. For example, I think the
most significant effect of your proposed scoring system is the role
played by stalemate in altering endgame evaluations. Engines tend to
be pretty weak on their own in the endgame because it's easier to code
up a tablebase than to search deeply. So, to get good results, I
think you'd have to recalculate all the tablebases, too.
> Based on simple knowledge about Chess, it seemed reasonable to view
> stalemate as almost as difficult to achieve as checkmate, bare king
> as a bit easier, and perpetual check as much easier.
In general positions, stalemate is *much* harder to achieve than
checkmate. For checkmate, I just need to prevent your king moving and
put it in check; for stalemate, I need to prevent *all* your pieces
from moving.
And I'm still wary that you're turning stalemate and perpetual check,
which are defensive resources in chess as it is played at the moment,
into attacking resources.
Almost every stalemate in chess occurs either because:
1. an inexperienced player (or a more experienced player in time
trouble) who has an overwhelming material advantage accidentally
leaves his opponent with no legal moves; or
2. more commonly, a player who is a pawn down in the endgame says,
`Look, either you try to promote that pawn and I slot in behind it
and you stalemate me, or you don't try to promote it and we agree a
draw because nothing is happening.'
It's not clear to me that either player in either situation should be
rewarded beyond the score of a draw.
Nobody ever says, `I know -- I'll try to stalemate my opponent' in
chess at the moment; rewarding stalemate as a kind of partial win
would change that and I'm not sure that the change would be as minor
as you think.
Likewise, perpetual check is used in two ways:
1. a player who thinks the game should be drawn gives perpetual check
to force the issue, rather than offering a draw and hoping for the
co-operation of the other player; or
2. more commonly, a player who is at a significant disadvantage uses
perpetual check to avoid losing and rescue a half point from an
otherwise bad position.
Neither stalemate nor perpetual check is treated in current chess as a
sort of `checkmate-lite'.
I'm also not convinced by the bare king thing. Consider the following
(admittedly rather artificial) position at the end of a pawn race:
White has a pawn on a7 his king on g2; Black has a pawn on h2 and his
king on b7. Whoever has the move can reduce his opponent to a bare
king by capturing the pawn but does either player really deserve more
than half a point? This, to me, seems to be more arbitrary than the
existing reciprocal zugzwang positions -- e.g., White Ke5 Pd4, Black
Kc4 Pd5 -- where whoever has the move loses, but loses because his
opponent promotes a pawn and checkmates him.
Dave.
--
David Richerby Transparent Slimy Chicken (TM): it's
www.chiark.greenend.org.uk/~davidr/ like a farm animal but it's covered in
goo and you can see right through it!
|
| | | | | |
Date: 11 Jun 2008 10:24:51
From: Quadibloc
Subject: Re: Does this so-called "draw problem" actually exist?
|
On Jun 11, 9:40=A0am, [email protected] (Andy Walker) wrote:
> =A0 =A0 =A0 =A0 Perpetual check is not, of course, a very well-defined
> concept. =A0If you merely mean repetition, then it's unclear, in
> general, who deserves the [minor] win in such cases.
I can provide a definition: perpetual check exists when one player is
willing and able to give check indefinitely. This definition existed
in older forms of Chess rules, even though it is not part of current
rules.
The player giving check is doing something offensive :) , and thus
gets the very minor win.
> =A0 =A0 =A0 =A0 Bared king and stalemate, "of course", used to be minor
> wins in old versions of chess, so the idea is not exactly original!
That's absolutely true, and people before me have suggested things
like 3/4 - 1/4 point splits for stalemate too.
My originality consists in giving Black extra credit for minor wins -
more so, the more minor the win is. And it's part of an attempt to
crib from _komidashi_ to boot.
John Savard
|
| | | | | | |
Date: 11 Jun 2008 20:35:28
From: David Richerby
Subject: Re: Does this so-called "draw problem" actually exist?
|
Quadibloc <[email protected] > wrote:
> On Jun 11, 9:40=A0am, [email protected] (Andy Walker) wrote:
>> Perpetual check is not, of course, a very well-defined
>> concept. If you merely mean repetition, then it's unclear, in
>> general, who deserves the [minor] win in such cases.
>
> I can provide a definition: perpetual check exists when one player
> is willing and able to give check indefinitely.
`Willing' is rather hard to quantify. And what about cases where the
checked player could, for example, block a check but would suffer
unacceptable material loss by so doing?
> This definition existed in older forms of Chess rules, even though
> it is not part of current rules.
But was perpetual check ever formally defined?
> The player giving check is doing something offensive :) , and thus
> gets the very minor win.
No! The player giving check is doing something defensive! He's
filibustering! He's stopping his opponent checkmating him by putting
the opponent in checks forever. Nobody ever gives perpetual check
from a position of strength.
Dave.
--
David Richerby Psychotic Laptop Flower (TM): it's
www.chiark.greenend.org.uk/~davidr/ like a flower that you can put on your
lap but it wants to kill you!
|
| | | |
Date: 10 Jun 2008 11:22:16
From: Quadibloc
Subject: Re: Does this so-called "draw problem" actually exist?
|
On Jun 10, 10:40=A0am, David Richerby <[email protected] >
wrote:
> The other top-ten players are running at something closer to 60%
> draws.
Even that is 40% non-draws, rather than, say, 10% non-draws or even
20% non-draws. Drawing half the time may be mildly annoying, but
indeed it doesn't qualify as dramatic evidence that chess is played
out.
But, as I noted, draws are only a symptom, not the problem itself. If
White always won - basically the situation Go had before komidashi -
that would be just as bad.
I think it's reasonable to say that the situation some (perhaps
unreasonable) people want is: White wins 41% of the time, Black wins
39% of the time, and it's a draw only 20% of the time. The idea being
that with draws in the minority, AND with either player being able to
win, there must be exciting, active play going on. If Black hardly
ever wins, even if White does win a fair amount of the time, in high-
level play, then there is already a... hint... of a problem.
What with Topolev being the person Kramink got the World Championship
from - and Viswarathan Anand, the current World Championship, having
gotten it from Kramnik, well, you don't get to be World Champion
without winning a few games!
The improving standard of Chess play over the centuries and decades -
with Steinitz being a watershed in that progress - means that the
'distance' between a win by White, a draw, and a win by Black has
increased - because now Chess players can navigate better around the
shoals of Chess, and so what was a tight squeeze is no longer so. A
reaction to that might be to include minor wins like stalemate for
fewer points, so that the events of a game will give players
advantages they can turn into a result of some kind, so that they have
something to fight for more often.
Solving the 'problem' by starting over with a 10 by 10 board isn't the
solution, because an improved standard of play is a good and valuable
thing we want to keep - but we want to get rid of the undesirable side-
effect of making games more predictable.
John Savard
|
| | | | |
Date: 11 Jun 2008 19:11:36
From: Guy Macon
Subject: Re: Does this so-called "draw problem" actually exist?
|
Quadibloc wrote:
>I don't know enough to offer any guarantees; I just think that I seem
>to have come up with an original idea with _some_ promise.
You always do. <grin >
|
| | | | |
Date: 11 Jun 2008 06:15:32
From: Quadibloc
Subject: Re: Does this so-called "draw problem" actually exist?
|
On Jun 11, 6:17 am, David Richerby <[email protected] >
wrote:
> In general positions, stalemate is *much* harder to achieve than
> checkmate. For checkmate, I just need to prevent your king moving and
> put it in check; for stalemate, I need to prevent *all* your pieces
> from moving.
That would be an issue if I awarded _more_ points for stalemate than
for checkmate.
But when checkmate puts 1 point on the board, and stalemate puts 1/5
of a point on the board, that would only tend to indicate that this
rule change, by itself, might have very little effect.
> And I'm still wary that you're turning stalemate and perpetual check,
> which are defensive resources in chess as it is played at the moment,
> into attacking resources.
But look at the pittance I'm awarding for them. So it still makes very
much sense to avoid them if checkmate is possible. (Suddenly, chess
programs would have to have four different "contempt factors".)
> It's not clear to me that either player in either situation should be
> rewarded beyond the score of a draw.
>
> Nobody ever says, `I know -- I'll try to stalemate my opponent' in
> chess at the moment; rewarding stalemate as a kind of partial win
> would change that and I'm not sure that the change would be as minor
> as you think.
> Neither stalemate nor perpetual check is treated in current chess as a
> sort of `checkmate-lite'.
>
> I'm also not convinced by the bare king thing.
Including perpetual check, because that is relatively easy to achieve,
is, on the one hand, the most dangerous part of the proposal. That
indeed could end up changing chess drastically, because it might mean
players would have to be constantly on their guard against that
possibility.
Bare king and stalemate both require the player achieving them to be
dominant, though, so if they're awarded points *much* less than given
for checkmate, I think it's reasonably sound to expect the change to
chess would be limited. Because I was afraid it might be too limited
to do any good - and because I was trying to imitate komidashi, which
made OTB draws impossible with the odd half-point - I dragged in
perpetual check.
It may *well* be that my proposal is too badly flawed to be salvaged
by re-jigging the scores - stalemate and bare king might be too
difficult, too "close" to checkmate to make any difference, and
perpetual check too easy to be included even as the most minor of
victories without destroying the game.
I don't know enough to offer any guarantees; I just think that I seem
to have come up with an original idea with _some_ promise.
John Savard
|
| | | | | |
Date: 11 Jun 2008 15:42:17
From: David Richerby
Subject: Re: Does this so-called "draw problem" actually exist?
|
Quadibloc <[email protected] > wrote:
> Including perpetual check, because that is relatively easy to
> achieve, is, on the one hand, the most dangerous part of the
> proposal. That indeed could end up changing chess drastically,
> because it might mean players would have to be constantly on their
> guard against that possibility.
I think perpetual check is by far the least significant of your
proposed changes.
> Bare king and stalemate both require the player achieving them to
> be dominant, though, so if they're awarded points *much* less than
> given for checkmate, I think it's reasonably sound to expect the
> change to chess would be limited.
I've pointed out that bare king leads to clearly equal positions being
scored as a `minor win' for one player or the other based on who has
the move and that changing the scoring for stalemate has what looks to
me like a significant effect on the endgame. You don't seem to have
responded to either of these points. You seem to be arguing based on
the opinion that stalemate and bare king (and maybe even perpetual
check, too) are a sort of `checkmate-lite' and you don't seem to be
addressing my arguments that they're completely different from check-
mate.
> It may *well* be that my proposal is too badly flawed to be salvaged
> by re-jigging the scores - stalemate and bare king might be too
> difficult, too "close" to checkmate to make any difference
But stalemate and bare king are a million miles from checkmate, as
I've been trying to explain! Making stalemate a `minor win' means
that a huge class of asymmetric pawn endings are no longer equal.
Making bare king a minor win means that KB-K, KN-K and KNN-K are no
longer equal, and nor are the large fraction of KP-K endgames that are
currently drawn. A large number of endgames where each player has one
man plus his king stop being equal and are tipped in the balance of
one player or the other, based on what is to me the arbitrary
criterion of who makes the first capture in any exchange. (This last
part can be fixed easily by giving the bared king one more move to
complete the exchange and declaring the game drawn in such cases.)
> [...] I seem to have come up with an original idea with _some_
> promise.
Absolutely.
Dave.
--
David Richerby Disgusting Psychotic Atlas (TM):
www.chiark.greenend.org.uk/~davidr/ it's like a map of the world but
it wants to kill you and it'll turn
your stomach!
|
| | | | | | |
Date: 11 Jun 2008 15:40:35
From: Andy Walker
Subject: Re: Does this so-called "draw problem" actually exist?
|
In article <fxy*[email protected] >,
David Richerby <[email protected] > wrote:
[to JS:]
>I think perpetual check is by far the least significant of your
>proposed changes.
Perpetual check is not, of course, a very well-defined
concept. If you merely mean repetition, then it's unclear, in
general, who deserves the [minor] win in such cases.
> [...] (This last
>part can be fixed easily by giving the bared king one more move to
>complete the exchange and declaring the game drawn in such cases.)
It would be fairer to give the bared king several moves;
eg W: Kc3 [to move], B: Kh1, Na1, Pa5.
>> [...] I seem to have come up with an original idea with _some_
>> promise.
>Absolutely.
Bared king and stalemate, "of course", used to be minor
wins in old versions of chess, so the idea is not exactly original!
I don't think it would make very much difference in practice;
most of my draws come from manifestly drawn positions where neither
side can make progress -- say, RPP each with all pawns well protected,
or opposite-coloured bishop endings, or level K&P endings where the
Ks block each other, etc., etc. No-one really wants to play those
out until the 50-move rule cuts in [or one of the players stumbles
into the third repetition] ....
--
Andy Walker
Nottingham
|
| | | | |
Date: 11 Jun 2008 12:45:18
From: David Richerby
Subject: Re: Does this so-called "draw problem" actually exist?
|
Quadibloc <[email protected] > wrote:
> David Richerby <[email protected]> wrote:
>> The other top-ten players are running at something closer to 60%
>> draws.
>
> Even that is 40% non-draws, rather than, say, 10% non-draws or even
> 20% non-draws. Drawing half the time may be mildly annoying, but
> indeed it doesn't qualify as dramatic evidence that chess is played
> out.
Yeah.
> I think it's reasonable to say that the situation some (perhaps
> unreasonable) people want is: White wins 41% of the time, Black wins
> 39% of the time, and it's a draw only 20% of the time. The idea
> being that with draws in the minority, AND with either player being
> able to win, there must be exciting, active play going on.
Sure, people might want that but it seems to come from the idea that a
draw is necessarily boring. They look at a draw as saying `Nobody
won' rather than `The result was too close to call in favour of one
player or the oher.' The interesting games are the close ones, not
the ones where the world champion checkmates a six year old in ten
moves.
Of course, everybody hates to watch games where the two players bash
out twenty book moves and agree a draw in a level but by no means
played-out position. But I wouldn't want to criticize those players
too harshly -- if there's nothing to play for and they're tired, I'd
rather see them not really play than make some colossal blunder.
(I'm not setting up a false dichotomy, here -- I'm just expressing my
preference between two of the available options.)
Even no-fight draws can be interesting. Look at game five of the 1993
match between Kasparov and Short[1]. Taken on its own, it's a pretty
dull game -- an eighteen move draw. But look at the context: Short,
at that stage, was 3.5-0.5 down after four games and has a migraine on
the day of game five, where he's playing black. He wheels out an
interesting novelty in the Nimzo-Indian and gets an effortless 18-move
draw. As I recall, Short used about twenty minutes on his clock and
Kasparov about an hour and a half.
I have to say, I'm coming round to your idea of tweaking the scoring
system to encourage more enterprising play. I'm not sure I like your
precise proposal but the general idea is an interesting one that is
certainly worthy of investigation.
Dave.
[1] http://www.chessgames.com/perl/chessgame?gid=1070677
--
David Richerby Permanent Simple Gnome (TM): it's like
www.chiark.greenend.org.uk/~davidr/ a smiling garden ornament but it has
no moving parts and it'll be there
for ever!
|
| | | | |
Date: 10 Jun 2008 21:23:47
From: David Kane
Subject: Re: Does this so-called "draw problem" actually exist?
|
"Quadibloc" <[email protected] > wrote in message
news:a00faaca-2705-4456-b14d-a974afa99bf2@j22g2000hsf.googlegroups.com...
>I think it's reasonable to say that the situation some (perhaps
>unreasonable) people want is: White wins 41% of the time, Black wins
>39% of the time, and it's a draw only 20% of the time. The idea being
>that with draws in the minority, AND with either player being able to
>win, there must be exciting, active play going on. If Black hardly
>ever wins, even if White does win a fair amount of the time, in high-
>level play, then there is already a... hint... of a problem.
I think you are missing part of the problem. Predictability (most
games will end as draws) is a factor. But more important is what
that situation does to the strategies that the players employ. When
the top players meet each other in tournaments, it is rare for them
to fully engage. It makes more sense to play half speed and hope
to pick up their points playing the bottom of the pack. There is no
return in looking for the full point against an Anand or a Kramnik.
If White won rather predictably then we wouldn't have that situation
(though it could harm fan interest for other reasons) because a draw
for White would be like a loss. Chess needs to create a situation
where we can count on a battle. I think I'm starting to appreciate
the merits of your proposal more - am still not sure whether the
valuing the various end conditions differently would make a
practical difference. Have you ever tried it?
|
| | | |
Date: 10 Jun 2008 13:35:54
From: David Richerby
Subject: Re: Does this so-called "draw problem" actually exist?
|
Quadibloc <[email protected] > wrote:
> When players are evenly matched, and are playing above a certain
> standard, not only do games have an expected outcome, but there is
> little that either player can do effectively to have a chance of
> varying from that outcome except through play sufficiently wild as
> to improve the result for his opponent, then we have the problem for
> which I don't have a simple name.
So, given that every game of chess between strong players is a near-
guaranteed draw, could you please explain how Kramnik (who is infamous
for playing dull, drawish chess) has won 52% of his games with white
since 2007? How Morozevich has won half his games with white and 43%
with black? How Topolov draws only 40% of his games?
Now, admittedly, in the last case, it would be unfair not to point out
that Topalov loses more games than the other people I've mentioned.
On the other hand, these stats do indicate that it's possible to be in
the top five in the world while still having a relatively high
proportion of decisive games.
Dave.
--
David Richerby Metal Goldfish (TM): it's like a fish
www.chiark.greenend.org.uk/~davidr/ that's made of steel!
|
| | |
Date: 09 Jun 2008 18:58:02
From: Quadibloc
Subject: Re: Does this so-called "draw problem" actually exist?
|
On Jun 9, 7:27 pm, Guy Macon <http://www.guymacon.com/ > wrote:
> Quadibloc wrote:
> >Given that chess games that epitomized the pre-Steinitz style were
> >much praised, and there was a lot of griping about what the Modern
> >style of Chess play did to the game from Steinitz onwards, I think
> >it's reasonable to say that this is "a problem", whether or not
> >solving it would automatically swell the membership rolls of the USCF.
>
> True, but I think it's also reasonable to say that this is *not*
> a problem. My beef is with those who seem to expect everyone to
> just assume without evidence that having fewer draws will swell
> the membership rolls of the USCF. In other words, they have
> failed to answer the question found in the subject of this thread;
> Does this so-called "draw problem" actually exist?
One of the other posts in this thread has pointed out that perhaps the
problem is misnamed. The problem isn't really too many draws, but not
enough exciting play.
These are two different problems - but they're easily confused,
because they are closely related.
If there is an expected outcome for a game - whether it's a safe win
for the first player (the way it was in Go before komidashi) - or a
draw which is hard to avoid unless the other player makes a mistake -
and getting a better outcome than the expected one is only a
possibility if you take big risks that usually lead to a worse
outcome, then playing defensively and coasting to the expected outcome
is the rational strategy.
It isn't so much the sight of 1/2 - 1/2 on the scoreboard that is
offending people. Rather, it's dull, defensive play that comes about
all too often because of the same characteristic of Chess that leads
to draws
John Savard
|
| | |
Date: 09 Jun 2008 12:20:42
From: Quadibloc
Subject: Re: Does this so-called "draw problem" actually exist?
|
On Jun 9, 11:48=A0am, Guy Macon <http://www.guymacon.com/ > wrote:
> Given the above facts, I find it difficult to assign the blame
> for the decline in US Chess to Steinitz (much too early) or to
> the rise of strong computers or the fall of the USSR (too late).
We can assign the credit for a certain peak in interest in Chess in
the U.S. to R. J. Fischer, just as we can assign the credit for a peak
in interest in classical piano music to Van Cliburn, or for a peak in
interest in science to Sputnik.
Given that chess games that epitomized the pre-Steinitz style were
much praised, and there was a lot of griping about what the Modern
style of Chess play did to the game from Steinitz onwards, I think
it's reasonable to say that this is "a problem", whether or not
solving it would automatically swell the membership rolls of the USCF.
A big enough problem to justify doing something to Chess? First, we
would have to settle the question of doing _what_ to Chess! Is it even
clear there's a way to modify Chess so that it becomes rational to
play in a more entertaining style as a way of winning?
John Savard
|
| | | |
Date: 10 Jun 2008 01:27:42
From: Guy Macon
Subject: Re: Does this so-called "draw problem" actually exist?
|
Quadibloc wrote:
>Given that chess games that epitomized the pre-Steinitz style were
>much praised, and there was a lot of griping about what the Modern
>style of Chess play did to the game from Steinitz onwards, I think
>it's reasonable to say that this is "a problem", whether or not
>solving it would automatically swell the membership rolls of the USCF.
True, but I think it's also reasonable to say that this is *not*
a problem. My beef is with those who seem to expect everyone to
just assume without evidence that having fewer draws will swell
the membership rolls of the USCF. In other words, they have
failed to answer the question found in the subject of this thread;
Does this so-called "draw problem" actually exist?
--
Guy Macon
<http://www.guymacon.com/ >
|
| | | | |
Date: 10 Jun 2008 06:16:38
From: Quadibloc
Subject: Re: Does this so-called "draw problem" actually exist?
|
On Jun 10, 6:35 am, David Richerby <[email protected] >
wrote:
> Quadibloc <[email protected]> wrote:
> > When players are evenly matched, and are playing above a certain
> > standard, not only do games have an expected outcome, but there is
> > little that either player can do effectively to have a chance of
> > varying from that outcome except through play sufficiently wild as
> > to improve the result for his opponent, then we have the problem for
> > which I don't have a simple name.
>
> So, given that every game of chess between strong players is a near-
> guaranteed draw, could you please explain how Kramnik (who is infamous
> for playing dull, drawish chess) has won 52% of his games with white
> since 2007? How Morozevich has won half his games with white and 43%
> with black? How Topolov draws only 40% of his games?
>
> Now, admittedly, in the last case, it would be unfair not to point out
> that Topalov loses more games than the other people I've mentioned.
> On the other hand, these stats do indicate that it's possible to be in
> the top five in the world while still having a relatively high
> proportion of decisive games.
Kramnik, at least, I can explain; he plays sufficiently better Chess
than virtually everyone else, with the possible exception of Anand
Viswarathan, that he should be able to win a few times.
Well, then, perhaps we can wait until Chess is a little closer to
being played out before considering drastic action. But if enough
other players are having more draws than these examples, it may still
be useful to explore alternatives to find ways to reduce the value of
"drastic".
John Savard
|
| | | | | |
Date: 10 Jun 2008 17:40:31
From: David Richerby
Subject: Re: Does this so-called "draw problem" actually exist?
|
Quadibloc <[email protected] > wrote:
> David Richerby <[email protected]> wrote:
>> So, given that every game of chess between strong players is a near-
>> guaranteed draw, could you please explain how Kramnik (who is infamous
>> for playing dull, drawish chess) has won 52% of his games with white
>> since 2007? How Morozevich has won half his games with white and 43%
>> with black? How Topolov draws only 40% of his games?
>
> Kramnik, at least, I can explain; he plays sufficiently better Chess
> than virtually everyone else, with the possible exception of Anand
> Viswarathan, that he should be able to win a few times.
Yes but Morozevich is better than everyone else apart from Kramnik and
Anand. And Topolov is better than everyone else apart from Kramnik,
Anand and Moro. At least, taking `better' to mean `higher rating',
i.e., `more successful'.
> Well, then, perhaps we can wait until Chess is a little closer to
> being played out before considering drastic action. But if enough
> other players are having more draws than these examples, it may
> still be useful to explore alternatives to find ways to reduce the
> value of "drastic".
The other top-ten players are running at something closer to 60%
draws.
Dave.
--
David Richerby Natural Atlas (TM): it's like a map of
www.chiark.greenend.org.uk/~davidr/ the world but it's completely natural!
|
| | | |
Date: 09 Jun 2008 21:40:49
From: David Richerby
Subject: Re: Does this so-called "draw problem" actually exist?
|
Quadibloc <[email protected] > wrote:
> Given that chess games that epitomized the pre-Steinitz style were
> much praised, and there was a lot of griping about what the Modern
> style of Chess play did to the game from Steinitz onwards, I think
> it's reasonable to say that this is "a problem"
Or, if you prefer, the "problem" is that so many of the games of the
pre-Steinitz era consist of a few dubious-to-awful moves followed by
a pretty combination while, on the other hand, plenty of games of
Alekhine, Bronstein, Keres, Tal, Korchnoi, Kasparov, Anand, Morozevich
and others (apologies if I missed your personal favourite) have been
much-praised.
> A big enough problem to justify doing something to Chess? First, we
> would have to settle the question of doing _what_ to Chess! Is it
> even clear there's a way to modify Chess so that it becomes rational
> to play in a more entertaining style as a way of winning?
Guy keeps suggesting that the way to influence people to play exciting
chess (by whatever definition of exciting you want to use) is to give
them incentives (tournament invites, prizes, and so on) to play that
way.
Dave.
--
David Richerby Cyber-Cheese (TM): it's like a lump
www.chiark.greenend.org.uk/~davidr/ of cheese that exists only in your
computer!
|
| | | | |
Date: 10 Jun 2008 01:40:52
From: Guy Macon
Subject: Re: Does this so-called "draw problem" actually exist?
|
David Richerby wrote:
>Guy keeps suggesting that the way to influence people to play exciting
>chess (by whatever definition of exciting you want to use) is to give
>them incentives (tournament invites, prizes, and so on) to play that
>way.
...with the conditional "assuming that whatever you define as
being exciting chess is desirable." I myself don't think that
it is. I think Chess isn't broken and doesn't need fixing.
Given that assumption, the question is whether to work towards
that goal with authoritarion, 1950s Soviet Union methods
(Fearless Leader decides what is best and imposes the new rules
on everyone) or with libertarian, 1770s USA methods (All proposed
solutions compete in the marketplace of ideas, and each person
who organizes a tournament, each person who chooses which
tournament to participate in, and each person having a casual
game all choose what game they want to play, what rules to
follow, what federation they want to join, etc.)
--
Guy Macon
<http://www.guymacon.com/ >
|
| |
Date: 06 Jun 2008 11:35:06
From: Quadibloc
Subject: Re: Does this so-called "draw problem" actually exist?
|
On Jun 1, 12:45=A0pm, Guy Macon <http://www.guymacon.com/ > wrote:
> I have seen plenty of people who have a personal preference that
> there be fewer draws, but that does not prove that there is a
> "draw problem" in the current rules. =A0Why is their preference
> more important than mine?
> I have seen plenty of claims that fewer draws will bring in more
> players or more paying spectators, but I haven't seen any evidence
> that these predictions will actually come to pass.
If something would bring in more paying spectators, then their views
would be important.
> In my opinion, the "solution" for those who prefer fewer draws
> at the top level is simple, straightforward, and far from being
> a new idea; simply don't invite players who draw "too often"
> (whatever that means to you), and do invite players who you
> consider to play "exciting" chess. "Problem" (if one actually
> exists) solved.
I think that players know that they're not likely to be invited
anywhere, unless they are "important", that is, highly rated. This is
achieved by winning, not losing. So a few extra invites isn't going to
motivate people to play recklessly.
As far as an individual tournament is concerned, yes, fewer draws
aren't going to help vastly. This is why I don't think the 1/3 - 1/3
draw rule will work.
Instead, a solution is needed with high visibility to improve the
reputation of chess as a whole; so whatever is done, it needs to be
effective in reducing draws during... the World Championship match.
And that's a tall order.
John Savard
|
| | |
Date: 06 Jun 2008 19:37:03
From: Quadibloc
Subject: Re: Does this so-called "draw problem" actually exist?
|
On Jun 6, 2:41 pm, David Richerby <[email protected] >
wrote:
> You're assuming that the draws are the problem. The problem is,
> surely, that Joe Public doesn't understand chess. Heck, even the
> chess enthusiast has a hard time understanding what goes on when two
> GMs sit down across a board.
That's the *main* problem.
But chess was a bit more popular back in the days of Anderssen and
Kieseritzky, of LaBourdonnais and MacDonnell, and chess enthusiasts
enjoyed it more. So I figure the people who complain about chess as it
is today want to bring those days back, but it is hard to figure out
how to do so.
I've tried to think of some new ideas that are a little different
from what has been suggested before, and which might come closer to
addressing the most serious of the addressable problems.
John Savard
|
| | | |
Date: 09 Jun 2008 15:58:01
From: David Richerby
Subject: Re: Does this so-called "draw problem" actually exist?
|
Quadibloc <[email protected] > wrote:
> But chess was a bit more popular back in the days of Anderssen and
> Kieseritzky, of LaBourdonnais and MacDonnell
[Citation needed]
> and chess enthusiasts enjoyed it more.
[Citation needed]
Dave.
--
David Richerby Beefy Monk (TM): it's like a man of
www.chiark.greenend.org.uk/~davidr/ God that's made from a cow!
|
| | |
Date: 06 Jun 2008 21:41:08
From: David Richerby
Subject: Re: Does this so-called "draw problem" actually exist?
|
Quadibloc <[email protected] > wrote:
> Guy Macon <http://www.guymacon.com/> wrote:
>> I have seen plenty of people who have a personal preference that
>> there be fewer draws, but that does not prove that there is a
>> "draw problem" in the current rules. Why is their preference
>> more important than mine?
>
>> I have seen plenty of claims that fewer draws will bring in more
>> players or more paying spectators, but I haven't seen any evidence
>> that these predictions will actually come to pass.
>
> If something would bring in more paying spectators, then their views
> would be important.
If increasing the number of paying spectators is the goal, I suggest
FIDE should abandon chess and set itself up as a football/American
football/baseball league.
> Instead, a solution is needed with high visibility to improve the
> reputation of chess as a whole; so whatever is done, it needs to be
> effective in reducing draws during... the World Championship match.
You're assuming that the draws are the problem. The problem is,
surely, that Joe Public doesn't understand chess. Heck, even the
chess enthusiast has a hard time understanding what goes on when two
GMs sit down across a board.
Dave.
--
David Richerby Mouldy Pointy-Haired Tool (TM): it's
www.chiark.greenend.org.uk/~davidr/ like a screwdriver that's completely
clueless but it's starting to grow
mushrooms!
|
| |
Date: 06 Jun 2008 10:43:34
From: Rich Hutnik
Subject: Re: Does this so-called "draw problem" actually exist?
|
On Jun 1, 2:45 pm, Guy Macon <http://www.guymacon.com/ > wrote:
> Ray Gordon wrote:
> >Tablebases are up to seven pieces now. I'd call that a threat.
>
> And you imagine that if the game was changed with a new promotion
> rule -- and the new rules became popular -- that nobody would
> create tablebases using the new rules?
>
> I have yet to see a convincing argument that this so-called
> "draw problem" in the rules of chess actually exists.
>
> I have seen plenty of people who have a personal preference that
> there be fewer draws, but that does not prove that there is a
> "draw problem" in the current rules. Why is their preference
> more important than mine?
Look at the fruit of the situation now. Is the percentages of draws
acceptable today? And, does this situation result in growth in people
playing? How about looking at the actual numbers and see. People
complain here about the U.S Chess Federation canning the magazine, and
they have financial issues. People think it is due to lack of
brilliance on the part of the leadership, as if chess is some sort of
perfection that shouldn't be looked at, only if the leadership did the
right things. Explain why poker isn't having a problem with growth
today, without saying, "Well people are idiots, and idiots like to
gamble". If you think people going to a game that is growing in
number of players are idiots, then explain how you would expect to get
growth? Call people idiots, or other words that mean the same thing,
or having that attitude, is done by people living in an intellectual
ghetto. The reality is, you aren't thinking growth, but rationalizing
why it isn't growing. Even thinking producing a U.S World Champ is
the magic bullet leads to this.
> I have seen plenty of claims that fewer draws will bring in more
> players or more paying spectators, but I haven't seen any evidence
> that these predictions will actually come to pass.
Does reducing the number of draws help or hurt interest? If it
generates more interest, then why shouldn't it be part of it. Let's
say chess has 10 issues total, and draws are 5% of the problem.
Explain the logic of going, "Well draws really don't have much of an
impact, so let's leave it as it!". If all the items on the list,
which are important to the growth of chess, and all need to be done,
all have this 5%, then you don't do ANYTHING. Such a mentality, to
me, is a rationalization for being lazy.
> I have seen evidence of top-level players drawing early without
> even trying to win (which is a distinct and separate issue from
> there supposedly being too many draws), but no evidence that this
> is caused by the present rules rather than, say, the reward system.
Then, you would argue there is a draw problem, just the reward system
is the reason why. This is a perfectly acceptable argument to make.
However, I have seen that not allowing players to ask for draws
reducing the number of draws by less than 10%. My question here is,
if too many draws IS an issue, why not do everything you can to
address them, from the compensation system to whatever else is needed,
even possible rules changes?
> Nobody who has created a rule change to address the supposed
> "draw problem" actually thinks that their new rule would prevail
> over the current rules in any sort of election/vote/poll.
>
> Nobody who has created a rule change to address the supposed
> "draw problem" actually thinks that their new rule would prevail
> over the current rules in any sort of free market such as
> announcing a tournament using the new rules and seeing how many
> players show up.
I will agree here that there is far too much dictatorial, from an
ivory tower, proposing of ideas, without actually them being taken up
and considered.
> I like new chess rules and variants on the game. I like reading
> about them, I like thinking about them, I like playtesting them,
> and I hope that someone can come up with something useful that
> becomes the new standard, as happened with "In Passing" pawn
> captures (and no, I am not going to start speaking french just
> so I can describe that rule...). But just because new rules
> appeal to me personally. that doesn't mean that there is anything
> wrong with the current rules.
>
> In my opinion, there is nothing wrong with the current rules.
> Chess is not broken, and does not require fixing.
And what is wrong with non-mad queens and non-mad bishops? What is
wrong with Shatranj? It was played by Persian royalty, and was a
perfectly acceptable game. Why make the queen and bishop mad, and
give them crazy mobility? There wasn't castling in the game either,
and pawns promoted to bishops only. Why change these rules? Maybe
the answer would be the same as why change chess rules, in order to
better serve the community of players.
> In my opinion, the "solution" for those who prefer fewer draws
> at the top level is simple, straightforward, and far from being
> a new idea; simply don't invite players who draw "too often"
> (whatever that means to you), and do invite players who you
> consider to play "exciting" chess. "Problem" (if one actually
> exists) solved.
Please repeat them here. Are you suggesting changes in the scoring
level? If so, that is fine. Any luck getting anyone to try them?
- Rich
|
| | |
Date: 09 Jun 2008 14:34:19
From: Quadibloc
Subject: Re: Does this so-called "draw problem" actually exist?
|
On Jun 9, 2:04=A0pm, "David Kane" <[email protected] > wrote:
> The biggest problem with your suggestion is that there
> are far simpler ways to reduce the return on playing
> drawishly. It makes sense to try them first.
As you pointed out, immediate incentives are needed. This is why I've
tried to avoid measures that place the player in a double-bind - such
as reducing the prize if a player draws too often, because if he isn't
the one in first place, he won't win anyways.
Even a measure like 1/3 - 1/3 for a draw - which has been tried - can
lead to a double-bind, because a draw is still better than losing, and
losing is the reward of pushing for the win when it hasn't been
earned.
That's why - through using komidashi for inspiration - I came to the
conclusion that a suggestion like mine may be what is needed. It is
necessarily artificial, in that Chess, unlike Go, doesn't provide
convenient hooks on which to peg degrees of victory. The idea is to
allow people to play aggressively, yet soundly - because they are able
to play aggressively to achieve a slight result on the scoreboard if
their positional play has achieved a slight advantage.
John Savard
|
| | |
Date: 09 Jun 2008 08:29:12
From: Quadibloc
Subject: Re: Does this so-called "draw problem" actually exist?
|
On Jun 9, 8:12=A0am, David Richerby <[email protected] >
wrote:
> The point of komidashi in Go is *not* to reduce draws. =A0As a side
> effect half-point komi makes draws impossible, but the purpose of
> introducing the system was that the first player (black, in Go) has
> the advantage of initiative in the initial position. =A0Between strong
> players, this meant that Black could win by a couple of points just by
> playing very defensively. =A0The point of komi was to force Black to
> play a more interesting game by turning those narrow wins into narrow
> losses.
And that's _precisely_ the object of my Dynamic Scoring scheme. In
chess, between strong players, Black - the second player - can draw by
playing very defensively. White's advantage of initiative isn't
sufficient to win, and trying to squeeze a win out of it leads to too
much risk of losing.
Making smaller advantages show up on the scoreboard, by adding
stalemate, bare King, and even perpetual check as "wins" means that
aggressive play which is still sound - which doesn't aim at a result
too far out of reach, like checkmate - is rewarded. Giving an
advantage to Black which is proportionately larger for the smaller
victories motivates even Black to try for a modest win - and motivates
White to play more aggressively, to bring the game into an area where
the wins and losses are awarded scores which don't favor Black as
heavily.
It's easy enough to conflate 'interesting play' and 'fewer draws', for
a number of reasons, but you are right the distinction needs to be
made.
John Savard
|
| | |
Date: 09 Jun 2008 07:11:02
From: Quadibloc
Subject: Re: Does this so-called "draw problem" actually exist?
|
On Jun 9, 4:21 am, David Richerby <[email protected] >
wrote:
> Half a dozen people, who imagine that everybody thinks like them,
> complaining about draws on an internet newsgroup constitutes a reason
> to reform chess?
Lots of people have been griping about draws, and about the lack of
obviously dramatic attacks in chess play, *ever since Steinitz*. But
this is _only_ a reason to reform chess if there were a way to do so
that
a) addresses the problem effectively, and
b) doesn't ruin Chess.
They solved the draw problem in Shogi with drops - but it's now
(almost) a game played only by children in Japan.
Fischerrandom - or Chess 960 - shakes up the openings, but would be
expected to do little to adjust the character of the game, and it does
make the board ugly and asymmetrical.
The two-move restriction, and its successors, were accepted as a
necessity in Checkers - in deeper trouble than Chess is - but lots of
people find it irksome, and Chess players will simply not stand for
such a thing.
Reform Chess? Accept this or that solution - even one of *mine*? No, I
won't urge that now - it's too soon. But - start putting some thought
into what an effective and palatable reform for Chess might be - yes;
I advocate that.
Yes, people like myself who have their own alternate versions of Chess
might well be dismissed as cranks; after all, even Seirawan and
Capablanca ended up being ignored when they suggested changed versions
of Chess. But if the world's foremost Grandmasters started *really
thinking* about the best way to change the rules, they ought to be
able to come up with a reform far better than anything I, or any other
individual, can come up with on my own.
John Savard
|
| | |
Date: 09 Jun 2008 07:01:11
From: Quadibloc
Subject: Re: Does this so-called "draw problem" actually exist?
|
On Jun 8, 11:41 pm, "David Kane" <[email protected] > wrote:
> That was a very long time ago. For decades,
> Black has played to draw, and White has played
> for a small advantage that also often draws.
Basically, ever since Steinitz.
> Given that reality, the 8-move draw is not
> *cheating* - it's an extremely minor variation
> of *normal*. That's why it will never be
> penalized.
Draws involving apparent collusion have already been penalized. But,
because, as you say, it's a small variation from "normal", penalizing
that kind of draw *won't solve the problem*.
My suggestion may not work - but it offers points on the scoreboard
for Black if Black can gain a slight advantage; and White can put
points on the board with a slight advantage as well, but the scoring
puts White at a disadvantage so as to spur him to seek a larger margin
of victory, commensurate with his first move advantage.
So it's a start - something that should be examined, and perhaps
revised quite extensively in its details - towards a scheme that
directly rewards aggressive play, rather than belatedly.
John Savard
|
| | | |
Date: 09 Jun 2008 16:55:52
From: Chess One
Subject: Re: Does this so-called "draw problem" actually exist?
|
Problem... for whom?
For whom does the 'draw problem' exist?
Who has a problem with it?
---
Those are the rather absent Subjects of the title to this thread.
Does the problem exist for players who cannot even understand the chess
content which brings about the draw? Who thinks that?
Or is it an abstracted one about the result of play ending in draws as if
the process was not understood but the result resented?
---
After so many threads with no subject [matter!] then it all seems rather
hypothetical. Who here can say what it means to them if top GMs draw, no-one
so far.
As for changing the rules of chess to accomnmodate this rather abstract
issue of top GM play to which no-one can quite say what it matters to them
personally - a second question is if 99.999% of actual chess players
consider draws 'a problem'? That is, their own draws.
I would generate some interest to read who thinks what and why.
Phil Innes
|
| | | |
Date: 09 Jun 2008 13:04:01
From: David Kane
Subject: Re: Does this so-called "draw problem" actually exist?
|
"Quadibloc" <[email protected] > wrote in message
news:f9a0568b-4863-411c-a805-10331fea5d2b@c58g2000hsc.googlegroups.com...
> On Jun 8, 11:41 pm, "David Kane" <[email protected]> wrote:
>
>> That was a very long time ago. For decades,
>> Black has played to draw, and White has played
>> for a small advantage that also often draws.
>
> Basically, ever since Steinitz.
>
>> Given that reality, the 8-move draw is not
>> *cheating* - it's an extremely minor variation
>> of *normal*. That's why it will never be
>> penalized.
>
> Draws involving apparent collusion have already been penalized. But,
> because, as you say, it's a small variation from "normal", penalizing
> that kind of draw *won't solve the problem*.
>
> My suggestion may not work - but it offers points on the scoreboard
> for Black if Black can gain a slight advantage; and White can put
> points on the board with a slight advantage as well, but the scoring
> puts White at a disadvantage so as to spur him to seek a larger margin
> of victory, commensurate with his first move advantage.
>
> So it's a start - something that should be examined, and perhaps
> revised quite extensively in its details - towards a scheme that
> directly rewards aggressive play, rather than belatedly.
I was not arguing against your proposal, though
I think many of your assertions are implausible
and in need of proof. While the idea is intriguing,
it doesn't sound like you've thought through the
details and, to me, it comes across as very
artificial.
The biggest problem with your suggestion is that there
are far simpler ways to reduce the return on playing
drawishly. It makes sense to try them first.
> John Savard
|
| | |
Date: 09 Jun 2008 01:15:28
From: help bot
Subject: Re: Does this so-called "draw problem" actually exist?
|
David Kane wrote:
> > As for chess tradition, there seem to be
> > enough "tradition" in favor of contesting games
> > that they once required draws -- even fighting
> > ones -- to be replayed. So much for that sort
> > of blather... .
> That was a very long time ago.
You want it both ways? Tradition when it suits
your whim, but dismissal of tradition when it
doesn't? Sorry, but you have to /choose/ one or
the other.
> For decades,
> Black has played to draw, and White has played
> for a small advantage that also often draws.
Then I would suggest not inviting him back.
You can get somebody else; try Mr. Weiss, or
Mr. Greene or Mr. Browne.
> Given that reality, the 8-move draw is not
> *cheating* - it's an extremely minor variation
> of *normal*. That's why it will never be
> penalized.
It is cheating /by definition/ if the game is
played under USCF's rules. We can't dictate
what they do /over there/, so why worry about
it.
> Drawishness can be dealt with by supplying
> *immediate* incentives for each side to play
> for more
Like maybe free chocolates at the board?
>- some anonymous bozo's belated
> disapproval not qualifying.
You're ad hominizing; have you seen a doctor
about this problem? A psychologist, I mean.
-- help bot
|
| | | |
Date: 09 Jun 2008 13:13:23
From: David Kane
Subject: Re: Does this so-called "draw problem" actually exist?
|
"help bot" <[email protected] > wrote in message
news:3f5432e8-02df-4549-a031-3f93dba5b47c@s50g2000hsb.googlegroups.com...
>
>
> David Kane wrote:
>
>> > As for chess tradition, there seem to be
>> > enough "tradition" in favor of contesting games
>> > that they once required draws -- even fighting
>> > ones -- to be replayed. So much for that sort
>> > of blather... .
>
>> That was a very long time ago.
>
> You want it both ways? Tradition when it suits
> your whim, but dismissal of tradition when it
> doesn't? Sorry, but you have to /choose/ one or
> the other.
>
>
>> For decades,
>> Black has played to draw, and White has played
>> for a small advantage that also often draws.
>
> Then I would suggest not inviting him back.
> You can get somebody else; try Mr. Weiss, or
> Mr. Greene or Mr. Browne.
The point is that nearly all players play that
way, because playing that way is optimal
given chess' inherent nature and the idiotic
scoring method that tournament organizers
use.
>
>
>> Given that reality, the 8-move draw is not
>> *cheating* - it's an extremely minor variation
>> of *normal*. That's why it will never be
>> penalized.
>
> It is cheating /by definition/ if the game is
> played under USCF's rules. We can't dictate
> what they do /over there/, so why worry about
> it.
It's not a *real* rule. It's just a bone
tossed to the mindless - who don't want to
face that it is *their* love affair with the draw
that produces these monstrosities.
>
>
>> Drawishness can be dealt with by supplying
>> *immediate* incentives for each side to play
>> for more
>
> Like maybe free chocolates at the board?
How about more points for wins, leading to
higher finishes, bigger prizes etc.?
>
>
>>- some anonymous bozo's belated
>> disapproval not qualifying.
>
> You're ad hominizing; have you seen a doctor
> about this problem? A psychologist, I mean.
>
Cite an example of a GM who
wanted to draw but claimed he was afraid
to because he was worried about
your opinion (or the anonymous
bozo of your choice)
|
| | |
Date: 08 Jun 2008 21:08:44
From: help bot
Subject: Re: Does this so-called "draw problem" actually exist?
|
On Jun 8, 10:02 pm, "David Kane" <[email protected] > wrote:
> > I think it is self-evident that the problem with
> > top players agreeing to *uncontested* draws
> > can be addressed as suggested above: by
> > not inviting abusers back again. This could
> > also draw in a bit of fresh blood, while ousting
> > the stale and the incurably corrupted.
> It is not that there are a few, rare,
> uncontested games.
Obviously. If these abuses were few and rare,
there would be no serious problem.
> The problem
> is that the scoring system reduces the
> contestedness of every game
Assertion noted. How about some evidence
to back it up?
At the lower levels, nobody seems to have
any problem with the scoring system doing
what you assert it does. (If you wish to revise
your assertion to apply only at the higher
levels, please tell us.)
Here is a problem I have noticed with the
scoring system as it applies to Swiss system
tourneys: in a one-day, four-round event, there
can be a lot of ties. In a two-day, five-round
event, there still can be ties, and there seems
to be a lot of uncontested draws in the final
rounds, designed to manipulate prize monies.
This is only *partly* due to the scoring of
games; anothe part is the relatively low
number of rounds; and another part is the
fact that players are deliberately agreeing
to draws to manipulate the prize money.
Note: that makes three separate issues.
> of which
> the 8-movers are just an extreme example.
> And it has been all but impossible to
> eliminate even the 8-movers by the
> method supplied.
If by "the method supplied" you mean not
inviting abusers back, it has yet to become
fashionable. There can be no question as
to whether it works, for just as we have seen
with lions culling the sickly from the herd, it
boils down to simple math. The only way it
could conceivably fail is if the fresh blood
brought in to replace the worst offenders
turned out to be, on average, just as bad!
> > It might surprise you to learn that it does
> > not require massive carnage in order for such
> > a strategy to be effective; in reality, just a few
> > players would need to be sanctioned in order
> > to keep the rest in line, and to effect a big
> > change in the way that business is done. It's
> > akin to the way that predators pick off the
> > sick, the slow, and the weak, thus improving
> > the prey species, over time.
> In the real world, nobody takes these sanctions
> seriously. They are completely ineffective
> because they go against decades and decades of
> chess tradition.
Good! If the worst offenders do not take the
threat seriously, it will be that much easier to
pick them off, one by one. Nothing is worse
than a cheater who knows you are trying to
catch him cheating; he will of course, stop
cheating until he believes the coast is clear,
then revert back to his normal self.
This reminds me of the sinking of the
Lusitania. Ads were prominently placed in
the newspaper warning passengers that a
state of war existed, and that British ships
could and would be sunk on sight. Seeing
this, the passengers panicked, but the good
captain reassured them that the ship was
too fast to be targeted by German subs. Of
course, he did not consider the possibility
of a sub just sitting there, waiting for him.
As for chess tradition, there seem to be
enough "tradition" in favor of contesting games
that they once required draws -- even fighting
ones -- to be replayed. So much for that sort
of blather... .
-- help bot
|
| | | |
Date: 08 Jun 2008 22:41:09
From: David Kane
Subject: Re: Does this so-called "draw problem" actually exist?
|
"help bot" <[email protected] > wrote in message
news:b026f65f-7326-4396-8f9f-51056f578503@z72g2000hsb.googlegroups.com...
>
> As for chess tradition, there seem to be
> enough "tradition" in favor of contesting games
> that they once required draws -- even fighting
> ones -- to be replayed. So much for that sort
> of blather... .
>
That was a very long time ago. For decades,
Black has played to draw, and White has played
for a small advantage that also often draws.
Given that reality, the 8-move draw is not
*cheating* - it's an extremely minor variation
of *normal*. That's why it will never be
penalized.
Drawishness can be dealt with by supplying
*immediate* incentives for each side to play
for more - some anonymous bozo's belated
disapproval not qualifying.
|
| | |
Date: 08 Jun 2008 20:46:13
From: help bot
Subject: Re: Does this so-called "draw problem" actually exist?
|
On Jun 8, 11:26 am, Quadibloc <[email protected] > wrote:
> No, I don't desire reckless attacks instead of brilliancies, but I
> think that brilliancy prizes, in themselves, don't produce
> brilliancies in great number over and above those that occur naturally
> in the course of trying to _win_.
There is a big difference between "brilliant" play
and trying to win. A while back, some fellow
showed up at our local club and we were informed
that he had beaten some famous GMs. After the
tourney was over, I went home and looked up some
of those games, and indeed, he "won", but only by
the opposite of brilliancy-- he apparently pressured
his opponents into bungling their winning positions
against him. Now, where I see an overlap between
what are perceived as "brilliant" games and what
players do in trying to win, is with GM Tal; he did
both at the same time-- inducing blunders and
attacking brilliantly; but he was a rare exception.
> For a number of reasons, going on the offensive involves risk. So one
> begins one's attack *when* one has accumulated sufficient positional
> advantage to make it sound - and if one needs enough advantage to
> obtain checkmate to make it worthwhile, then, in most games, "when" is
> "never" and a draw ensues. Hence, my scheme.
Nice try. You simply messed up by confounding
positional play as the opposite of aggression; the
reality is that there can be aggressive positional
players, and there can be defensive ones, since
these are two very different concepts. I think the
aggressive positional players have been very
successful, while the defensive ones have
difficulties in winning /tournaments/.
In reality, humans do not play perfect chess, so
what you have done above is build a straw man,
or maybe just a false dilemma. There is "risk" in
playing defensively as well, and often as not, an
unsound attack may work because the defender
must successfully anticipate every conceivable
attack, while the attacker need only find *one*
winning shot. Anyway, that's what the top GMs
say, and it seems to apply at the lower levels as
well.
The fact remains that the problem in top-level
chess is that the players themselves are being
rewarded well enough for drawing, and any
rules forbidding prearranged outcomes are not
being enforced. Thus, the simplest solution is
for the organizers to stop inviting back those
players who are the worst offenders. (This is
because we already know they aren't going to
start enforcing the rules against uncontested
draws-- it seems they haven't the guts.)
One problem would be loud screams of pain
from the first "victims". But if this is done in a
fair manner, those "victims" will just point
themselves out as the root of the problem we
are trying to solve; their recorded draws will
serve as the arbiter-- were they or were they
not "contested"?
I see another problem here, in that while all
(chess playing) men are created equal, some
seem to be a little more equal than others. As
we have seen in certain world championship
cycles, sometimes the rules are bent -- or
discarded altogether -- when the powers that
be want a certain player to get to the top. It's
not a pretty picture, but even acknowledging
such problems, the fact remains that the
draw problem would be reduced, even if just
those offenders who are disliked were not
invited back again.
They say the threat is stronger than the
execution; well then, the mere threat of not
being invited back should have an impact, in
itself. I know that if they threatened to keep
me out of the Midwest Duffer's Championship
next year, I would certainly think twice before
agreeing to draw in the last round with Rob
Mitchell on board one, to split the trophy.
-- help bot
|
| | |
Date: 08 Jun 2008 08:31:36
From: Quadibloc
Subject: Re: Does this so-called "draw problem" actually exist?
|
On Jun 8, 7:43 am, help bot <[email protected] > wrote:
> It might surprise you to learn that it does
> not require massive carnage in order for such
> a strategy to be effective;
If one's only problem is with the few extra draws caused by lazy or
dishonest players, that would work. But if one's goal is to make Chess
look like it did before Steinitz, achieving that goal would not be
possible without "massive carnage" because then one would be trying to
get players to play in a way that has been shown to be unsound and
unworkable.
Making aggressive play rational once again requires more, and measures
with the double-bind problem are foredoomed to be of limited
effectiveness at most.
John Savard
|
| | |
Date: 08 Jun 2008 08:26:15
From: Quadibloc
Subject: Re: Does this so-called "draw problem" actually exist?
|
On Jun 8, 7:25 am, help bot <[email protected] > wrote:
> On Jun 7, 9:09 pm, Quadibloc <[email protected]> wrote:
>
> > Also, note that draws _in themselves_ are not necessarily the whole
> > problem; defensive, positional play instead of aggressive, attacking
> > play might be the "real" problem
>
> Positional play is not in any way associated with
> either defense or offense; nor is recklessness
> inseparably connected to attack or defense.
>
> Note that if wild, attacking play is desired, there
> are theme tournaments in which, say, the King's
> Gambit or the Smith Morra Gambit could be
> required; those who are unfamiliar with the theory
> would invariably land in treacherous waters.
>
> Monies could be reallocated from 1st, 2nd, etc.
> to 1st brilliancy, 2nd brilliancy, best attack, most
> reckless attack, 2nd most reckless attack, etc.--
> if indeed that is what you desire.
No, I don't desire reckless attacks instead of brilliancies, but I
think that brilliancy prizes, in themselves, don't produce
brilliancies in great number over and above those that occur naturally
in the course of trying to _win_.
For a number of reasons, going on the offensive involves risk. So one
begins one's attack *when* one has accumulated sufficient positional
advantage to make it sound - and if one needs enough advantage to
obtain checkmate to make it worthwhile, then, in most games, "when" is
"never" and a draw ensues. Hence, my scheme.
John Savard
|
| | |
Date: 08 Jun 2008 06:43:34
From: help bot
Subject: Re: Does this so-called "draw problem" actually exist?
|
On Jun 7, 9:29 pm, "David Kane" <[email protected] > wrote:
> > No. I am saying that *IF* you believe that tournament chess has
> > too many draws, simply don't invite players who (you believe) draw
> > "too often" (whatever that means to you), and do invite players
> > who play the way you want them to play.
> That "solution" would only solve a problem in which draws were
> the result of some aberrant playing style. In fact, playing in a way
> that produces lots of draws is optimal. It's a basic flaw in chess
> tournament design that has nothing to do with a player's individual
> preference.
You are sadly mistaken.
I have seen two primary causes of draws:
1) Incompetence, whereby winning positions
are routinely bungled by weak players;
2) Prize money manipulation, whereby the
leading scorers conspired to divvy up prizes
to suit their whims.
The only other time I have noticed a problem
with lots of draws occurring is when all the
players were of /very similar strength/, so
closely matched that even after hard struggles,
the outcomes included a significant number of
draws; even here, the tally was invariably
tainted by the addition of a few uncontested
draws, added into the mix.
I think it is self-evident that the problem with
top players agreeing to *uncontested* draws
can be addressed as suggested above: by
not inviting abusers back again. This could
also draw in a bit of fresh blood, while ousting
the stale and the incurably corrupted.
It might surprise you to learn that it does
not require massive carnage in order for such
a strategy to be effective; in reality, just a few
players would need to be sanctioned in order
to keep the rest in line, and to effect a big
change in the way that business is done. It's
akin to the way that predators pick off the
sick, the slow, and the weak, thus improving
the prey species, over time.
-- help bot
|
| | | |
Date: 08 Jun 2008 19:02:30
From: David Kane
Subject: Re: Does this so-called "draw problem" actually exist?
|
"help bot" <[email protected] > wrote in message
news:60d9241f-4a94-45d4-b74b-172e27d37d1d@c65g2000hsa.googlegroups.com...
>
> I think it is self-evident that the problem with
> top players agreeing to *uncontested* draws
> can be addressed as suggested above: by
> not inviting abusers back again. This could
> also draw in a bit of fresh blood, while ousting
> the stale and the incurably corrupted.
It is not that there are a few, rare,
uncontested games. The problem
is that the scoring system reduces the
contestedness of every game - of which
the 8-movers are just an extreme example.
And it has been all but impossible to
eliminate even the 8-movers by the
method supplied.
> It might surprise you to learn that it does
> not require massive carnage in order for such
> a strategy to be effective; in reality, just a few
> players would need to be sanctioned in order
> to keep the rest in line, and to effect a big
> change in the way that business is done. It's
> akin to the way that predators pick off the
> sick, the slow, and the weak, thus improving
> the prey species, over time.
In the real world, nobody takes these sanctions
seriously. They are completely ineffective
because they go against decades and decades of
chess tradition.
|
| | |
Date: 08 Jun 2008 06:25:54
From: help bot
Subject: Re: Does this so-called "draw problem" actually exist?
|
On Jun 7, 9:09 pm, Quadibloc <[email protected] > wrote:
> Also, note that draws _in themselves_ are not necessarily the whole
> problem; defensive, positional play instead of aggressive, attacking
> play might be the "real" problem
Positional play is not in any way associated with
either defense or offense; nor is recklessness
inseparably connected to attack or defense.
Note that if wild, attacking play is desired, there
are theme tournaments in which, say, the King's
Gambit or the Smith Morra Gambit could be
required; those who are unfamiliar with the theory
would invariably land in treacherous waters.
Monies could be reallocated from 1st, 2nd, etc.
to 1st brilliancy, 2nd brilliancy, best attack, most
reckless attack, 2nd most reckless attack, etc.--
if indeed that is what you desire.
-- help bot
|
| | |
Date: 08 Jun 2008 04:07:29
From: Quadibloc
Subject: Re: Does this so-called "draw problem" actually exist?
|
On Jun 7, 10:16 pm, "David Kane" <[email protected] > wrote:
> "Quadibloc" <[email protected]> wrote in message
> news:3e289698-ca51-41b8-9a21-203e1b4073c7@d77g2000hsb.googlegroups.com...
> > On Jun 7, 7:29 pm, "David Kane" <[email protected]> wrote:
>
> >> That "solution" would only solve a problem in which draws were
> >> the result of some aberrant playing style. In fact, playing in a way
> >> that produces lots of draws is optimal. It's a basic flaw in chess
> >> tournament design that has nothing to do with a player's individual
> >> preference.
>
> > I agree with that except for one thing: the flaw is not in tournament
> > design. Playing in a way that produces lots of draws is optimal during
> > matches as well.
>
> Matches are so rare these days that focusing on them doesn't make
> sense. Personally, I am much more tolerant of draws in that setting.
> Sure the draws are generally forgettable but the decisive games aren't.
But are you reading about them in the newspaper, or are you in the
audience? What if the match goes on for 60 games instead of 24
(presumably as the result of an attempt to reduce draws in a misguided
fashion that failed)?
> I am unconvinced that your approach would work, and because
> the scoring method requires relearning endgame theory, I don't
> see any way to even test whether it would help.
It's true that komidashi didn't yield its full benefits until Go
Seigen advanced the understanding of Go.
But I have anticipated this objection to some extent. Stalemate is
awarded, roughly, a 3/5 - 2/5 split of points. So winding up with a
stalemate when you could have had a checkmate is still heavily
penalized.
This means that just about everything in existing endgame theory
remains true, there's just a new level of endgame theory added on.
> Aside from your proposal, the most reliable approach to reduce
> draws in matches would be to do things that would decrease
> the quality of the games - faster time control, fewer rest days etc.
> While I think we could move along that axis a little, it's not
> a promising option because almost everybody does want
> the World Championship to be decided by high quality
> matches.
I think that this has already been done to an extent - faster time
controls in all but the highest-level events are already a reality -
I'm opposed to that option. If people want mediocre Chess, they can
find that at home.
My goal, which may well be a vain one, is to have the world's top
Grandmasters, when playing with generous time controls, start off with
good, sound positional play of excellent quality... but, because of
the multiple victory conditions, and the incentives to take some
controlled risks, to end up most of the time finding a way to have a
sufficient advantage to launch a sound attack that leads to the value
of the advantage being realized in the score.
So the Grandmaster entertains us by sacrificing his Queen and both
Rooks, and the fact that his reward is the pittance accorded a
perpetual check, rather than the full point of checkmate, hardly
excites our sympathies, cruel and selfish spectators that we are.
John Savard
|
| | |
Date: 07 Jun 2008 18:45:29
From: Quadibloc
Subject: Re: Does this so-called "draw problem" actually exist?
|
On Jun 7, 7:29 pm, "David Kane" <[email protected] > wrote:
> That "solution" would only solve a problem in which draws were
> the result of some aberrant playing style. In fact, playing in a way
> that produces lots of draws is optimal. It's a basic flaw in chess
> tournament design that has nothing to do with a player's individual
> preference.
I agree with that except for one thing: the flaw is not in tournament
design. Playing in a way that produces lots of draws is optimal during
matches as well.
Playing in such a way was _found_ to be optimal for Chess by Steinitz.
Even if my proposal to change scoring would be ineffective or
unworkable, this is, as you've noted, where one has to tinker.
John Savard
|
| | | |
Date: 07 Jun 2008 21:16:09
From: David Kane
Subject: Re: Does this so-called "draw problem" actually exist?
|
"Quadibloc" <[email protected] > wrote in message
news:3e289698-ca51-41b8-9a21-203e1b4073c7@d77g2000hsb.googlegroups.com...
> On Jun 7, 7:29 pm, "David Kane" <[email protected]> wrote:
>
>> That "solution" would only solve a problem in which draws were
>> the result of some aberrant playing style. In fact, playing in a way
>> that produces lots of draws is optimal. It's a basic flaw in chess
>> tournament design that has nothing to do with a player's individual
>> preference.
>
> I agree with that except for one thing: the flaw is not in tournament
> design. Playing in a way that produces lots of draws is optimal during
> matches as well.
Matches are so rare these days that focusing on them doesn't make
sense. Personally, I am much more tolerant of draws in that setting.
Sure the draws are generally forgettable but the decisive games aren't.
I am unconvinced that your approach would work, and because
the scoring method requires relearning endgame theory, I don't
see any way to even test whether it would help.
Aside from your proposal, the most reliable approach to reduce
draws in matches would be to do things that would decrease
the quality of the games - faster time control, fewer rest days etc.
While I think we could move along that axis a little, it's not
a promising option because almost everybody does want
the World Championship to be decided by high quality
matches.
One thing to keep in mind is that match players will
still spend most of their time playing tournaments -
so the players will be learning a more combative style
for tournaments, and that will likely carry over into
matches.
>
> Playing in such a way was _found_ to be optimal for Chess by Steinitz.
> Even if my proposal to change scoring would be ineffective or
> unworkable, this is, as you've noted, where one has to tinker.
>
> John Savard
|
| | |
Date: 07 Jun 2008 18:09:01
From: Quadibloc
Subject: Re: Does this so-called "draw problem" actually exist?
|
On Jun 7, 6:25 pm, Guy Macon <http://www.guymacon.com/ > wrote:
> Rich Hutnik wrote:
>
> >Look at the fruit of the situation now. Is the percentages of draws
> >acceptable today?
>
> Yes. Perfectly acceptable. Chess is not broken, and does not
> require fixing.
>
> >And, does this situation result in growth in people
> >playing? How about looking at the actual numbers and see.
>
> And you expect us all to accept without evidence your assumption
> that lack of growth of chess is the effect and "too many" draws
> are the cause? Got any evidence to back up that assertion?
I can understand your frustration, but some things are well-known
enough they can be accepted without trying to go beyond anecdotal
evidence.
People have been whining about what has become of Chess ever since
Steinitz. It *is* reasonable to believe that Chess would be perceived
as "more exciting" if we had more games like Anderssen-Kieseritzky
nowadays. One can note the existence of griping about draws in Chess,
the efforts of the Japanese to reduce draws in Go, and even the
overtime rules in ice hockey (i.e., "sudden death overtime") as
evidence of a general desire for games to produce a conclusive
resolution.
The trouble is, though, that it is a fact that with the additional
knowledge about Chess that Steinitz provided, it would be irrational
for a chessplayer seeking to win to play in that kind of dramatic
attacking style; he would give more in openings to his opponent than
he would gain for himself in winning chances.
1/3 - 1/3 for a draw could work in tournaments. But the general
public's image of chess would be most profoundly affected by the World
Championship match, which is a match, not a tournament.
And other proposals, such as diminishing the prize fund if there are
too many draws, or fiddling about with time controls, create a double-
bind, in which the strongest pull is still the need to be a winner
rather than a loser. (Even 1/3 - 1/3 in tournaments is not entirely
free of the double-bind effect.)
There are more debatable factors, of course. It might be that a
successful strategy for reducing draws would have only a minor effect,
compared to that of the loss of Soviet state support for Chess, or the
emergence of so many rival distractions these days.
Also, note that draws _in themselves_ are not necessarily the whole
problem; defensive, positional play instead of aggressive, attacking
play might be the "real" problem, but some measures to reduce draws
would also address that problem by giving players something to fight
over more often. Giving players a tiny fraction of a point even for
inflicting perpetual check on an opponent - and making that fraction
significantly larger for Black as, being disadvantaged, it's more
difficult for him to engage in attacking play - was a notion I've
floated, inspired by komidashi.
John Savard
|
| | | |
Date: 09 Jun 2008 15:12:44
From: David Richerby
Subject: Re: Does this so-called "draw problem" actually exist?
|
Quadibloc <[email protected] > wrote:
>On Jun 7, 6:25 pm, Guy Macon <http://www.guymacon.com/> wrote:
>> Rich Hutnik wrote:
>>> And, does this situation result in growth in people playing? How
>>> about looking at the actual numbers and see.
>>
>> And you expect us all to accept without evidence your assumption
>> that lack of growth of chess is the effect and "too many" draws
>> are the cause? Got any evidence to back up that assertion?
>
> I can understand your frustration, but some things are well-known
> enough they can be accepted without trying to go beyond anecdotal
> evidence.
If it's so well-known, it shouldn't be hard to cite some reasonably
authoritative figure who says it!
I would question the whole idea that the popularity of chess is
adversely affected by the high draw rate at the highest levels of the
game. The greatest potential for growth in chess is with those people
who know nothing at all about the game or maybe learnt the rules as a
kid, or those people who play the occasional social game against
friends but have never really thought hard about the game. The sort
of people who either can't play at all or whom any competent player
should be able to beat just by waiting for them to leave a piece en
prise or get forked. But these are the sort of people who couldn't
even tell you who the world champion is -- it seems absurd to me to
suggest that they're being put off from getting more involved with
chess by the high rate of draws among the top GMs. How can they be
put off by something they don't even know about?
The only group who might be being put off by the high proportion of GM
draws are the fairly casual players
> People have been whining about what has become of Chess ever since
> Steinitz. It *is* reasonable to believe that Chess would be
> perceived as "more exciting" if we had more games like
> Anderssen-Kieseritzky nowadays. One can note the existence of
> griping about draws in Chess, the efforts of the Japanese to reduce
> draws in Go, and even the overtime rules in ice hockey (i.e.,
> "sudden death overtime") as evidence of a general desire for games
> to produce a conclusive resolution.
The point of komidashi in Go is *not* to reduce draws. As a side
effect half-point komi makes draws impossible, but the purpose of
introducing the system was that the first player (black, in Go) has
the advantage of initiative in the initial position. Between strong
players, this meant that Black could win by a couple of points just by
playing very defensively. The point of komi was to force Black to
play a more interesting game by turning those narrow wins into narrow
losses.
Dave.
--
David Richerby Slimy Psychotic Atom Bomb (TM): it's
www.chiark.greenend.org.uk/~davidr/ like a weapon of mass destruction but
it wants to kill you and it's covered
in goo!
|
| | |
Date: 08 Jun 2008 00:35:56
From: Guy Macon
Subject: Re: Does this so-called "draw problem" actually exist?
|
Quadibloc wrote:
>Starting over from scratch, though, so that even Grandmasters would be
>feeling their way along, would give Chess a bigger issue than any it
>has now; until a large body of knowledge is built up
That large body of knowledge could be built up a lot faster by the
computers (think thousand of them all chugging away 24/7 building up
an opening book) than could be built up by any human.
>As a game for people to play at home for their own amusement, Chess
>doesn't need fixing - and if something else, like Checkers, or Chinese
>Checkers, or Monopoly, or even Snakes and Ladders is more suitable for
>the occasion, there are plenty of choices.
I agree 100%
>Maybe a new game of Chess will evolve the way modern Chess evolved
>from Shatranj - people in, say, Poland will start playing Chess at
>home giving the Queen the Knight's move as well, or requiring a player
>to make a capturing move whenever a Pawn capture is available - and
>then, decades later, it will be found that this game is more profound
>and subtle than regular Chess, and so it supplants it. In our age of
>mass communications, however, I view the odds of such a development as
>vanishingly small.
I do as well, but stranger things have hapened. I suspect that if
it happens, it will be because of the efforts of someone like you,
inventing new variants and really thinking about the rule sets to
avoid gross imbalance or an obvious winning strategy. It's a real
long shot, of course.
|
| | |
Date: 08 Jun 2008 00:25:38
From: Guy Macon
Subject: Re: Does this so-called "draw problem" actually exist?
|
Rich Hutnik wrote:
>Look at the fruit of the situation now. Is the percentages of draws
>acceptable today?
Yes. Perfectly acceptable. Chess is not broken, and does not
require fixing.
>And, does this situation result in growth in people
>playing? How about looking at the actual numbers and see.
And you expect us all to accept without evidence your assumption
that lack of growth of chess is the effect and "too many" draws
are the cause? Got any evidence to back up that assertion?
>> I have seen plenty of claims that fewer draws will bring in more
>> players or more paying spectators, but I haven't seen any evidence
>> that these predictions will actually come to pass.
>
>Does reducing the number of draws help or hurt interest? If it
>generates more interest, then why shouldn't it be part of it.
You are doing it again. You write "*IF* it generates more interest"
without making any attempt to show that it does. Are we supposed to
accept that on blind faith just because you say so?
>> I have seen evidence of top-level players drawing early without
>> even trying to win (which is a distinct and separate issue from
>> there supposedly being too many draws), but no evidence that this
>> is caused by the present rules rather than, say, the reward system.
>
>Then, you would argue there is a draw problem, just the reward system
>is the reason why. This is a perfectly acceptable argument to make.
It is not, however an argument that I made. You are confusing
the questions of whether something exists, what the cause of it
is, and whether it is undesirable. Those are three entirely
seperate questions. Please read what I write and try not to put
words in my mouth.
>> In my opinion, there is nothing wrong with the current rules.
>> Chess is not broken, and does not require fixing.
>
>And what is wrong with non-mad queens and non-mad bishops? What is
>wrong with Shatranj?
I have no idea what yiou are takling about. Are you imagining that
my claim that Chess is not broken has anything to do with some
other game or variant? Or with drag racing or Mixed Martial Arts?
>It was played by Persian royalty, and was a perfectly acceptable game.
>Why make the queen and bishop mad, and give them crazy mobility?
>There wasn't castling in the game either, and pawns promoted to
>bishops only. Why change these rules?
Somebody changed the rules for Shatranj? News to me. The development
of Chess didn't change the rules of Shatranj any more than the
development of Smess changed the rules of Chess.
>> In my opinion, the "solution" for those who prefer fewer draws
>> at the top level is simple, straightforward, and far from being
>> a new idea; simply don't invite players who draw "too often"
>> (whatever that means to you), and do invite players who you
>> consider to play "exciting" chess. "Problem" (if one actually
>> exists) solved.
>
>Please repeat them here.
Please repeat what here? Was the above not clear?
>Are you suggesting changes in the scoring level?
No. I am saying that *IF* you believe that tournament chess has
too many draws, simply don't invite players who (you believe) draw
"too often" (whatever that means to you), and do invite players
who play the way you want them to play.
|
| | | |
Date: 07 Jun 2008 18:29:06
From: David Kane
Subject: Re: Does this so-called "draw problem" actually exist?
|
"Guy Macon" <http://www.guymacon.com/ > wrote in message
news:[email protected]...
>
>
>
> Rich Hutnik wrote:
>
>>Are you suggesting changes in the scoring level?
>
> No. I am saying that *IF* you believe that tournament chess has
> too many draws, simply don't invite players who (you believe) draw
> "too often" (whatever that means to you), and do invite players
> who play the way you want them to play.
>
That "solution" would only solve a problem in which draws were
the result of some aberrant playing style. In fact, playing in a way
that produces lots of draws is optimal. It's a basic flaw in chess
tournament design that has nothing to do with a player's individual
preference.
But people will always invent theories (unsupported by
evidence) in order to avoid facing reality.
|
| | |
Date: 07 Jun 2008 15:03:25
From: Quadibloc
Subject: Re: Does this so-called "draw problem" actually exist?
|
On Jun 7, 3:07 pm, "David Kane" <[email protected] > wrote:
> Actually, there are many proposals that do not change the
> underlying rules of chess one iota, yet are greeted with
> the same reflexive "no change ever, no matter what"
> chorus.
There is another comment that should be made here. In fairness to the
naysayers, it isn't necessarily obvious that Chess is in deep trouble.
Checkers was in deep trouble that was too obvious to deny, and so the
two-move restriction was adopted, despite that being a change that
was, and still is, very unpalatable to many players.
The game of Go has an inherent flexibility that allows it to have a
convenient handicap system. This same flexibility allowed it to bring
in _komidashi_ as an answer to defensive play by the first player
leading dependably to a win by three points. (Komi increased to 6 1/2
points over time; the full fruits of _komidashi_ were not realized
until Go Seigen and his leading opponents arrived at a school of play
suited to balancing risk and benefit in extracting the maximum benefit
from the first move advantage. Shusaku is the great Go player who can
probably be compared to Steinitz in his impact on the game's theory,
so one can say that Go is now in its postmodern period.) Because Go is
flexible this way, change is much milder and easier - and less
controversial and more possible when it is needed.
Chess doesn't have the flexibility of Go, and it's in less deep
trouble than Checkers was. The collapse of the Soviet Union, and rival
distractions from TV to video games to the Internet are problems minor
changes won't solve - so it's unclear that change is worth the bother.
And the people who say there are problems are usually the ones with a
pet project of their own modified version of Chess to flog, to boot.
So it's hardly any wonder that change meets disinterest.
John Savard
|
| | | |
Date: 07 Jun 2008 18:22:41
From: David Kane
Subject: Re: Does this so-called "draw problem" actually exist?
|
"Quadibloc" <[email protected] > wrote in message
news:[email protected]...
> On Jun 7, 3:07 pm, "David Kane" <[email protected]> wrote:
>
>> Actually, there are many proposals that do not change the
>> underlying rules of chess one iota, yet are greeted with
>> the same reflexive "no change ever, no matter what"
>> chorus.
>
> There is another comment that should be made here. In fairness to the
> naysayers, it isn't necessarily obvious that Chess is in deep trouble.
>
"No change" is only the correct course of action in the case
that chess is perfectly optimized. Nobody has ever
provided evidence in support of that position, which I find
patently absurd.
> Checkers was in deep trouble that was too obvious to deny, and so the
> two-move restriction was adopted, despite that being a change that
> was, and still is, very unpalatable to many players.
>
> The game of Go has an inherent flexibility that allows it to have a
> convenient handicap system. This same flexibility allowed it to bring
> in _komidashi_ as an answer to defensive play by the first player
> leading dependably to a win by three points. (Komi increased to 6 1/2
> points over time; the full fruits of _komidashi_ were not realized
> until Go Seigen and his leading opponents arrived at a school of play
> suited to balancing risk and benefit in extracting the maximum benefit
> from the first move advantage. Shusaku is the great Go player who can
> probably be compared to Steinitz in his impact on the game's theory,
> so one can say that Go is now in its postmodern period.) Because Go is
> flexible this way, change is much milder and easier - and less
> controversial and more possible when it is needed.
>
> Chess doesn't have the flexibility of Go, and it's in less deep
> trouble than Checkers was. The collapse of the Soviet Union, and rival
> distractions from TV to video games to the Internet are problems minor
> changes won't solve - so it's unclear that change is worth the bother.
>
> And the people who say there are problems are usually the ones with a
> pet project of their own modified version of Chess to flog, to boot.
> So it's hardly any wonder that change meets disinterest.
>
> John Savard
|
| | |
Date: 07 Jun 2008 14:48:05
From: Quadibloc
Subject: Re: Does this so-called "draw problem" actually exist?
|
On Jun 7, 3:07 pm, "David Kane" <[email protected] > wrote:
> Count me in the group that has zero interest
> in the "new pieces, bigger board" ideas. Whatever
> the merits of such games, they aren't chess and
> would have little to no continuity with chess' long
> history. But even maintaining the basic play of
> chess, there is a lot that can be done to spice up
> interest, I believe.
That may be. I think that discouraging draws through methods that
reward the players for active play in some way independently of
helping them win, though, will have limited effectiveness. Given that,
I came up with a points system that, unavoidably, does change the
rules of Chess a little - by treating stalemate and bare King as non-
drawing outcomes with partial credit - as I've mentioned already,
perhaps in replies to you.
John Savard
|
| | | |
Date: 07 Jun 2008 18:18:55
From: David Kane
Subject: Re: Does this so-called "draw problem" actually exist?
|
"Quadibloc" <[email protected] > wrote in message
news:[email protected]...
> On Jun 7, 3:07 pm, "David Kane" <[email protected]> wrote:
>
>> Count me in the group that has zero interest
>> in the "new pieces, bigger board" ideas. Whatever
>> the merits of such games, they aren't chess and
>> would have little to no continuity with chess' long
>> history. But even maintaining the basic play of
>> chess, there is a lot that can be done to spice up
>> interest, I believe.
>
> That may be. I think that discouraging draws through methods that
> reward the players for active play in some way independently of
> helping them win, though, will have limited effectiveness. Given that,
> I came up with a points system that, unavoidably, does change the
> rules of Chess a little - by treating stalemate and bare King as non-
> drawing outcomes with partial credit - as I've mentioned already,
> perhaps in replies to you.
I don't believe your proposal would dramatically change
the risk-reward situation facing the player. But that
can be changed by altering scoring.
|
| | |
Date: 07 Jun 2008 10:57:19
From: Quadibloc
Subject: Re: Does this so-called "draw problem" actually exist?
|
On Jun 6, 9:27 pm, "David Kane" <[email protected] > wrote:
> Amen. People who make these arguments really need to
> evaluate their own motives.
Although there's some truth to that, resistance to change usually has
quite legitimate motives.
Chess took a long time to build up to its current level of popularity,
recognition, and respect. That's because, over centuries, the game was
played by many people who devoted great effort in acquiring expertise,
building up a large body of knowledge about the game.
While some may feel that one part of that corpus of knowledge -
opening theory - is better done without, that's not a universal
opinion.
The usual suggestion for changing Chess - whether or not it has
anything to do with solving specific problems - is to enlarge the
board and add an extra piece or two. Even with grandmasters like Bird
and Capablanca behind the idea, it hasn't gone anywhere, and this is
true whether the pieces being added are very powerful, as in their
form, or of modest power, as in Duke Chess or the more recent
proprietary game of Omega Chess.
Hexagonal Chess and Three-Dimensional Chess belong to a separate
category.
Starting over from scratch, though, so that even Grandmasters would be
feeling their way along, would give Chess a bigger issue than any it
has now; until a large body of knowledge is built up, so that there
are players who are taken seriously as experts in the game, the World
Championship wouldn't be a big deal, and so on.
As a game for people to play at home for their own amusement, Chess
doesn't need fixing - and if something else, like Checkers, or Chinese
Checkers, or Monopoly, or even Snakes and Ladders is more suitable for
the occasion, there are plenty of choices.
Maybe a new game of Chess will evolve the way modern Chess evolved
from Shatranj - people in, say, Poland will start playing Chess at
home giving the Queen the Knight's move as well, or requiring a player
to make a capturing move whenever a Pawn capture is available - and
then, decades later, it will be found that this game is more profound
and subtle than regular Chess, and so it supplants it. In our age of
mass communications, however, I view the odds of such a development as
vanishingly small.
Even though I've tried to come up with somewhat different and original
ideas to address the issues, I do not harbor illusions about how
likely either my ideas, or anyone else's, are to be adopted.
Also, it might be noted that Shogi solved the draw problem with drops,
but ended up becoming a less profound game as a result; while there is
still some serious tournament play of Shogi, it falls further behind
Go than Chinese Chess does.
John Savard
|
| | | |
Date: 07 Jun 2008 14:07:54
From: David Kane
Subject: Re: Does this so-called "draw problem" actually exist?
|
"Quadibloc" <[email protected] > wrote in message
news:[email protected]...
> On Jun 6, 9:27 pm, "David Kane" <[email protected]> wrote:
>
>> Amen. People who make these arguments really need to
>> evaluate their own motives.
>
> Although there's some truth to that, resistance to change usually has
> quite legitimate motives.
>
> Chess took a long time to build up to its current level of popularity,
> recognition, and respect. That's because, over centuries, the game was
> played by many people who devoted great effort in acquiring expertise,
> building up a large body of knowledge about the game.
>
> While some may feel that one part of that corpus of knowledge -
> opening theory - is better done without, that's not a universal
> opinion.
>
> The usual suggestion for changing Chess - whether or not it has
> anything to do with solving specific problems - is to enlarge the
> board and add an extra piece or two. Even with grandmasters like Bird
> and Capablanca behind the idea, it hasn't gone anywhere, and this is
> true whether the pieces being added are very powerful, as in their
> form, or of modest power, as in Duke Chess or the more recent
> proprietary game of Omega Chess.
Actually, there are many proposals that do not change the
underlying rules of chess one iota, yet are greeted with
the same reflexive "no change ever, no matter what"
chorus.
Count me in the group that has zero interest
in the "new pieces, bigger board" ideas. Whatever
the merits of such games, they aren't chess and
would have little to no continuity with chess' long
history. But even maintaining the basic play of
chess, there is a lot that can be done to spice up
interest, I believe.
|
| | |
Date: 06 Jun 2008 22:08:12
From: Rich Hutnik
Subject: Re: Does this so-called "draw problem" actually exist?
|
On Jun 6, 11:27 pm, "David Kane" <[email protected] > wrote:
> > Does reducing the number of draws help or hurt interest? If it
> > generates more interest, then why shouldn't it be part of it. Let's
> > say chess has 10 issues total, and draws are 5% of the problem.
> > Explain the logic of going, "Well draws really don't have much of an
> > impact, so let's leave it as it!". If all the items on the list,
> > which are important to the growth of chess, and all need to be done,
> > all have this 5%, then you don't do ANYTHING. Such a mentality, to
> > me, is a rationalization for being lazy.
>
> Amen. People who make these arguments really need to
> evaluate their own motives.
When as many people are complaining about the same thing, as they are
now, this is a sign something should be done somehow. People may not
have the right answer, but at least the issues should be acknowledge.
It is easy for people, if a list of things are a problem, to end up
dismissing them all one by one, so you have nothing left and things
remain how they are.
Whether it be draws or not as the issue, people should take concern
over what is going on with the U.S Chess Federation currently, and
realize there is a fire somewhere, and try to deal with everything,
and work for even the small things, to make them work.
If people had the same attitude they have to addressing issues with
the chess world, as they do with chess, they would stop playing chess,
because fighting over a single pawn would be too much work, and "not
really a problem anyhow".
- Rich
|
| | | |
Date: 09 Jun 2008 11:21:08
From: David Richerby
Subject: Re: Does this so-called "draw problem" actually exist?
|
Rich Hutnik <[email protected] > wrote:
> When as many people are complaining about the same thing, as they
> are now, this is a sign something should be done somehow.
Half a dozen people, who imagine that everybody thinks like them,
complaining about draws on an internet newsgroup constitutes a reason
to reform chess?
Dave.
--
David Richerby Indelible Dish (TM): it's like a fine
www.chiark.greenend.org.uk/~davidr/ ceramic dish but it can't be erased!
|
| | |
Date: 06 Jun 2008 20:27:57
From: David Kane
Subject: Re: Does this so-called "draw problem" actually exist?
|
"Rich Hutnik" <[email protected] > wrote in message
news:e4c8e91c-abcd-45b1-9722-feb60cb33b5a@x41g2000hsb.googlegroups.com...
> On Jun 1, 2:45 pm, Guy Macon <http://www.guymacon.com/> wrote:
>> Ray Gordon wrote:
>> >Tablebases are up to seven pieces now. I'd call that a threat.
>>
>> And you imagine that if the game was changed with a new promotion
>> rule -- and the new rules became popular -- that nobody would
>> create tablebases using the new rules?
>>
>> I have yet to see a convincing argument that this so-called
>> "draw problem" in the rules of chess actually exists.
>>
>> I have seen plenty of people who have a personal preference that
>> there be fewer draws, but that does not prove that there is a
>> "draw problem" in the current rules. Why is their preference
>> more important than mine?
>
> Look at the fruit of the situation now. Is the percentages of draws
> acceptable today? And, does this situation result in growth in people
> playing? How about looking at the actual numbers and see. People
> complain here about the U.S Chess Federation canning the magazine, and
> they have financial issues. People think it is due to lack of
> brilliance on the part of the leadership, as if chess is some sort of
> perfection that shouldn't be looked at, only if the leadership did the
> right things. Explain why poker isn't having a problem with growth
> today, without saying, "Well people are idiots, and idiots like to
> gamble". If you think people going to a game that is growing in
> number of players are idiots, then explain how you would expect to get
> growth? Call people idiots, or other words that mean the same thing,
> or having that attitude, is done by people living in an intellectual
> ghetto. The reality is, you aren't thinking growth, but rationalizing
> why it isn't growing. Even thinking producing a U.S World Champ is
> the magic bullet leads to this.
>
>> I have seen plenty of claims that fewer draws will bring in more
>> players or more paying spectators, but I haven't seen any evidence
>> that these predictions will actually come to pass.
>
> Does reducing the number of draws help or hurt interest? If it
> generates more interest, then why shouldn't it be part of it. Let's
> say chess has 10 issues total, and draws are 5% of the problem.
> Explain the logic of going, "Well draws really don't have much of an
> impact, so let's leave it as it!". If all the items on the list,
> which are important to the growth of chess, and all need to be done,
> all have this 5%, then you don't do ANYTHING. Such a mentality, to
> me, is a rationalization for being lazy.
Amen. People who make these arguments really need to
evaluate their own motives.
>> I have seen evidence of top-level players drawing early without
>> even trying to win (which is a distinct and separate issue from
>> there supposedly being too many draws), but no evidence that this
>> is caused by the present rules rather than, say, the reward system.
>
> Then, you would argue there is a draw problem, just the reward system
> is the reason why. This is a perfectly acceptable argument to make.
> However, I have seen that not allowing players to ask for draws
> reducing the number of draws by less than 10%. My question here is,
> if too many draws IS an issue, why not do everything you can to
> address them, from the compensation system to whatever else is needed,
> even possible rules changes?
>
>> Nobody who has created a rule change to address the supposed
>> "draw problem" actually thinks that their new rule would prevail
>> over the current rules in any sort of election/vote/poll.
>>
>> Nobody who has created a rule change to address the supposed
>> "draw problem" actually thinks that their new rule would prevail
>> over the current rules in any sort of free market such as
>> announcing a tournament using the new rules and seeing how many
>> players show up.
>
> I will agree here that there is far too much dictatorial, from an
> ivory tower, proposing of ideas, without actually them being taken up
> and considered.
>
>> I like new chess rules and variants on the game. I like reading
>> about them, I like thinking about them, I like playtesting them,
>> and I hope that someone can come up with something useful that
>> becomes the new standard, as happened with "In Passing" pawn
>> captures (and no, I am not going to start speaking french just
>> so I can describe that rule...). But just because new rules
>> appeal to me personally. that doesn't mean that there is anything
>> wrong with the current rules.
>>
>> In my opinion, there is nothing wrong with the current rules.
>> Chess is not broken, and does not require fixing.
>
> And what is wrong with non-mad queens and non-mad bishops? What is
> wrong with Shatranj? It was played by Persian royalty, and was a
> perfectly acceptable game. Why make the queen and bishop mad, and
> give them crazy mobility? There wasn't castling in the game either,
> and pawns promoted to bishops only. Why change these rules? Maybe
> the answer would be the same as why change chess rules, in order to
> better serve the community of players.
>
>> In my opinion, the "solution" for those who prefer fewer draws
>> at the top level is simple, straightforward, and far from being
>> a new idea; simply don't invite players who draw "too often"
>> (whatever that means to you), and do invite players who you
>> consider to play "exciting" chess. "Problem" (if one actually
>> exists) solved.
>
> Please repeat them here. Are you suggesting changes in the scoring
> level? If so, that is fine. Any luck getting anyone to try them?
>
> - Rich
|
| |
Date: 02 Jun 2008 20:38:35
From: David Richerby
Subject: Re: Does this so-called "draw problem" actually exist?
|
Guy Macon <http://www.guymacon.com/ > wrote:
> Ray Gordon wrote:
>> Tablebases are up to seven pieces now. I'd call that a threat.
>
> And you imagine that if the game was changed with a new promotion
> rule -- and the new rules became popular -- that nobody would create
> tablebases using the new rules?
Well, exactly. The difficulty of creating tablebases isn't
significantly altered by minor changes to the rules.
> I have yet to see a convincing argument that this so-called "draw
> problem" in the rules of chess actually exists.
Me neither.
> In my opinion, the "solution" for those who prefer fewer draws at
> the top level is simple, straightforward, and far from being a new
> idea; simply don't invite players who draw "too often" (whatever
> that means to you), and do invite players who you consider to play
> "exciting" chess. "Problem" (if one actually exists) solved.
You are Luis Rentero and I claim my five pounds. :-)
Dave.
--
David Richerby Mentholated Beer (TM): it's
www.chiark.greenend.org.uk/~davidr/ like a refreshing lager but it's
invigorating!
|
| |
Date: 02 Jun 2008 08:00:01
From: Sanny
Subject: Re: Does this so-called "draw problem" actually exist?
|
> In my opinion, there is nothing wrong with the current rules.
> Chess is not broken, and does not require fixing.
Now people do not like chess as much they like playing other games on
computer with 3d graphics.
Chess has no music no fun just sit and wait for opponent move. While
Racing games and Shooter games are more interesting to play. Thats the
main reasion people are not much inclined towards Chess.
Slowly with Time people will forget playing Chess. As Computer will
always win So no point in playing. A game where it is already known
that Opponent will win is of no interest.
Computers have Spoiled the Chess Game. Earlier if some one was better
than 10 guys in the town he used to feel great but now people will
tell him to play against computer and he will loose all the games and
then he lost the interest.
Chess is no more that Atractive. As you can never win a Computer.
Bye
Sanny
Do not play at GetClub as now, you can never win.
http://www.GetClub.com/Chess.html
|
| | |
Date: 06 Jun 2008 11:38:20
From: Quadibloc
Subject: Re: Does this so-called "draw problem" actually exist?
|
On Jun 2, 9:00=A0am, Sanny <[email protected] > wrote:
> Chess has no music no fun just sit and wait for opponent move. While
> Racing games and Shooter games are more interesting to play. Thats the
> main reasion people are not much inclined towards Chess.
Yes, there are some problems that cannot be fixed without replacing
Chess with something that definitely is not Chess. Those, I won't
worry about.
John Savard
|
| | |
Date: 06 Jun 2008 10:31:15
From: Rich Hutnik
Subject: Re: Does this so-called "draw problem" actually exist?
|
On Jun 2, 11:00 am, Sanny <[email protected] > wrote:
> > In my opinion, there is nothing wrong with the current rules.
> > Chess is not broken, and does not require fixing.
>
> Now people do not like chess as much they like playing other games on
> computer with 3d graphics.
>
> Chess has no music no fun just sit and wait for opponent move. While
> Racing games and Shooter games are more interesting to play. Thats the
> main reasion people are not much inclined towards Chess.
When you fail to consider the era you are in, and think you have some
sort of perfection that needs to be worshipped, you won't get growth.
When you force people to bow to a game, rather than the game serve the
people, you won't get growth.
> Slowly with Time people will forget playing Chess. As Computer will
> always win So no point in playing. A game where it is already known
> that Opponent will win is of no interest.
A car can beat people in a straight up quarter mile. Perhaps people
should see chess as a GAME where you challenge PEOPLE to see who is
superior, not some sort of puzzle you have to figure out.
> Computers have Spoiled the Chess Game. Earlier if some one was better
> than 10 guys in the town he used to feel great but now people will
> tell him to play against computer and he will loose all the games and
> then he lost the interest.
>
> Chess is no more that Atractive. As you can never win a Computer.
A game, when it fails to entertain, is spoiled. The idea is to be the
top HUMAN player, and be surprised and entertained by the battle.
People also would be helpful if they waive the belief they want to
play the EXACT SAME GAME that was played by people several hundred
years ago. How can one even say they do, when you have a lot more
theory and so on developed?
- Rich
|
|
