DR. ELO'S LETTER TO GM EVANS

[Searching the Net I found this interesting letter from Arpad Elo
in the Chesscafe archives.]

June 19, 1987

To: Larry Evans Reno, Nevada

Dear Larry:

Thank you very much for your letter of May 29 and the inclosed
literature on CHESSGATE. The latter is a sad commentary on the
state of US chess and the USCF. I am very much afraid that the
coming elections will not change the situation much. Most of
the voting delegates are the clubhouse political types who will
elect their own kind to the Policy Board. And those on the PB
will be ever jealous of their small powers and suspicious of
the masters. I hope I am wrong about the election, so let's
wait and see.

the Kalme claims, I was not convinced by his arguments and
I don't think he convinced anybody. At that time (1975) I made
an extensive probabilistic analysis of the situation which
compared various likely events under the two systems of the
match and without doubt the Fischer conditions favored the
champion --- and in retrospect rightly so. This analysis runs
to eight pages and if you are interested I could send you a
copy. A shorter version appeared in the magazine CHESS but
unfortunately I have no copy either of the article or of the
magazine.

Further in retrospect/there is a simple argument in favor of
the Fischer conditions without any elaborate probabilistic
calculations. I owe credit to one of our local players (Al
Losoff) for putting me on the train of thought which I
regrettably missed twelve years ago. It goes as follows:

1. If the match at any time stands at 9:9 it indicates an
equality between the contestants.

2. Thus the next decisive game could go either way since the
contestants have equal probabilities of gaining the next
decisive game.

3. Thus a win by either player in the next decisive game shows
no superiority any more than a single throw of a coin shows a
superiority of heads or tails.

4. A 9:9 result then is properly declared a drawn match since
the challenger has shown no clear superiority.

I find this intuitive approach more convincing than any fancy
probabilistic calculation and I only wish I had had the insight
to see it 12 years ago. Who knows, we might have seen a
Fischer-Karpov match. Or would Bobby have found other excuses
for not playing?

Cordially,

/s/ Arpad E. Elo

PS: Dr Euwe was Correct in his remark. AEE

wrote:
DR. ELO'S LETTER TO GM EVANS

[Searching the Net I found this interesting letter from Arpad Elo
in the Chesscafe archives.]

June 19, 1987

To: Larry Evans Reno, Nevada

Dear Larry:

Thank you very much for your letter of May 29 and the inclosed
literature on CHESSGATE. The latter is a sad commentary on the
state of US chess and the USCF. I am very much afraid that the
coming elections will not change the situation much. Most of
the voting delegates are the clubhouse political types who will
elect their own kind to the Policy Board. And those on the PB
will be ever jealous of their small powers and suspicious of
the masters. I hope I am wrong about the election, so let's
wait and see.

the Kalme claims, I was not convinced by his arguments and
I don't think he convinced anybody. At that time (1975) I made
an extensive probabilistic analysis of the situation which
compared various likely events under the two systems of the
match and without doubt the Fischer conditions favored the
champion --- and in retrospect rightly so. This analysis runs
to eight pages and if you are interested I could send you a
copy. A shorter version appeared in the magazine CHESS but
unfortunately I have no copy either of the article or of the
magazine.

Further in retrospect/there is a simple argument in favor of
the Fischer conditions without any elaborate probabilistic
calculations. I owe credit to one of our local players (Al
Losoff) for putting me on the train of thought which I
regrettably missed twelve years ago. It goes as follows:

1. If the match at any time stands at 9:9 it indicates an
equality between the contestants.

2. Thus the next decisive game could go either way since the
contestants have equal probabilities of gaining the next
decisive game.

3. Thus a win by either player in the next decisive game shows
no superiority any more than a single throw of a coin shows a
superiority of heads or tails.

4. A 9:9 result then is properly declared a drawn match since
the challenger has shown no clear superiority.

I find this intuitive approach more convincing than any fancy
probabilistic calculation and I only wish I had had the insight
to see it 12 years ago. Who knows, we might have seen a
Fischer-Karpov match. Or would Bobby have found other excuses
for not playing?

Cordially,

/s/ Arpad E. Elo

PS: Dr Euwe was Correct in his remark. AEE


Dr. Euwe mentioned a magazine called "Chess" --
what sort of magazine was this, and who/where was
it published?

There is a lot of interesting material in the ChessCafe
archives, which is (I expect) a lot easier to search than
a collection of old magazines.

-- help bot

_
"Fischer refused to negotiate or compromise and his
stubbornness is what killed the match - nothing or nobody
else." - GM Evans (1975)
_
"Karpov deserves to be world champion"
- GM Evans (1975)
_
"... Evans ... still holds that Fischer's stubbornness
and nothing else killed the match. ... If Fischer had
not been so stubborn, there would have been a
match. ..." - Larry Parr (28 May 2002 08:11:07 GMT)
_
"... Fischer must be slanged -- as both GM Evans
and I have so slanged in thousands of words -- for
producing asinine conditions. ..." - Larry Parr
(05 Jun 2002 07:12:26 GMT)

helpbot wrote (4 Nov 2006 16:34:27 -0800):

7 Dr. Euwe mentioned a magazine called "Chess" --
7 what sort of magazine was this, and who/where was
7 it published? ...

_
http://www.chess.co.uk/mag.html
_
http://www.soszynski.btinternet.co.uk/sccc/history.html
_
http://en.wikipedia.org/wiki/Baruch_Harold_Wood
_
http://www.chessgames.com/perl/chessplayer?pid=58959

VVVVVVVVVVVVVVVVVVVVVVVVVVVVVVVVVVV
Path:
g2news2.google.com!news3.google.com!border1.nntp.d ca.giganews.com!nntp.giganews.com!local01.nntp.dca .giganews.com!nntp.adelphia.com!news.adelphia.com. POSTED!not-for-mail
NNTP-Posting-Date: Sat, 04 Nov 2006 18:36:32 -0600
From: Jud McCranie
Newsgroups: rec.games.chess.misc
Subject: Elo on Fischer's conditions vs. Karpov
Date: Sat, 04 Nov 2006 19:36:30 -0500
Message-ID:
References: . com
X-Newsreader: Forte Agent 4.1/32.1088
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Lines: 25
NNTP-Posting-Host: 68.71.90.12
X-Trace:
sv3-VL9U23Ercxq/wVXvd++DelfNZdUmxaUBZ5ngZAADcj0wO2kWQ6CqmP0UedH9QU ywM5gfdtym7jC/6/a!pgAwFUKqpXwhAn7cslQBmkQRid6+Lq1ExPwlVoJTRKGVw3T4 dPm/uQgZMDzod5Hz/nqZ651KawUx!V8ewx7s8gqL6zwnThtoMn09Dnw==
X-Complaints-To:
X-DMCA-Complaints-To:
X-Abuse-and-DMCA-Info: Please be sure to forward a copy of ALL headers
X-Abuse-and-DMCA-Info: Otherwise we will be unable to process your
complaint properly
X-Postfilter: 1.3.32

On 4 Nov 2006 14:58:54 -0800, "
wrote:

1. If the match at any time stands at 9:9 it indicates an
equality between the contestants.

3. Thus a win by either player in the next decisive game shows
no superiority any more than a single throw of a coin shows a
superiority of heads or tails.

This doesn't make any sense to me. Whomever wins the next game as won
10 to 9.

4. A 9:9 result then is properly declared a drawn match since
the challenger has shown no clear superiority.

If the challenger won the next game, it would be 10 to 9, and he would
have shown a clear superiority.

If it is a match until one player wins 10 games, then the challenger
needs to have the opportunity to win that 10th game, if it is 9 to 9.
I would have expected better logic by Elo, but he was pretty old at
that time.
---
Replace you know what by j to email
AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA

VVVVVVVVVVVVVVVVVVVVVVVVVVVVVVVVVVV
Path:
g2news2.google.com!postnews.google.com!m73g2000cwd .googlegroups.com!not-for-mail
From: "help bot"
Newsgroups: rec.games.chess.misc
Subject: Elo on Fischer's conditions vs. Karpov
Date: 4 Nov 2006 18:10:17 -0800
Organization: http://groups.google.com
Lines: 38
Message-ID: .com
References: . com

NNTP-Posting-Host: 71.2.9.152
Mime-Version: 1.0
Content-Type: text/plain; charset="iso-8859-1"
X-Trace: posting.google.com 1162692623 21189 127.0.0.1 (5 Nov 2006
02:10:23 GMT)
X-Complaints-To:
NNTP-Posting-Date: Sun, 5 Nov 2006 02:10:23 +0000 (UTC)
In-Reply-To:
User-Agent: G2/1.0
X-HTTP-UserAgent: Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US;
rv:1.8.0.7) Gecko/20060909 Firefox/1.5.0.7,gzip(gfe),gzip(gfe)
Complaints-To:
Injection-Info: m73g2000cwd.googlegroups.com; posting-host=71.2.9.152;
posting-account=9BIrBw0AAAB7iLUVDUl7NdGh5tPTwbP2


Jud McCranie wrote:
On 4 Nov 2006 14:58:54 -0800, "
wrote:

1. If the match at any time stands at 9:9 it indicates an
equality between the contestants.

3. Thus a win by either player in the next decisive game shows
no superiority any more than a single throw of a coin shows a
superiority of heads or tails.

This doesn't make any sense to me. Whomever wins the next game as won
10 to 9.

4. A 9:9 result then is properly declared a drawn match since
the challenger has shown no clear superiority.

If the challenger won the next game, it would be 10 to 9, and he would
have shown a clear superiority.

If it is a match until one player wins 10 games, then the challenger
needs to have the opportunity to win that 10th game, if it is 9 to 9.
I would have expected better logic by Elo, but he was pretty old at
that time.


My impression was that either Dr. Elo was making some
kind of odd joke, or else his logic cells were just burnt out.

In particular, step number 3 falls flat on its face when it
mistakenly tries to equate a toss of a coin (random chance)
to GM Fischer winning that 10th game. (Insert Blairian
quote here, of GM Fischer stating that chess is a game of
*skill*.)

-- help bot
AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA

VVVVVVVVVVVVVVVVVVVVVVVVVVVVVVVVVVV
Path:
g2news2.google.com!postnews.google.com!f16g2000cwb .googlegroups.com!not-for-mail
From: "Wlodzimierz Holsztynski (Wlod)"
Newsgroups: rec.games.chess.misc
Subject: Elo on Fischer's conditions vs. Karpov
Date: 4 Nov 2006 18:43:18 -0800
Organization: http://groups.google.com
Lines: 62
Message-ID: .com
References: . com

NNTP-Posting-Host: 66.81.67.46
Mime-Version: 1.0
Content-Type: text/plain; charset="iso-8859-1"
X-Trace: posting.google.com 1162694602 25901 127.0.0.1 (5 Nov 2006
02:43:22 GMT)
X-Complaints-To:
NNTP-Posting-Date: Sun, 5 Nov 2006 02:43:22 +0000 (UTC)
In-Reply-To:
User-Agent: G2/1.0
X-HTTP-UserAgent: Mozilla/5.0 (Macintosh; U; PPC Mac OS X Mach-O;
en-US; rv:1.8.1) Gecko/20061010 Firefox/2.0,gzip(gfe),gzip(gfe)
Complaints-To:
Injection-Info: f16g2000cwb.googlegroups.com; posting-host=66.81.67.46;
posting-account=pwtqtw0AAACw59dt7zOnp9M9tGBpr6Lp

On 4 Nov 2006 14:58:54 -0800, "
wrote:

(a citation from Al Losoff follows)

1. If the match at any time stands at 9:9 it
indicates anequality between the contestants.

3. Thus a win by either player in the next decisive game shows
no superiority any more than a single throw of a coin shows a
superiority of heads or tails.

To this Jud McCranie responds with:

This doesn't make any sense to me. Whomever
wins the next game as won 10 to 9.

4. A 9:9 result then is properly declared a drawn match
since the challenger has shown no clear superiority.

If the challenger won the next game, it would be 10 to 9,
and he would have shown a clear superiority.

Where is your treshold?

Let's consider, for the sake of argument, a fast
chess match, 20m+20s, 4 games per day,
5 days a week (20 games per week). They
play till one of them wins 32 games. Would you
still object to declaring a drawn match in the
case of the 31:31 score?

***

We need to accept the reality.
We need to accept the fact that
occasionally the competitors are
so well matched that it is the question
of good luck which of them wins.
In such situations we need to decide
about the our customs, and not about
what is right and what is wrong in the
chess sense. As long as the solution
is fair, things are fine, and we need to
be philosophical about the necessarily
imperfect comparisons--there is no such
thing as a perfect one.

The classical (old?) chess tradition
requires that challenger should show
a clear dominance over the old champion
in order to replace him/her. If we want to
uphold this tradition then indeed
a provision like 9:9 is reasonable.

But is a match till 10 wins reasonable?
That's a separate question.

Regards,

Wlod
AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA

VVVVVVVVVVVVVVVVVVVVVVVVVVVVVVVVVVV
Path:
g2news2.google.com!news3.google.com!border1.nntp.d ca.giganews.com!nntp.giganews.com!local01.nntp.dca .giganews.com!nntp.adelphia.com!news.adelphia.com. POSTED!not-for-mail
NNTP-Posting-Date: Sat, 04 Nov 2006 20:48:31 -0600
From: Jud McCranie
Newsgroups: rec.games.chess.misc
Subject: Elo on Fischer's conditions vs. Karpov
Date: Sat, 04 Nov 2006 21:48:29 -0500
Message-ID:
References: . com

.com
X-Newsreader: Forte Agent 4.1/32.1088
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Lines: 18
NNTP-Posting-Host: 68.71.90.12
X-Trace:
sv3-fnkxAkCr/eABnMx6LU4QuHp63SI6mZmoeA9vIZlosqLwernTvy05oGIG1OP 3eEoig9JkrlOUqt4LGxp!mbNsscIIYzsn6V9aimGwh5pwMxoUj qkGGrcgJkiWGGVIY1nWKkGFPtR5gi2novIIJuNjB8cnnTtN!1s lKIU7XHWjvjFCO6yQN7IWZYQ==
X-Complaints-To:
X-DMCA-Complaints-To:
X-Abuse-and-DMCA-Info: Please be sure to forward a copy of ALL headers
X-Abuse-and-DMCA-Info: Otherwise we will be unable to process your
complaint properly
X-Postfilter: 1.3.32

On 4 Nov 2006 18:10:17 -0800, "help bot"
wrote:

In particular, step number 3 falls flat on its face when it
mistakenly tries to equate a toss of a coin (random chance)
to GM Fischer winning that 10th game. (Insert Blairian
quote here, of GM Fischer stating that chess is a game of
*skill*.)

Right - there is a fatal flaw - a false assumption. The part about if
it is tied 9-9, each player has an equal chance of being the first to
win 10 games. False - the better player has the best chance to be the
first to win a 10th game. To illustrate this even better, carry it to
the extreme. Suppose it gets to 1-1. Then the players are equal, and
each has an equal chance to win the next game?? If it gets to 1-1,
should the champion retain his title?
---
Replace you know what by j to email
AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA

VVVVVVVVVVVVVVVVVVVVVVVVVVVVVVVVVVV
Path:
g2news2.google.com!postnews.google.com!i42g2000cwa .googlegroups.com!not-for-mail
From: "Wlodzimierz Holsztynski (Wlod)"
Newsgroups: rec.games.chess.misc
Subject: Elo on Fischer's conditions vs. Karpov
Date: 4 Nov 2006 19:14:40 -0800
Organization: http://groups.google.com
Lines: 56
Message-ID: .com
References: . com

.com

NNTP-Posting-Host: 66.81.67.46
Mime-Version: 1.0
Content-Type: text/plain; charset="iso-8859-1"
X-Trace: posting.google.com 1162696484 30430 127.0.0.1 (5 Nov 2006
03:14:44 GMT)
X-Complaints-To:
NNTP-Posting-Date: Sun, 5 Nov 2006 03:14:44 +0000 (UTC)
In-Reply-To:
User-Agent: G2/1.0
X-HTTP-UserAgent: Mozilla/5.0 (Macintosh; U; PPC Mac OS X Mach-O;
en-US; rv:1.8.1) Gecko/20061010 Firefox/2.0,gzip(gfe),gzip(gfe)
Complaints-To:
Injection-Info: i42g2000cwa.googlegroups.com; posting-host=66.81.67.46;
posting-account=pwtqtw0AAACw59dt7zOnp9M9tGBpr6Lp

Jud McCranie wrote:

Right - there is a fatal flaw - a false assumption. The part about if
it is tied 9-9, each player has an equal chance of being the first to
win 10 games. False - the better player has the best chance to be the
first to win a 10th game. To illustrate this even better, carry it to
the extreme. Suppose it gets to 1-1.

No, the extreme would be: suppose it gets to 0:0.

Then the players are equal, and each has
an equal chance to win the next game??

This argument is funny, it's a good joke,
but it's not a good argument.

If it gets to 1-1,
should the champion retain his title?

Then Topalov should get five points
for free, and the right to soil Kramnik's
bathroom.

Seriously, when players are equally
matched then there is no way to
select one as the clearly better player.
Every solution will favorite a non-chessical
consideration or it's just like a toss
of a coin; say the match went like this:

1:0, 1:3, 3:4, 4:5,
7:5, 7:8, 8:9, ...

Can you really claim that one of the players
is clearly better. The result depends on the
lucky/unlucky cut-off interruption of the match.

On the other hand, if it went like:

0:1, 0:2, ..., 0:8,
1:8, 2:8, ..., 8:8

then, under the (very) regular circumstances,
it means that the first player, who was losing
for a long time to the second player, has learned
how to handle his/her opponent, and now that
player prevails over the opponent. That's why
long matches start to be something else, they
would not be about the present strongest
player, but about the present and future best
player (if a match were a year long or longer).

Regards,

Wlod
AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA

VVVVVVVVVVVVVVVVVVVVVVVVVVVVVVVVVVV
Path:
g2news2.google.com!news2.google.com!news3.google.c om!border1.nntp.dca.giganews.com!nntp.giganews.com !local01.nntp.dca.giganews.com!nntp.adelphia.com!n ews.adelphia.com.POSTED!not-for-mail
NNTP-Posting-Date: Sat, 04 Nov 2006 21:16:50 -0600
From: Jud McCranie
Newsgroups: rec.games.chess.misc
Subject: Elo on Fischer's conditions vs. Karpov
Date: Sat, 04 Nov 2006 22:16:49 -0500
Message-ID:
References: . com

.com
X-Newsreader: Forte Agent 4.1/32.1088
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Lines: 39
NNTP-Posting-Host: 68.71.90.12
X-Trace:
sv3-Q7CwJNpxvOAbaCkHXN2c1SXgu6FR8fCN1+SkX5NoJb68d+G82R ebG2TPjCb/g0m0UTG1CEOq3946uHR!/QxZvk16elZ1w3jtxWtu97RFQ3GNRLBRI72c/R4LY6gm+0k3rHT/m9Kzh/+tBVFQOWWBBRLsEXnI!OxFrXAeFqMt6DAPC7tRWkeNT2g==
X-Complaints-To:
X-DMCA-Complaints-To:
X-Abuse-and-DMCA-Info: Please be sure to forward a copy of ALL headers
X-Abuse-and-DMCA-Info: Otherwise we will be unable to process your
complaint properly
X-Postfilter: 1.3.32

On 4 Nov 2006 18:43:18 -0800, "Wlodzimierz Holsztynski (Wlod)"
wrote:

If the challenger won the next game, it would be 10 to 9,
and he would have shown a clear superiority.

Where is your treshold?

Well, fischer's proposal was for the first one to win 10 games. Let
the match go until that threshold is met.

Let's consider, for the sake of argument, a fast
chess match, 20m+20s, 4 games per day,
5 days a week (20 games per week). They
play till one of them wins 32 games. Would you
still object to declaring a drawn match in the
case of the 31:31 score?

Yes. If it is the first one to win 32 games, play until one wins 32
games. If the match rule is for the first one to win 100 games, let
the match go until one wins 100 games.

The classical (old?) chess tradition
requires that challenger should show
a clear dominance over the old champion
in order to replace him/her. If we want to
uphold this tradition then indeed
a provision like 9:9 is reasonable.

No it isn't. If it is a match until the first player wins x number of
games (draws not counting) then that should be it. Otherwise, play a
limited number of games, draws counting, and if it is tied at the end
of those games, the champion retains the championship.

But is a match till 10 wins reasonable?

I think that is too many.
---
Replace you know what by j to email
AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA