| |
Main
Date: 07 Dec 2008 15:27:32
From: Wlodzimierz Holsztynski (Wlod)
Subject: Higher ratings? Yes. Stronger players? No. Inflation? No.
|
Why ratings of the strongest players are higher
today than 50 years ago? Is Moro stronger
than Tal? Who knows. Perhaps not. So, is it
rating inflation? You may jump to this
conclusion but it is not really about any
inflation.
There are many more players today
than 50 years ago. This means that the
strongest player are fed points from more
players than in the past. After all, you need
to win those points from someone. Thus
there is a greater total pool of points today,
the strongest players are the beneficiaries
of this situation and that's why they have
higher ratings than the top players of the past.
Regards,
Wlod
|
|
| |
Date: 09 Dec 2008 07:53:48
From:
Subject: Re: Higher ratings? Yes. Stronger players? No. Inflation? No.
|
On Dec 9, 4:15=A0am, Offramp <[email protected] > wrote:
> On Dec 8, 10:04=A0pm, [email protected] wrote:
>
> > =A0 I have great respect for Dr. Nunn, and I know that his doctorate is
> > in mathematics. Still, I don't think his research goes nearly far
> > enough to establish his thesis with any confidence. Certainly chess
> > knowledge has advanced since 1911, but to the extent that all masters
> > today are on average 350 points better than their 1911 counterparts?
>
> To be honest, 350 points over 90 years is only a 4 Elo points a year
> increase.
The amount of change per year is irrelevant. I would just like to
see Nunn research more than one player and more than one event, and
offer some real data rather than an educated guess, before I'll accept
his conclusion.
|
| |
Date: 09 Dec 2008 01:15:26
From: Offramp
Subject: Re: Higher ratings? Yes. Stronger players? No. Inflation? No.
|
On Dec 8, 10:04=A0pm, [email protected] wrote:
> =A0 I have great respect for Dr. Nunn, and I know that his doctorate is
> in mathematics. Still, I don't think his research goes nearly far
> enough to establish his thesis with any confidence. Certainly chess
> knowledge has advanced since 1911, but to the extent that all masters
> today are on average 350 points better than their 1911 counterparts?
To be honest, 350 points over 90 years is only a 4 Elo points a year
increase. Isn't that an increase of just c. 8%?
Nunn also mentions that S=FCchting in today's terms would be about BCF
187. As a comparison, Paul Littlewood is currently rated 215 (http://
grading.bcfservices.org.uk/engtopplayers.php), or Malcolm Pein 214.
That is a difference of about 30 points in 90 years, 0.3 points a
year.
That to me looks quite likely, I must say.
|
| |
Date: 08 Dec 2008 14:04:05
From:
Subject: Re: Higher ratings? Yes. Stronger players? No. Inflation? No.
|
On Dec 8, 11:46=A0am, Offramp <[email protected] > wrote:
> On Dec 8, 3:05=A0pm, [email protected] wrote:
>
> > On Dec 8, 12:35=A0am, Offramp <[email protected]> wrote:
> > > On Dec 7, 11:27=A0pm, "Wlodzimierz Holsztynski (Wlod)"
> > > Biel, however, is barely addressed, other than mention that
> > > positions of interest have been dispersed throughout the book."
>
> > =A0 IM Richard Forster gave that part of Nunn's book quite a stern
> > critique in his ChessCafe column of May 1999, "Jewels from Carlsbad
> > 1911." The column is not available online, but I included it in the
> > best-of-ChessCafe anthology "Heroic Tales":http://uscfsales.com/item.as=
p?cID=3D19&PID=3D256
>
> Not specifically Forster, but some chess antiquarians seem to take
> great umbrage at any criticism of old players' ability. When it comes
> to overlaps between the worlds of chess and mathematics I, in general,
> will side with GM Dr Nunn.
>
> > =A0 For example, Nunn claims that the average strength of the Carlsbad
> > 1911 tournament was only Elo 2130.
>
> It is all based on Hugo S=FCchting.
Isn't that a rather thin foundation? To base the strength of the
entire tournament on one contestant's result, and his play in just
that one event?
> Nunn: "In order to be more specific about Karlsbad, take one player:
> Hugo S=FCchting (1874-1916). At Karlsbad he scored 11.5/25 or 'minus 2',
> as they say these days - a perfectly respectable score. Having played
> over all his games at Karlsbad I think that I can confidently state
> that his playing strength was not greater than Elo 2100 (BCF 187) -
> and that was on a good day and with a following wind."
How did Nunn arrive at that figure? Did he compare S=FCchting's
frequency and severity of error at Carlsbad to a reasonably large
sample of modern 2100-players, and find they were the same? That seems
to me about the only reasonably sound method available to arrive at
such a conclusion. Did Nunn explain in any detail his methodology? The
quote you cite makes it sound like no more than an educated guess,
rather than anything at all rigorous, either empirically or
mathematically.
Elo, who presumably used a much bigger sample of S=FCchting's games,
gives him a rating of 2450. So if Nunn is correct, the knowledge
chessplayers have accumulated since 1911 is worth about 350 Elo
points.
> Nunn is talking about what S=FCchting's rating would be TODAY. Remember
> that Nunn is also something of an expert in the Elo system; he wrote
> that article for the BCM. He has experience as a chess player playing
> ALL sorts of players. He has played through all of S=FCchting's games
> from Karlsbad and made his estimate.
I have great respect for Dr. Nunn, and I know that his doctorate is
in mathematics. Still, I don't think his research goes nearly far
enough to establish his thesis with any confidence. Certainly chess
knowledge has advanced since 1911, but to the extent that all masters
today are on average 350 points better than their 1911 counterparts?
|
| |
Date: 08 Dec 2008 08:46:50
From: Offramp
Subject: Re: Higher ratings? Yes. Stronger players? No. Inflation? No.
|
On Dec 8, 3:05=A0pm, [email protected] wrote:
> On Dec 8, 12:35=A0am, Offramp <[email protected]> wrote:
> > On Dec 7, 11:27=A0pm, "Wlodzimierz Holsztynski (Wlod)"
> > Biel, however, is barely addressed, other than mention that
> > positions of interest have been dispersed throughout the book."
>
> =A0 IM Richard Forster gave that part of Nunn's book quite a stern
> critique in his ChessCafe column of May 1999, "Jewels from Carlsbad
> 1911." The column is not available online, but I included it in the
> best-of-ChessCafe anthology "Heroic Tales":http://uscfsales.com/item.asp?=
cID=3D19&PID=3D256
Not specifically Forster, but some chess antiquarians seem to take
great umbrage at any criticism of old players' ability. When it comes
to overlaps between the worlds of chess and mathematics I, in general,
will side with GM Dr Nunn.
> =A0 For example, Nunn claims that the average strength of the Carlsbad
> 1911 tournament was only Elo 2130.
It is all based on Hugo S=FCchting.
Nunn: "In order to be more specific about Karlsbad, take one player:
Hugo S=FCchting (1874-1916). At Karlsbad he scored 11.5/25 or 'minus 2',
as they say these days - a perfectly respectable score. Having played
over all his games at Karlsbad I think that I can confidently state
that his playing strength was not greater than Elo 2100 (BCF 187) -
and that was on a good day and with a following wind."
Nunn is talking about what S=FCchting's rating would be TODAY. Remember
that Nunn is also something of an expert in the Elo system; he wrote
that article for the BCM. He has experience as a chess player playing
ALL sorts of players. He has played through all of S=FCchting's games
from Karlsbad and made his estimate.
> Forster points out that this would
> make the TPR of the winner, Richard Teichmann, only Elo 2300. Yet
> Teichmann's historical Elo is 2570, and Carlsbad 1911 was the best
> result of Teichmann's career. Forster says "You can have many
> reservations about the Elo system and the calculation of historical
> ratings, but something must be very rotten in the state of Denmark if
> Teichmann's performan in what was undoubtedly the best tournament of
> his life is 270 points below his peak five-year average."
> =A0 To accept Nunn's logic, we'd have to revise Elo's historical ratings
> for players of this time downward by at least 300 points. I tend to
> think rather that Nunn underestimates them.
Nunn writes about Biel:
"The Fritz search revealed relatively little, many of the points
raised had already been examined in the players' own notes. [Nunn said
he found that Vidmar had missed 85-90% of the errors in his tournament
book]...."
He goes on to say that the reason the standard of play was so poor was
for three main reasons:
"The first was a tendency to make serious oversights."
"The second problem area was an inclination to adopt totally the wrong
plan."
"The third main problem area was that of endgame play."
Now, look - none of us is very much, if at all, more intelligent than
an ancient Egyptian stonemason. But some of us could show him advances
in geometry that would make his head spin.
> =A0 BTW, on the subject of comparing players of different eras, Forster
> had an interesting perspective:
>
> =A0 "To comfort myself and all other fans of the past masters, I should
> state that in my opinion questions like 'How would Lasker do if he
> lived today?' do not make much sense anyway. Every person is a product
> of his times and an Emanuel Lasker today would probably not even have
> started playing chess. He might have invented Windows 98 instead ...
> An extraordinary player is one who is far ahead of his contemporaries,
> not somebody winning in an intergalactic, time-spanning fantasy
> tournament."
I am sure Forster is right.
|
| |
Date: 08 Dec 2008 07:05:13
From:
Subject: Re: Higher ratings? Yes. Stronger players? No. Inflation? No.
|
On Dec 8, 12:35=A0am, Offramp <[email protected] > wrote:
> On Dec 7, 11:27=A0pm, "Wlodzimierz Holsztynski (Wlod)"
>
>
>
>
>
> <[email protected]> wrote:
> > Why ratings of the strongest players are higher
> > today than 50 years ago? =A0Is Moro stronger
> > than Tal? Who knows. Perhaps not. So, is it
> > rating inflation? You may jump to this
> > conclusion but it is not really about any
> > inflation.
>
> > There are many more players today
> > than 50 years ago. This means that the
> > strongest player are fed points from more
> > players than in the past. After all, you need
> > to win those points from someone. Thus
> > there is a greater total pool of points today,
> > the strongest players are the beneficiaries
> > of this situation and that's why they have
> > higher ratings than the top players of the past.
>
> > Regards,
>
> > =A0 =A0 Wlod
>
> See the chapter called The Test of Time in John Nunn's Chess Puzzle
> Book. It is reviewed here:http://www.chesscafe.com/text/nunnquiz.txt
>
> "One of the features that differentiates this work from other quiz
> books is the "Test of Time" chapter. Named after Kasparov's book
> in which he analyzes early games from his own career, Nunn takes
> a look at games from a present day invitational GM tournament,
> 1993 Biel Interzonal, and compares the quality of play to an
> invitational event of approximate strength from a previous era,
> Karlsbad 1911. Nunn writes that his "main interest is in assessing
> how much the overall level of chess has changed since the pre-
> First World War period" (page 67).
>
> His methodology of comparing the games consisted of using the
> computer software "Fritz 5" to check each game in both
> tournaments (about 800 total games) for obvious blunders. Nunn
> then reviewed the findings, one blunder at a time. What he found
> was somewhat surprising: "the old players were much worse than I
> expected. The blunders thrown up by Fritz were so awful that I
> looked at a considerable number of complete games 'by hand',
> wondering if the Fritz results really reflected the general standard
> of play. They did" (page 68). At the end of the chapter, he includes
> 30 problems from Karlsbad, most of about moderate difficulty,
> comparable to the problems found in the rest of the book.
>
> Biel, however, is barely addressed, other than mention that
> positions of interest have been dispersed throughout the book."
IM Richard Forster gave that part of Nunn's book quite a stern
critique in his ChessCafe column of May 1999, "Jewels from Carlsbad
1911." The column is not available online, but I included it in the
best-of-ChessCafe anthology "Heroic Tales": http://uscfsales.com/item.asp?c=
ID=3D19&PID=3D256
For example, Nunn claims that the average strength of the Carlsbad
1911 tournament was only Elo 2130. Forster points out that this would
make the TPR of the winner, Richard Teichmann, only Elo 2300. Yet
Teichmann's historical Elo is 2570, and Carlsbad 1911 was the best
result of Teichmann's career. Forster says "You can have many
reservations about the Elo system and the calculation of historical
ratings, but something must be very rotten in the state of Denmark if
Teichmann's performan in what was undoubtedly the best tournament of
his life is 270 points below his peak five-year average."
To accept Nunn's logic, we'd have to revise Elo's historical ratings
for players of this time downward by at least 300 points. I tend to
think rather that Nunn underestimates them.
BTW, on the subject of comparing players of different eras, Forster
had an interesting perspective:
"To comfort myself and all other fans of the past masters, I shuold
state that in my opinion questions like 'How would Lasker do if he
lived today?' do not make much sense anyway. Every person is a product
of his times and an Emanuel Lasker today would probably not even have
started playing chess. He might have invented Windows 98 instead ...
An extraordinary player is one who is far ahead of his contemporaries,
not somebody winning in an intergalactic, time-spanning fantasy
tournament."
|
| |
Date: 07 Dec 2008 21:35:59
From: Offramp
Subject: Re: Higher ratings? Yes. Stronger players? No. Inflation? No.
|
On Dec 7, 11:27=A0pm, "Wlodzimierz Holsztynski (Wlod)"
<[email protected] > wrote:
> Why ratings of the strongest players are higher
> today than 50 years ago? =A0Is Moro stronger
> than Tal? Who knows. Perhaps not. So, is it
> rating inflation? You may jump to this
> conclusion but it is not really about any
> inflation.
>
> There are many more players today
> than 50 years ago. This means that the
> strongest player are fed points from more
> players than in the past. After all, you need
> to win those points from someone. Thus
> there is a greater total pool of points today,
> the strongest players are the beneficiaries
> of this situation and that's why they have
> higher ratings than the top players of the past.
>
> Regards,
>
> =A0 =A0 Wlod
See the chapter called The Test of Time in John Nunn's Chess Puzzle
Book. It is reviewed here:
http://www.chesscafe.com/text/nunnquiz.txt
"One of the features that differentiates this work from other quiz
books is the "Test of Time" chapter. Named after Kasparov's book
in which he analyzes early games from his own career, Nunn takes
a look at games from a present day invitational GM tournament,
1993 Biel Interzonal, and compares the quality of play to an
invitational event of approximate strength from a previous era,
Karlsbad 1911. Nunn writes that his "main interest is in assessing
how much the overall level of chess has changed since the pre-
First World War period" (page 67).
His methodology of comparing the games consisted of using the
computer software "Fritz 5" to check each game in both
tournaments (about 800 total games) for obvious blunders. Nunn
then reviewed the findings, one blunder at a time. What he found
was somewhat surprising: "the old players were much worse than I
expected. The blunders thrown up by Fritz were so awful that I
looked at a considerable number of complete games 'by hand',
wondering if the Fritz results really reflected the general standard
of play. They did" (page 68). At the end of the chapter, he includes
30 problems from Karlsbad, most of about moderate difficulty,
comparable to the problems found in the rest of the book.
Biel, however, is barely addressed, other than mention that
positions of interest have been dispersed throughout the book."
|
| |
Date: 07 Dec 2008 22:28:02
From: Kenneth Sloan
Subject: Re: Higher ratings? Yes. Stronger players? No. Inflation? No.
|
C-
Wlodzimierz Holsztynski (Wlod) wrote:
> Why ratings of the strongest players are higher
> today than 50 years ago? Is Moro stronger
> than Tal? Who knows. Perhaps not. So, is it
> rating inflation? You may jump to this
> conclusion but it is not really about any
> inflation.
>
> There are many more players today
> than 50 years ago. This means that the
> strongest player are fed points from more
> players than in the past. After all, you need
> to win those points from someone. Thus
> there is a greater total pool of points today,
> the strongest players are the beneficiaries
> of this situation and that's why they have
> higher ratings than the top players of the past.
>
> Regards,
>
> Wlod
--
Kenneth Sloan [email protected]
Computer and Information Sciences +1-205-932-2213
University of Alabama at Birmingham FAX +1-205-934-5473
Birmingham, AL 35294-1170 http://KennethRSloan.com/
|
|
