Reply
 
LinkBack Thread Tools Display Modes
  #1   Report Post  
Old March 10th 04, 07:47 PM
Dr. John E. Upham
 
Posts: n/a
Default ELO Rating Inflation?

Having not been chess active since 1978, one thing that staggers me is the
number of players who are now "stronger" than Fischer, Hubner, Portisch,
Mecking (all giants of the game) ever were.

Indeed, all sorts of people are 2700 strength. I'd guess there must be a
factor than can be applied to correct back to pre Kasparov (Weinstein) days?

I guess this is going over old ground, but is there a collected wisdom on
how to correct this state of affairs?

My yardstick used to be:

National Master: 2200+
IM: 2400+
IGM: 2500+
Candidates: 2600+
WC: 2700+

Have I missed the point somewhere?

Regards, John Upham


  #2   Report Post  
Old March 11th 04, 04:37 AM
Entropyfoe
 
Posts: n/a
Default ELO Rating Inflation?

see
http://www.chessmetrics.com

He discusses the very real inflation, proposing just a fixed adjustment to
calibrate out the inflation.

Very interesting.
-J



Having not been chess active since 1978, one thing that staggers me is the
number of players who are now "stronger" than Fischer, Hubner, Portisch,
Mecking (all giants of the game) ever were.

Indeed, all sorts of people are 2700 strength. I'd guess there must be a
factor than can be applied to correct back to pre Kasparov (Weinstein) days?

I guess this is going over old ground, but is there a collected wisdom on
how to correct this state of affairs?

My yardstick used to be:

National Master: 2200+
IM: 2400+
IGM: 2500+
Candidates: 2600+
WC: 2700+

Have I missed the point somewhere?

Regards, John Upham




entr0pyf0e
  #3   Report Post  
Old March 11th 04, 04:24 PM
Vaughn Sturtevant
 
Posts: n/a
Default ELO Rating Inflation?

I have a question for the mathematicians: Doesn't the ELO system presume
that all levels play all levels and that there is a normal distribution of
levels? What happens then when just grandmasters just play grandmasters or
more practically when class B players play just class B players? Obviously
if you started 20 grandmasters at 1200 without any mix of other strengths,
none of them would ever reach 2700. Thus what effect on ratings does the
tendency to play by classes have on the individual player and on the system?


"Entropyfoe" wrote in message
...
see
http://www.chessmetrics.com

He discusses the very real inflation, proposing just a fixed adjustment to
calibrate out the inflation.

Very interesting.
-J



Having not been chess active since 1978, one thing that staggers me is

the
number of players who are now "stronger" than Fischer, Hubner, Portisch,
Mecking (all giants of the game) ever were.

Indeed, all sorts of people are 2700 strength. I'd guess there must be a
factor than can be applied to correct back to pre Kasparov (Weinstein)

days?

I guess this is going over old ground, but is there a collected wisdom

on
how to correct this state of affairs?

My yardstick used to be:

National Master: 2200+
IM: 2400+
IGM: 2500+
Candidates: 2600+
WC: 2700+

Have I missed the point somewhere?

Regards, John Upham




entr0pyf0e



  #4   Report Post  
Old March 11th 04, 04:37 PM
Terry
 
Posts: n/a
Default ELO Rating Inflation?


"Dr. John E. Upham" wrote in message
...
Having not been chess active since 1978, one thing that staggers me is the
number of players who are now "stronger" than Fischer, Hubner, Portisch,
Mecking (all giants of the game) ever were.

Indeed, all sorts of people are 2700 strength. I'd guess there must be a
factor than can be applied to correct back to pre Kasparov (Weinstein)

days?

I guess this is going over old ground, but is there a collected wisdom on
how to correct this state of affairs?

My yardstick used to be:

National Master: 2200+
IM: 2400+
IGM: 2500+
Candidates: 2600+
WC: 2700+

Have I missed the point somewhere?

Regards, John Upham


I dont believe there has been any rating inflation.

The top twenty players of today are probably
stronger then the greats you mentioned.

Dont forget that the top players now use
computers for their preparations.

Regards
]






  #5   Report Post  
Old March 11th 04, 06:54 PM
Kenneth Sloan
 
Posts: n/a
Default ELO Rating Inflation?

"Vaughn Sturtevant" writes:

I have a question for the mathematicians: Doesn't the ELO system presume
that all levels play all levels and that there is a normal distribution of
levels? What happens then when just grandmasters just play grandmasters or
more practically when class B players play just class B players? Obviously
if you started 20 grandmasters at 1200 without any mix of other strengths,
none of them would ever reach 2700.



Why is this "obvious"? It's not obvious to me.


Thus what effect on ratings does the
tendency to play by classes have on the individual player and on the system?


--
Kenneth Sloan
Computer and Information Sciences (205) 934-2213
University of Alabama at Birmingham FAX (205) 934-5473
Birmingham, AL 35294-1170
http://www.cis.uab.edu/info/faculty/sloan/


  #6   Report Post  
Old March 12th 04, 06:31 AM
Aces123
 
Posts: n/a
Default ELO Rating Inflation?

Kenneth Sloan wrote in message ...
"Vaughn Sturtevant" writes:

I have a question for the mathematicians: Doesn't the ELO system presume
that all levels play all levels and that there is a normal distribution of
levels? What happens then when just grandmasters just play grandmasters or
more practically when class B players play just class B players? Obviously
if you started 20 grandmasters at 1200 without any mix of other strengths,
none of them would ever reach 2700.



Why is this "obvious"? It's not obvious to me.

Assume that you have 5 people and they all would have a rating of say
2700 if they competed in chess. If they only play against each other
and started at 1200 and are of equal strenght then they would all
remain at 1200. Their superior knowledge of chess would never show up
in the ratingsystem since they only play against players with an equal
superior understanding of chess.

/Aces123
  #7   Report Post  
Old March 12th 04, 10:00 AM
Newsmail
 
Posts: n/a
Default ELO Rating Inflation?


"Dr. John E. Upham" wrote in message
...
Having not been chess active since 1978, one thing that staggers me is the
number of players who are now "stronger" than Fischer, Hubner, Portisch,
Mecking (all giants of the game) ever were.


You are making some assumptions when you say that being higher rated than
someone from a different era means stronger. The elo rating system is
designed to compare players of the same era as it is only a relative
numerical means of displaying a summary of one's perfomance. And as a side
point, the only people to get a higher rating than fischer's 2780 are
kasparov and kramnik. Not that this necessarily means they are "stronger" or
more dominant now, than fischer was in his time.


Indeed, all sorts of people are 2700 strength. I'd guess there must be a
factor than can be applied to correct back to pre Kasparov (Weinstein)

days?

I guess this is going over old ground, but is there a collected wisdom on
how to correct this state of affairs?

My yardstick used to be:

National Master: 2200+
IM: 2400+
IGM: 2500+
Candidates: 2600+
WC: 2700+

Have I missed the point somewhere?

Regards, John Upham




Also, there are new factors to consider such as the increased popularity of
the game, and the gigantic influence that computers have had on the game.
One theory is that computers have helped to facilitate and speed up the
development of strong players as more people have access to huge databases
of games, training software, and engine analysis.


Jason Repa


  #8   Report Post  
Old March 12th 04, 08:50 PM
chessplayer
 
Posts: n/a
Default ELO Rating Inflation?

"Dr. John E. Upham" wrote in message .. .
Having not been chess active since 1978, one thing that staggers me is the
number of players who are now "stronger" than Fischer, Hubner, Portisch,
Mecking (all giants of the game) ever were.

Indeed, all sorts of people are 2700 strength. I'd guess there must be a
factor than can be applied to correct back to pre Kasparov (Weinstein) days?

I guess this is going over old ground, but is there a collected wisdom on
how to correct this state of affairs?

My yardstick used to be:

National Master: 2200+
IM: 2400+
IGM: 2500+
Candidates: 2600+
WC: 2700+

Have I missed the point somewhere?

Regards, John Upham



An extremely interesting point. Just because a player is rated higher
today, does that mean that he is better than a player of yesteryear
who was rated lower.

It is true that today more and more people are playing chess. Many
more youngsters are taking up the game and therefore, we are even
seeing grandmasters as little as age 12. Of course computers are
playing a big part in improving one's play. With the sophisticated
computers and computer programs available today anyone who can afford
the latest pc with the latest Fritz etc. can avail himself or herself
to playing against someone of Grandmaster capabilities. Can you
imagine the youngsters 30 or 40 years ago or even 20 years ago to have
this opportunity. Therefore, the youngsters of today are probably even
playing a better quality of chess.

We could further argue that the world champion of today, (Whether
Kasparov or Kramnik) could probably have beaten any world champion of
the past. However, that would be like comparing apples and oranges. If
we see the play of today's players and compare them to the play of
players of the past we would probably find that these players (of the
present) had a superior knowledge and therfore game. However, that
comparision would be unfair as the players of today have all the vast
knowledge of the past available to them. Therefore, obviously they
should be a level above. Add to that the availability of computers and
you certainly open up a whole new dimension.

The only way to have settled who was the greater player would have
been to have something like a chess IQ. At least even a normal IQ test
would have given us some indication. This way we could have argued
that had Fischer or Capablanca or even Morphy been a youngster today
availing himself of the latest computers he could have beaten
Kasparov.

As things stand, due to the current generation availing themselves of
all that the past generations have accomplished they should be able to
play better. When we add computers to this argument we see that the
current lot also can avail themselves to play against highly
sophisticated opponents from a very young age. Also, they can get to
try out new theories against these highly sophisticated opponents.
Therefore, their play should be better.

More and more people are playing chess and more and more youngsters
are getting better at chess at younger ages than before. I myself know
of an eight year old who was rated as high as 2027 on FIDE. He is
currently at 1997. I also know of a six year old who is as good as any
1800 player I have seen. My daughter when she was only five could have
given any 1600 player a run for his or her money.

In conclusion I would like to say that yes, today's players are better
than thier counterparts from early years. This is of course due to the
fact that each new generation has the benefit of what the generation
did before them. Also, the addition of computers has totally changed
the whole scenario for chess and chessplayers.

However, if one could somehow find out what the IQ's were of all the
players of the past one could have somehow come to a conclusion that
had this player (like Capablanca, Fischer, Morphy etc.) be a youngster
today he might in all probability have beaten anyone including
Kasparov.

Regards,

Chessplayer
Reply
Thread Tools
Display Modes

Posting Rules

Smilies are On
[IMG] code is Off
HTML code is Off
Trackbacks are On
Pingbacks are On
Refbacks are On



All times are GMT +1. The time now is 07:16 PM.

Powered by vBulletin® Copyright ©2000 - 2019, Jelsoft Enterprises Ltd.
Copyright 2004-2019 ChessBanter.
The comments are property of their posters.
 

About Us

"It's about Chess"

 

Copyright © 2017