Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Posting-Version: version B 2.10.1 (Denver Mods 7/26/84) 6/24/83; site drutx.UUCP
Path: utzoo!watmath!clyde!burl!ulysses!mhuxr!mhuxt!houxm!mtuxo!drutx!connors
From: connors@drutx.UUCP (ConnorsPA)
Newsgroups: net.games.chess
Subject: Chess Ratings
Message-ID: <23@drutx.UUCP>
Date: Tue, 10-Dec-85 17:30:42 EST
Article-I.D.: drutx.23
Posted: Tue Dec 10 17:30:42 1985
Date-Received: Thu, 12-Dec-85 04:18:18 EST
Organization: AT&T Information Systems Laboratories, Denver
Lines: 42

[]

The system used to rate tournament chess was invented
by Elo in 1959/1960. It was first used by the USCF in 1960,
and was adopted in 1970 by FIDE for international
ratings. The USSR is just now starting to use this system.

It is based on statistical probability theory, with
ratings following a Normal distribution with a
mean of 1500 and a standard deviation of 200.

Roughly speaking:
2700+  Kasparov, Karpov, etc. ( None active in the US. Unless Fischer.....)
2500+  Grandmaster strength (48 in the US)
2400+  International Master strength (76 in the US)
2200+  National Master (642 in the US)
2000+  Expert (2324 in the US)
1500+  Better than average. (Total of 15,000 in the US)

Using the properties of the Normal distribution you can estimate
the probable outcome of various tournament games. For example:

1) Kasparov (2715) versus Grandmaster (2500)
	Kasparov wins 3 out of 4.
2) Kasparov versus International Master (2400):
	Kasparov wins 17 out of 20.
3) Kasparov versus Expert (2000):
	Kasparov drops 1 game in 200.
4) Kasparov versus Average player (1500):
	Average player won't win unless Kasparov dies in mid-game,
		thus losing on time.

Interestingly, this rating system is quite capable of being
applied to other sports, especially where the competition
is of an individual or one-on-one nature. We could find out
exactly how good John McInroe is at tennis, or Arnold
Palmer at golf.

		Paul Connors
	Email:	ihnp4!drutx!connors
	Phone:	(303)-538-4047