On Apr 6, 1:06 am, "David Kane" wrote:

Both the USCF and FIDE rating systems have elements not
anticipated by Elo. Both essentially use the Elo algorithm and
are Elo systems.

The USCF is further from Elo than FIDE largely because it has a more
difficult population to cover. Being different from Elo is not a bad
thing. Although people can and do quibble over details in the way
organizations do ratings (I am one of these people, BTW), no one
could reasonably claim that Elo rating was the ultimate in ratings.

From what I can tell, the Elo system was not wildly different
from the Harkness system and there is nothing to Sloan's
fabrication that Elo is specifically designed for OTB chess
rather than correspondence chess. The Harkness tables are
different in detail from the Elo versions. Does anyone know the
logical basis of the Harkness tables, or what statistical
problems Harkness ratings had? It may be in Elo's book but
it has been a while since I've read it.

I suggest that you buy my book, "The Blue Book and Encyclopedia of
Chess" by Kenneth Harkness, because it contains a complete and
detailed explanation of the Harkness Rating System.

http://www.amazon.com/dp/0923891927

Sam Sloan