The Ratings Committee has had another busy year.  In fact,
our e-mail discussions totaled over 24000 lines this past
year.  The majority of committee work this year focused on 
proposing and evaluating replacements to the recently defunct
title system.  Last year's Delegates Meeting resulted in the
removal of the USCF Title System, leaving the USCF with no
system for rewarding achievement.  Despite the committee's
strong preference to have fixed the existing title system,
we have proposed a new title system which was submitted to
the Policy Board in February.  An executive summary is below.
The discussion that led to the new system was 
preceded by extensive discussion on Bill Goichberg's proposed
"Life Achievement Point" system, which was ultimately rejected by 
the committee.  The proposed "Life Achievement Awards" system is 
currently being reviewed by several other USCF committees before further
action is taken.  The details of the system can be found on 
my ratings web page,

http://math.bu.edu/people/mg/ratings.html

The following is the "Executive Summary" of the Life
Achievement proposal:

"USCF recognizes Life Achievement by a series of awards.  There is one
award for each rating Class, ranging from Senior Master to Class J.
Awards are based on maintaining an established rating in the Class for a
significant number of games.  The number of games required depends on
the Class.  Awards are automatically sent by the USCF office if the Life
Achievement award is for a Class no more than one Class below the
player's current rating.  Other (lower Class) awards may be issued upon
specific written request. This system is an extension of the Original
Life Master award; requirements for the Original Life Master award
remain unchanged."

Please refer to the URL above for details.

The Ratings Committee has also been finalizing the details of the 
new rating system, with several committee members involved in 
the testing and debugging of the system.  At the time of submitting
this report, the USCF has a version of the new rating system
programmed, but not fully debugged.  Unfortunately, the long-time
USCF programmer George Wang has recently left the USCF, and the USCF
office has decided to hold off making major programming changes to 
the rating system until a replacement was found and new office software 
implemented.  The most current version of the rating system details
can be found at the URL above.  

The Ratings Committee was charged with the task of evaluating the
current practice of establishing "money" rating floors.  The policy
of establishing rating floors for winners of large prizes (over $1000)
was motivated by the need to protect membership against players who
artificially manipulate their ratings to remain as low as possible
(a.k.a. sandbaggers).  On the June rating list, there were a total
of 34 players at their money floor.  The overwhelming majority
of Ratings Committee members object to the use of money floors
on the grounds that the rating system should not be used as an
anti-sandbagging device.  Other suggestions and proposals have
arisen and been rejected as a result of our discussions, including the 
ill-fated "Current Class" proposal which would categorize players
into classes (roughly corresponding to their ratings) on which they
would be sectioned and paired in tournaments.  The approach that
the committee is more attracted to involves identifying players
that may be suspected of sandbagging based on their performances in
relation to money prizes earned, and reporting this
information to tournament directors.  The Committee is still 
considering various alternatives.  Our response this year to the Policy 
Board is as follows:

"The Ratings Committee will develop, in consultation with the Executive and
Technical Directors, methods to test the reliability of rating information,
with the goal of detecting and correcting situations where a member's rating
is inaccurate due to errors in the tournament report or extraordinary
fluctuations in results.  This proposal is in response to the Board's 
desire to abolish dollar floors when an acceptable alternate method to 
combat sandbagging is found."

The Committee was also asked to propose a solution to the problem
that Quick chess ratings and standard ratings seemed to be out of sync.
Indeed, an analysis of ratings data has revealed large discrepancies
in ratings between the two systems.  The source of the problem 
appears to be that players who compete in standard events do not
compete regularly in quick events, so that a quick chess rating usually
is not current relative to the standard rating.  Several proposals
have been suggested, including forcing quick ratings to be within
a specified rating difference away from standard ratings.  So far,
there has not been substantial agreement on how to handle the problem,
though the Committee continues to address this issue.  In the meantime,
the Committee has agreed by a large majority to increase the multiplier 
in the quick chess established rating formula from 0.6K to 1.0K while
the current rating system is in place, thus allowing greater fluctuation
in ratings.

Another small task accomplished by the Ratings Committee
this year was to make sure that the absolute floor at 100
was implemented.  Last year a motion was passed that the absolute
floor be set to 100, but after implementation there were still 
ratings below 100.  This has now been repaired in the ratings code.