This past year, the Ratings Committee (RC) tackled a variety of tasks.
Some of these tasks were connected to charges from last year, and
several new issues arose.  Discussions on the US Chess forums
triggered several advisory questions, and we thank members for their
input on helping to improve the rating and title systems.  We describe
the achievements of the RC over the past year, including both major
and minor issues that have been addressed.  We also continue to
monitor the rating system for systematic changes, and we describe the
results of this year's analysis below.

An ongoing issue that the RC has been addressing is the way in which
players rated under the Northwest Scholastic rating system (NWSRS) are
to be sectioned in US Chess tournaments if they are US Chess-unrated.
Last year, we provided guidance for sectioning such players with a
rating conversion formula.  The RC was asked this year by our
Executive Board (EB) liaison to review a white paper presenting an
analysis that derived a different set of formulas to convert NWSRS
ratings to a US Chess rating scale.  While we found much to like about
the paper, we concluded that several technical problems with the
method reduced our enthusiasm at the formulas the author presented.
Our recommendation was to use the formulas the RC derived last year
and described in the 2015 RC report.

Late in 2015, the RC was asked to comment on whether it was sensible
to rate games in which one player was simultaneously playing multiple
players.  Specifically, we were asked to comment on the following two
issues: (1) If a player competed in two sections of the same
tournament simultaneously, should the games be rated?  (2) Can a clock
simultaneous exhibition be rated if the "simul giver" has the same amount 
of time on the clock as other players in the simultaneous exhibition?  
The essential recommendation by the RC was "no" to both questions.  The basic
principle at work is that ratings are measuring ability in the context
of one-on-one games.  If a player is competing in multiple games
simultaneously, then the player's strength is not likely at the same
level as if he/she is playing one-on-one games.  Thus updating ratings
based on results of such games is not necessarily reflective of
one-on-one game ability.

Several motions were passed this year that involved RC discussion.
Last June, the RC recommendation that the bonus point threshold of
B=12 was implemented.  This was an increase from the prior
year's value of B=10.  Larger values of B result in less
opportunity for a player to earn bonus points.  The reason for the
increase of the threshold was that we have been starting to observe
rating inflation.  The tuning of the bonus point threshold is one of
our main tools in controlling inflation and deflation.  A second
motion that passed this year was related to money floors.  A player
had complained about receiving a money prize-based floor from earning
a moderate prize (> $2000) in a large prize fund tournament.  The RC
was asked whether the current rules were too strict in creating floors
based on prizes.  The general attitude by most members on the RC is
that money floors distort the rating system, and that any opportunity
to weaken money floors would help improve the rating system.  Together
with the EB liaison, the RC advised that increasing the money floor
threshold to $4000 would be acceptable.  A motion to increase the
money floors passed, and as of April 1, 2016, the $4000 prize
threshold for awarding floors was put in use.  The rating system white
paper now reflects this change in the rules.

An advance delegate motion from last year, ADM 15-45, was referred
back to the ratings committee.  The ADM requested wording for the new
official rules language about rating floors.  The RC suggested using
the following wording, which will be presented for a vote to the
delegates this year:

   "Rating floor. Each rated player has a rating floor. Every
   player has a personal absolute floor between 100 and 150. A player
   with an established rating has a rating floor possibly higher than the
   absolute floor.  Higher rating floors exist at 1200, 1300, 1400, ... , 2100.  
   A player's rating floor is calculated by subtracting 200 points from 
   the highest attained established rating after rounding to the nearest 
   integer, and then using the floor at or just below. For example, if 
   an established player's highest rating was 1941, then subtracting 200 
   yields 1741, and the floor just below is 1700. Thus the player's rating 
   cannot go below 1700. If a player's highest established rating were 
   1999.51, then subtracting 200 from the integer-rounded rating of 2000 
   yields 1800 which is the player's floor. If an established player's 
   highest rating was 1388, then subtracting 200 yields 1188, and the 
   next lowest floor is the player's absolute floor, which is this 
   player's current floor. A player who achieves the Original Life 
   Master (OLM) title is given a floor of 2200. The OLM title is 
   earned by playing 300 games, not necessarily consecutive, with an 
   established rating of 2200 or higher. A player's rating floor can 
   also change if he or she wins a large cash prize."

An issue that was discussed with the RC chair this past year was the
increase in Blitz events that are 20 or more rounds long.  The rating
programs were set up to support up to 32 rounds, but this may change
given the longer Blitz events that are being submitted for rating.
The RC chair was queried on whether the rating formulas needed to
address longer events.  The response was that the mathematical
underpinnings of the rating formulas are particularly sound in longer
events, so that no concern was necessary.  A related issue was whether
the formulas needed to change for Blitz matches if the match length
restriction were increased.  This is an area of further study, but
based on previous work by the RC chair the formulas should change by
dividing the difference between the total attained score and total
winning expectancy by the square root of the number of games against
the same opponent.  Currently, the rating system has no such
adjustment for competing against the same opponent multiple times.

Last July the EB liaison to the RC asked our opinion about the US
Chess partnering online sites, ICC and chess.com, to eventually offer
access to actual US Chess over-the-board Blitz or Quick ratings.  We
already have online Blitz and Quick ratings which are separate systems
from the over-the-board systems.  The RC had been asked about this
early on when ICC was being considered initially for hosting rated
events.  Our response was that over-the-board ratings should not be
impacted by results of ICC games for various reasons (cheating
potential being the most obvious).

Throughout the year, several minor changes were made to the rating
system white paper and the title system white paper.  These changes
were made to reduce the ambiguity of some of the descriptions of the
rating system, and to more accurately reflect the system that is
currently implemented.  Some of the changes included (1) correcting
the date to which the norm and title computations are retroactively
computed, (2) clarifying that norms can be earned only in regular time
control over-the-board events, and (3) clarifying that the age-based
rating formula applies to children even younger than 3 years old.
Finally, the rating system, norm and title system, and approximation
formulas document now include the new US Chess logo and refer only to
US Chess and not the old abbreviation USCF.

Each year the RC performs a set of diagnostic analyses to monitor
trends in the rating pool.  Overall rating levels deflated from the
mid-1990s through 2000 when rating floors were decreased by 100 points
without a counteracting inflationary mechanism.  With the new rating
system implemented in 2001, ratings started to re-inflate.  The
RC's goal has been to re-inflate and then maintain rating
levels roughly where they were at the end of 1997.  Our analyses have
focused on players with established ratings who have been active over
the current and previous three years and who are aged 35-45 years old
in the current year.  Based on our analyses of rating changes, current
ratings are still uncomfortably higher relative to where they were at
the end of 1997.  With the change in the "effective N" formula and 
K-factor formula made effective in May 2013, players' ratings are 
experiencing noticeable inflation relative to the formulas previously 
in place.  Last year's increase to the bonus point threshold has slowed 
down the inflation, but based on this year's analysis the RC has recommended
that no further change should be made even though ratings have
increased.  We will continue to monitor changes in the pool, and in
particular whether the inflation slows down or reverses in the
upcoming year.