This past year was relatively light for the Ratings Committee (RC).  
The year was marked with several transitions, including the welcoming 
of our new Executive Board (EB) liaison Mike Hoffpauir, and the unfortunate 
retirement of Mike Nolan with whom the RC has been working for many 
years.  His retirement is well-deserved, but he will be gravely missed.  
The RC had several issues to address this year, some of which resulted 
in updating the rating system technical specifications as we describe 
below.  We also continue to monitor the rating system for systematic 
changes, and we describe below the results of this year’s analyses.

At the Ratings Workshop at the 2016 US Open, participants at the 
meeting proposed the idea that ratings should be accompanied by a 
simple measure of ratings uncertainty.  Such a measure would highlight 
whether a player’s rating was a reliable indicator of his/her playing 
strength, and could be used by players and tournament directors to refine 
the information available about player strength.  This idea is not new - 
in fact, a rating system invented by the chair of the RC over 20 years 
ago has ratings accompanied by a measure of uncertainty, and this rating 
system is used on other leagues for head-to-head competition.  A proposal 
was made to the EB stemming from the Ratings Workshop which was 
subsequently tabled by the EB.  In January, 2017, the RC chair raised 
the issue again with the EB liaison to propose at the January EB meeting.  
A motion was proposed by Mike Hoffpauir in which the EB would request 
the RC to discuss and report on the need to develop a quantitative 
measure of ratings uncertainty.  The vote was one in favor, and five 
against.  The issue was therefore formally dropped due to a lack of 
interest by the EB.

One of the ratings-related developments at the start of the 2017 calendar 
year was the roll-out of the Universal Rating System (URS), a project 
sponsored primarily by Grand Chess Tour (GCT).  The URS is now in use 
for rating, ranking, and inviting elite players to GCT-sponsored events.  
The RC chair was one of the team members that developed the URS.  The 
EB invited the RC chair to discuss some of the specifics of the URS to 
have a better sense for how the new rating system would co-exist with 
the US Chess rating system.  During the January 2017 EB meeting conference 
call, the RC chair answered some of the EB’s questions about the system.  
The RC chair clarified that the URS was not intended to become a substitute 
for the US Chess rating system.  The details of the URS are spelled out 
on the site universalrating.com.

Several minor issues regarding the workings of the rating system were 
raised over the past year.  One issue was triggered as a result of a 
FIDE player obtaining a US Chess membership but not having (yet) played 
in a US Chess tournament.  The rating system specifications were a 
little ambiguous about which US Chess players would have their ratings 
updated as a result of playing in foreign FIDE-rated events.  The 
intention was to have the procedure apply only to players with established 
ratings, but this was never included in the technical specifications. 
This matter was discussed by the RC, and the conclusion was to amend 
the rating system description to clarify the original intention of the 
rule.  The technical specification of the system now includes language 
that clarifies that updating players’ ratings from results in foreign 
FIDE-rated events applies only to US Chess members with established ratings.

A second minor issue involved Mike Nolan noticing towards the end of 
2016 that 78 players had ratings between 1 and 99.  This was due to 
various ratings database-related issues from the 1980s through 2004, 
and connected to changes on the web-server where ratings could remain 
below 100.  Mike and I agreed that these players’ ratings should be 
moved up to 100, the minimum possible rating.  In actuality, a player’s 
absolute floor according to the technical specifications is higher, 
but these 78 players’ ratings would increase above 100 once they played 
in their next tournament.

The third minor issue was related to initializing online ratings.  
A question arose late in 2016 whether online Quick Chess (QC) ratings 
initialized from online blitz ratings should be based on 10 games.  
A discussion among the RC members acknowledged that online blitz ratings 
initialized from online QC ratings were based on 0 games, so why the 
asymmetry?  The result of the RC discussion was to reset the number of 
games on which online QC ratings would be initialized from blitz ratings 
(assuming initialization from blitz ratings according to the order of 
precedence described in the technical specifications) to 0.  This 
change has been made to the technical specifications.  

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 rising 
for this "stable" group of players, and that measures in the last 
couple of years have not counteracted the inflationary tendency.  
Depending on the summary measure one uses, ratings among 35-45 year 
olds who compete regularly have increased about 13-14 rating points 
from Dec 2015 to Dec 2016, and are now on average about 30 points 
higher than they were in 1997.  The RC recommends to increase the 
bonus point threshold, currently at B=12, to B=14.  We will work with 
the executive director to make this change.  We will continue to monitor 
changes in the pool, and in particular whether the inflation slows down 
or reverses in the upcoming year after the bonus point threshold change 
is implemented.