Provided by MARK E. GLICKMAN, Ph.D.
Over the past eight months, the Ratings Committee (RC) has been busy with a variety of tasks. In addition to taking on several new assignments, we resumed our diagnostic analyses of the rating pool and continued our work on developing a measure of uncertainty for ratings. Our efforts have yielded some noteworthy results, which we summarize below.
In the late summer of 2022, the EB liaison to the RC shared correspondence from John Fawole, a former Nigerian chess champion and current FIDE master, regarding a project in Nigeria to establish a local rating system. While the project was initially abandoned due to difficulties in implementing a proper rating system, there has been recent interest in rebooting it with the goal of adopting the US Chess rating system for use by the Nigeria Chess Federation. The RC chair corresponded with Mr. Fawole to learn more about the rating system specifications, but it soon became apparent that he was also interested in developing a tournament and ratings management system, rather than just a set of formulas for rating chess players. The RC chair directed Mr. Fawole towards freeware options for tournament management as a starting point for his efforts.
For more than two years, the rating system specifications document on the US Chess website was a “tracked changes” version. This particular description of the rating system includes changes from 2020 that involved initializing ratings from one system to another (e.g., from Quick Chess ratings to Regular ratings), factoring in the staleness of a rating from lack of activity. The RC chair noticed late in 2022 that a finalized version of the document had not yet been uploaded, so after making minor wording corrections, was able to work with the US Chess office to provide a version with the tracked changes accepted. The updated document is accessible from the US Chess web site.
The EB liaison to the RC indicated early in 2023 that US Chess was planning to revamp the Chess Coach Certification Program. As part of that effort, they needed summary statistics for the distribution of established over-the-board (OTB) ratings for US Chess members over the age of 18, which was provided by the RC chair. Further information about the progress of the Certification Program can be found elsewhere.
A question arose in March 2023 in the context of estimating the rating for an unrated player in the All-Girls National Chess Championship about converting English Chess Federation (ECF) ratings to the US Chess scale. The most recent edition of the official US Chess rulebook includes the formula US-Rating = 8×ECF + 700. This formula no longer works because the ECF in 2020 converted their ratings to the Elo scale where they are now approximately in sync with FIDE ratings. We expect that any revision to the official rules will correct this out-of-date conversion.
The RC has been working on an ongoing project to develop a measure of ratings uncertainty, dating back to the 2016 US Open when interest was expressed during a ratings workshop and from other sources that ratings should include a simple measure of uncertainty. Tasked by the EB in early 2018, the RC's goal was to create a measure that could determine the reliability of a player's rating, providing valuable information for players and tournament directors. An initial proposal was drafted by the RC chair and discussed by the entire committee in spring 2019. However, due to a change in leadership later that year, the project was put on hold for two years. Upon another leadership change in 2021, the project was revived, and the new RC chair proposed an alternative approach, differing from the one suggested in 2019. Although this alternative version was not fully developed, the current RC chair, who assumed office in 2022, has made efforts to reconcile the differences in both approaches. In the upcoming months, the RC chair will collaborate with the committee members to determine the most effective direction to pursue in order to achieve their goal of creating a reliable measure for rating uncertainty.
The RC performed a set of diagnostic analyses to monitor trends in the rating pool, resuming a practice that had been regularly carried out pre-pandemic. 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, it appears that the mean rating for this group has dropped by about 20 rating points in the past year. One possible explanation for this decline is the lack of OTB US Chess tournament activity in 2020-2021, coupled with the ongoing improvement of scholastic players. In other words, after two years, players with stable ratings may be losing points to younger, improving players who had not been able to compete in OTB games. The RC voted 6-1 in favor of lowering the bonus point threshold by 2 points to account for the lack of OTB activity over the past two years, and have submitted this recommendation to the EB liaison.
