The Ratings Committee (RC) had a busy year, and expect future work to become even more involved now that RC is charged with monitoring five rating systems (the regular, quick and blitz over-the-board systems, and the quick and blitz online systems on the Internet Chess Server [ICC]). Most of the issues confronted by the RC were in an advisory capacity, most notably our weighing in on issues involving match rules and on rating translation from regional rating systems to the USCF system. One of our main outstanding tasks, the development and testing a method to connect the Quick Chess (QC), blitz, and regular over-the-board rating systems, has been on hold to address the more pressing issues we have tackled over the year. From a recent analysis performed by Mike Nolan, blitz ratings are noticeably lower on average than regular ratings for players having both, so we will need to monitor this discrepancy over time. In this report, we describe our main achievements over the year, and several of the minor issues we have addressed. The RC was approached last summer before the US Open to address issues relating to the comparison of ratings from the Chess Express rating system (CXR) and the Northwest Scholastic rating system (NWSRS) to the USCF system. The issue arose because players competing in scholastic tournaments who were USCF-unrated would frequently have ratings in other regional systems like the NWSRS and CXR, and the Executive Board (EB) felt that converting ratings from these systems to the USCF scale would be helpful particularly for sectioning and prize eligibility. Mike Atkins, one of the liaisons to the RC, sent us data on players having ratings in both the NWSRS and USCF systems. A formula was derived to convert NWSRS ratings onto the USCF scale: If 500 < NWSRS <= 1400, then USCF = NWSRS - 200 If 1400 < NWSRS <= 2000, then USCF = NWSRS*1.333 - 666 If NWSRS >= 2000, then USCF = NWSRS The analysis that produced this conversion consisted of players who had sufficiently reliable ratings (based on game counts) in both the NWSRS and USCF systems. We offered this conversion to be used for initializing USCF-unrated players. ratings with NWSRS ratings. The RC was asked to opine about the number of games on which to base a converted rating. The recommendation by the committee was to having the converted rating be based on 0 games. Subsequently, it was made clear by the EB that the main purpose of the conversion was for prize eligibility and sectioning, with initializing ratings in the USCF system being of secondary importance. We responded to the board by explaining that the conversion formula above is not particularly reliable on a case-by-case basis. The range of USCF ratings for players with a small window of NWSRS ratings is large. The conversion formula can only be interpreted .on average.. Given that the conversion from NWSRS ratings to USCF ratings is not particularly reliable, one question is whether it is better to section players with converted NWSRS ratings who are USCF-unrated (or with stale USCF ratings) rather than not use the NWSRS information at all and just treat the players as unrated, or even use an age-imputed rating for sectioning. The RC recommendation was to use the conversion formula rather than an age-based rating for players with 50+ NWSRS rated games. When the NWSRS rating is based on fewer than 30-40 NWSRS games, then the age-based rating described in the USCF rating system documentation is probably just as reliable. Based on the discussion of these issues, a revision of delegate motion DM 12-28 was to be reexamined. The motion requires players to disclose ratings derived from other over-the-board systems (such as the NWSRS and CXR systems), and if the specific system has an associated rating translation formula like the one above then this converted rating can be used as information relevant for sectioning. A newly formed committee was created in August 2014 to study the delegate motion, and RC representative Tom Doan agreed to participate on the committee. In May 2014, Mike Nolan brought up an issue to the EB and the RC regarding match rules. The rules state that matches are only eligible between players with established ratings. Although provisionally rated players are not supposed to play in rated matches, sometimes it happens anyway. For example, an Under 1200 section may only have two players in it, and one of them with a provisional rating. Such occurrences are not uncommon, and they have been rated without concern for the match rules. If such .matches. involving provisional players are to continue, a question that was raised is whether to limit the rating change for provisional players. The specific question the RC was whether to change the current rules wording from .A player may gain or lose no more than 50 rating points in a match. to .A player with an established rating may gain or lose no more than 50 rating points in a match.. The reaction of the RC was that this problem is not a ratings issue; it is an issue of inappropriately scheduling repeated games between players when at least one is provisional. No change was made to the rules wording. The RC chair suggested a possible future change to the rating system to address this issue. The idea is to incorporate into the rating formulas that repeated games against a single opponent convey less information about each player.s rating than when competing against multiple opponents, so that the rating change in the former case should be smaller than in the current rating formulas. The vice-chair of the Senior committee referred a matter to the RC. The question was whether seniors who request that their rating floor be lowered should have their petition honored by the USCF office. The rationale for the request was that some seniors may not have playing strengths as high as in previous years, so these players would be over-rated and not likely to compete realistically for section prizes in tournaments. The USCF office has traditionally handled such requests on a case-by-case basis. The issue was raised, however, whether the USCF office should be provided greater guidance through formulas that the RC might derive. For example, the formulas to estimate a senior.s actual rating strength might be a function of the player.s age, some indication of tournament activity, the length of time on the player.s rating floor, highest attained rating, and so on. The feeling of the RC chair was that it would be more realistic to continue to give the USCF office discretion to make floor-lowering decisions for senior players, and that a formulaic approach may be more rigid than necessary. The motion to store ratings and compute rating changes in floating point accuracy was approved by the EB in February 2014. Mike Nolan implemented the conversion since the last RC report and now ratings are being stored with floating point accuracy. Several minor issues arose this past year related to the conversion. For example, one question was how the conversion to floating point values affected triggering a rating floor. The decision was first to round the rating to the nearest integer (using conventional rounding rules) and then apply the rating floor algorithm of subtracting 200 from the highest integer-rounded rating. Thus, if a player.s highest attained established rating was 1899.5, which would round to 1900 using conventional rounding rules, the player.s new floor would be 1900.200 = 1700. Similarly, a player attaining a floating point rating of 2199.5 (or higher) would be rounded to 2200 for the purpose of being considered a master. Several minor issues were addressed over the past year that are worthy of mention. First, a question arose about whether two players could compete in a blitz match if they had no blitz ratings but did have established regular ratings. In this instance, the decision was to disallow such matches because the intent of the rules was that players were required to have established ratings in the system in which a match took place. Second, the RC was alerted to the change in the FIDE rating system rules in which players could receive ratings with as few as 5 games instead of 9. This would have implications on the rating conversion formula when estimating USCF ratings from FIDE ratings and also on updating USCF ratings from FIDE events played outside the US. Finally, the rating system specifications underwent a major rewriting this past January. The main change in the rating system specifications document was to provide better organization of the way in which ratings from each of the five rating systems (over-the-board and online) are initialized. Additionally, several unrelated corrections were made to the document. The ratings document can be downloaded from http://www.glicko.net/ratings/rating.system.pdf . 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 now 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. The concern of ongoing rating inflation due to the new formulas has resulted in the RC proposing to change the bonus point threshold from B=10 (implemented March 2014) to B=12 in the rating formula. We will work with the EB to approve the change.