The Ratings Committee (RC) this year was charged with two major tasks. The first was to investigate methods for monitoring systematic drift in ratings. There has been some anecdotal evidence that ratings had been on the decline in the 1990s, and in response to this assertion the RC has begun examining approaches to detect significant deflation or inflation in ratings. The second task was to prepare a proposal for a norm-based title system. The Delegates in 2002 passed a motion to adopt a Title system, giving the Executive Board the authority to implement it. As many USCF members may recall, a norm-based title system was in effect in the early 1990s, but contained a flaw that enabled players to earn titles easily in events with many rounds. The flaw was corrected in the mid-1990s, but the system was discontinued. We now introduce a revised version of the norm-based title system for the delegates' consideration. The monitoring system that the RC has been examining is based on tracking the rating behavior of players who have been established, active, and in an age group that would have an overall tendency to remain unchanged in playing strength. Elo's gerontology work suggests that players between 30 and 45 years old do not improve or become worse, on average (of course, there is plenty of room for individual variation). Our data analyses were based on annual rating lists from December, 1992 through December, 2002. After trying various criteria for inclusion in our analyses, we focused on examining rating changes for players between 35 and 45 years old in year "Y" who were established and active for a specified number of years prior to year "Y". The goal of such an analysis was to determine the relationship between rating changes and the year "Y". For example, we were interested in examining typical rating changes from 2001 to 2002 for players active in the previous 5 years (1995 to 1999), compared to typical rating changes from 2000 to 2001 for players active in the previous 5 years (1994 to 1998). Some interesting patterns emerged from our analyses. First, regardless of the number of years a player was active prior to year "Y", average rating gains in 1996 through 2000 were negative, and in 2001 and 2002 they were positive. Essentially this means active players' ratings (no matter how "active" is defined) were decreasing steadily from 1996 through 2000, and then were increasing in 2001 and 2002. This finding is not surprising: In 1996, the 100-point rating floors were dropped without a counteracting measure to stem off deflation, so ratings started to drop. In 2000, the new rating system was implemented with a low bonus threshold for two years that intended to re-inflate ratings back to the 1996 levels. A second finding of our work is that rating changes from year to year for active players tend to be positively skewed. This means that, regardless of whether there were rating gains or losses in a year, there were more active established players whose ratings increased substantially over one year (e.g., more than 200 rating points) than decreased substantially. We also noticed that the conclusions of our analyses did not change by examining the 35 to 40 years olds separately from the 40 to 45 year olds. The overall conclusion of our work is that monitoring should involve the analysis of rating changes for established players active at least four or five years, though the exact mechanism for determining when the rating system needs adjustment has not yet been worked out. We plan to continue our monitoring work this coming year. In the 1994 and 1996 RC reports, we introduced and developed a reconstruction of the USCF Title System which had been approved for implementation by the Policy Board. The Title system was constructed to reward players with titles based on qualifying performances. Before the system was implemented, interest developed in devising a method of rewarding achievement by keeping ratings above specified thresholds. This led to the Life Achievement proposal, which was approved by a new Executive Board in 1998. While neither system was implemented, the current focus has been to improve and implement the Title system from the mid-1990s. This year we reintroduce that system with refinements as a proposal for issuing titles to players competing in USCF events. The basic premise of the proposed Title system is to award permanent titles based on sustained performances at particular rating levels. To be more concrete, a player who is vying for the expert title would need to demonstrate several qualifying tournament performances in which his game results would be considered notable for someone rated 2000. For each qualifying performance, a single norm or multiple norms are awarded. Once either three, four, or five norms are collected (more norms are required for higher titles), a title for that level is issued. Norms and titles cannot be lost through poor performance or inactivity. The new main feature of the currently proposed system is to add a minimum rating threshold, so that a player who earns five norms for the expert title must also have, or have had, an established rating of at least 2000. The proposed system has no effect on the original Life Master title, which requires 300 games above a rating of 2200. The complete proposal can be accessed online as a pdf file at http://math.bu.edu/people/mg/ratings/titles2003.pdf The committee has engaged in two other activities worthy of mention. First, the bonus point threshold, which was set in the year 2001 to reinflate ratings for the next two years, was supposed to be reset to its normal value in January 2003. Apparently, this has not yet happened (as of May, 2003). The RC is currently working with the USCF office to make sure the change is implemented soon. The other activity is that the RC will plan to work with the USCF office on the redesign of the USCF relational database environment. The RC recognizes the importance of this work in its impact on the administration of ratings, so we will assist with the conversion.