The US Chess Federation implemented a change in the K-factor for ratings calculations. The multiplicative parameter K governs how quickly the ratings catch up to a rapidly improving (or deteriorating) player. A high K opens the door for large fluctuations in ratings from one tournament to the next. A small K represents a stable rating, which may not be accurate for someone moving up the ranks.
Statistically, the chess skill of a master varies less from day to day than that of a novice. The rating system accounts for this trend by decreasing the K-factor as the rating increases. This article ignores the secondary effects of the number of rounds played and bonus points. The plot above assumes 4 rounds. K decreases slightly for 5 or 6 rounds.
The biggest change affects USCF members rated between 1500 and 2200, who will see an increase in K of 20% or more. Those between 1900 and 2100 can expect a 40% boost in the volatility of their rating. No doubt, all of the elite juniors should welcome this news.
The second tweak targets only masters who inflate their rating at action tournaments (G/30 through G/60). They can still play, but will no longer gain or lose as many rating points. Indeed, master ratings should be determined by the results at serious (slow) time controls.
No comments:
Post a Comment
I have allowed public comments again, but I must approve a comment before it appears. My decision is final. Here are a few requirements:
1. Keep your comments on chess.
2. Please make constructive remarks, both if you agree or disagree.
3. Avoid insults, name-calling, accusations, profanity and adult language.
I would prefer that you sign your name, but this is not a requirement. That said, I am more likely to reject a negative comment from an anonymous poster.