Is it possible for a USCF ratings floor to be higher than a player’s highest ever rating? If I recall correctly, there used to be a policy that winning an under-2000 prize of $1,000 or greater creates a floor of 2000. Is that still in place?
Assuming yes, consider the following case. An 18 year old player has been steadily progressing since he was 12 or so and goes from something like 1850 to 1929 by winning a class section in which he wins over a $1,000. The rules say he gets a floor of 2000, so I suppose he should be happy that he’s made expert. On the other hand, his next tournament he ends up with a post-rating of 1925, so he’s figuring that his real strength might be pretty darn close to his non-floor rating. He’d just as soon enter Class A sections until he really gets up to 2000 strength. How do you suppose the office would respond to a petition to remove his 2000 floor? After all, his record doesn’t look like he’s been sandbagging.
It’s a bit more complicated than that, but yes, someone rated under 2000 who wins a class prize of $1000 or more is supposed to get a higher floor. (It may not be 2000, depending on the cutoff of the section in which the prize was won.)
If the player’s post-event rating is below the floor, the post-event rating will be raised up to the new floor.
It can take several weeks for this to happen, though, first the tournament must be submitted for rating along with the list of class prizes of $1000 or more. (With Walter Brown still out of the office, I’m not sure who’s handling the class prize floors today.)
Once a ‘class prize’ floor is in place, it would take a significant record of below-the-floor performance in order for the USCF to consider lowering that floor.
It is a virtual certainty that an A player who wins a $1000 class prize is actually playing at expert strength, if not master.
Tournaments with large class prizes always attract hordes of under-rated players. Some are deliberate sandbaggers, others have recently experienced a legitimate quantum leap in their playing strength.
You can bet dollars to donuts (oops, that’s about an even bet these days) that big-money class winners in multi-section events are playing at least 200-300 points above their published ratings.
That’s an interesting perspective. You could argue that the bump to 2000 is for two reasons: preventing sandbagging (at least in the future), and acknowledging that at such an event, many players are under-rated.
Just for fun, I added a class (200) to each of his opponents’ ratings and calculated his post-rating. It would go 1847==> 1986, so almost but not quite to expert.
Theres no doubt he was playing well that weekend (performance rating of 2225). The question is how much above his “real” strength he was playing that weekend.
The best data I can imagine using to estimate his “real” strength is the linear regression of his supplement ratings on the date of the supplement. That regression fits well (R^2 = .95), and predicts a rating of… about 2000 for Aug, 2005. So, what the heck. Call him an expert.