What do you think about the proposed changes to initializing player ratings?
I do not wonder about the calculations for online rating, which use a carry over from other rating systems. This is similar to what was done in the past when US Chess first established Quick ratings. However, I do wonder if rating floors will be carried over. Way back when Quick ratings were first established I was a regular rated A class player with an 1800 floor. I have never been over 2000 in the Quick rating system, but I did have a 1800 floor for Quick chess. I know this for a fact as there was a Quick rated event where my Quick rating should have gone below 1800, but did not. The next Quick rated event would have brought it back over 1800, but it clearly showed that my Regular floor was carried over into the Quick rated system. Should and will floors be carried over into online ratings?
Larry Cohen
I vaguely recall you stating in a previous thread that your quick rating has never been over 2000 and I showed you from your tournament history that your quick rating has been over 2000.
Your quick rating got over 2000 from this event, which it says you directed as well!: uschess.org/msa/XtblMain.php?199305132000
Although floors are kept internally in all 6 ratings systems, they aren’t generally published except for OTB regular ratings.
Moreover, a floor doesn’t exist until a player’s rating is established. And because a rating initialized from other data is not yet an established rating, it means there is no floor in that ratings system yet. So floors are NOT carried over from other ratings systems at this time, the floor is only set once the player’s rating in THAT ratings system becomes established.
Money prize floors are exceptions, I guess, but I’m not sure we’ve ever had a player earn a money prize floor before a regular OTB rating became established.
But floors are political devices, not mathematical ones.
Why?
In part because they’d clutter up the page even more than it already is, and also because it isn’t clear that except possibly for OTB regular ratings the floors serve a useful purpose.
Many people, myself included, think floors exist primarily to combat sandbagging and other forms of ratings abuse, mainly in reference to large money events which, until recently, have only been in OTB regular rated events. The 2200 OLM floor, guarded for many years by Jerry Hanken, is said to impact participation by older masters, but I’m not aware of any statistical evidence of that.
I think there may have also been concerns tied to the long-running debate over inconsistencies between OTB regular ratings and other ratings systems. This started out as regular/quick issue but now there are 15 possible two-way inconsistencies.
An issue that as far as I know has never been discussed is whether floors from one ratings system should impact floors in other ratings systems. (This applies both to peak rating based floors and money prize floors.)
Clearly there are issues that policy-setters may want to review before we expand the visibility of floors in other ratings systems.