I recently contacted the US Chess office about a player, uschess.org/msa/MbrDtlMain.php?30255425, who signed up for a tournament I was helping to run whose FIDE ID was not linked to his US Chess records. The office added his FIDE ID to his record and also manually applied his converted FIDE rating to his US OTB Chess regular rating.
The issue is don’t think his converted FIDE rating should have been manually applied to his US OTB Chess regular rating since this caused his initial US Chess OTB regular rating to be based just on his converted FIDE rating of 1700 based on 5 games but the current rating regulations should have taken into account both his FIDE and US Chess online regular rating in determining his initial US Chess OTB regular rating rating. Am I correct here?
I don’t think so. It’s possible that his FIDE rating should have been used to initialize his Online Regular rating and then that rating to initialize his OTB regular, but I think that would be a mistake. OTB should trump everything.
After his FIDE ID was added to his record, his FIDE rating was used to initialize is US Chess online regular rating, uschess.org/msa/MbrDtlRtgSupp.php?30255425. This is why his US Chess online regular rating changed from 1545(P08) to 1633(P13) without him playing any more US Chess online rated games.
Whether OTB should trump everything is irrelevant here. I am asking what should happen based on the current rating regulations. Based on my reading of the current rating regulations, his initial US Chess OTB regular rating should have taken into account both his FIDE rating and US Chess online regular rating in determining his initial US Chess OTB regular rating.
As I understand it, here’s how the current blending formula should apply here. Tom, feel free to correct me as needed.
His first event (online regular rated) should be initialized using his FIDE rating, re-scaled to US Chess. This would produce a pre-event rating based on 5 games since his FIDE rating is under 2100.
For his first OTB event, the initialization would be blended using the post-event online regular rating AND his FIDE rating, again re-scaled. I haven’t done the math but I suspect this would produce a pre-event rating based on 10 games.
I’ll advise the ratings department, there has been some turnover there and not everyone may be familiar with how the blending formula works, they’ve only been around for a little over a year.
It gets more complicated with Canadian ratings, because we don’t have an online time-series history of them like we do for FIDE and US Chess ratings.
Yes, and it looks like this was done correctly after his FIDE ID was added to his record, uschess.org/msa/MbrDtlRtgSupp.php?30255425, since his initial post event US Chess online regular rating changed from 1545(P08) to 1633(P13) without him playing any more US Chess online rated games.
Yes, this is my understanding as well. It looks like since his converted FIDE rating was incorrectly manually added to his US Chess OTB regular rating, the system initialized his US Chess OTB regular rating based only on his FIDE rating.
Thanks Mike. I just emailed the ratings department on this as well.
Inserting the adjusted FIDE rating hasn’t been necessary for OTB regular rated events for some time, because we have a database of FIDE ratings by month going back over a decade. It may have still been used for some other ratings systems up until the blended formula went into effect, last year, I think.
You seem to be talking about both apples and oranges, and lemons and limes.
As far as I know FIDE never uses U.S. Chess ratings (nor ratings from any other nation’s rating system) to initialize FIDE ratings. FIDE just waits until the player has played in a FIDE-rated event or two and then assigns a rating based on their own internal calculations. Somebody in the know please correct me if I’m wrong about this.
Going the other way around, I’m pretty sure U.S. Chess occasionally uses a player’s FIDE rating to initialize his U.S. Chess rating, but only if the player would otherwise be unrated in U.S. Chess (or perhaps provisional based on fewer than N games, and I don’t know what N is).
Above all, I’m pretty sure U.S. Chess never tries to tell FIDE what a player’s initial FIDE rating should be.
Adding a player’s FIDE ID to his U.S. Chess record, in cases where the former was missing for any reason, is (or would be) simply a useful service U.S. Chess performs (or could perform) for everybody’s benefit (the player, the TD, FIDE, etc).
This is correct as far as I know but this has nothing to do with what this discussion is about Bill.
The current rating regulations state:
“If a player has no rating under a given system, as many as seven other sources may provide rating information: the five other US Chess ratings, FIDE (Section 2.1) and cfc (Canadian, Section 2.2). These may vary in quality because of the number of games on which they are based, the age of the ratings and general quality of the rating system or its conversion. A weighted average of these will be used as an initial estimate of the player’s rating—the weights will be expressed in terms of the number of ”games” worth of information the rating provides. If none of these ratings exist, the initial rating will be the age-based rating (Section 2.3) with the tournament games having been played, N, set to 0.”
Therefore, for the player I cited, I believe his initial US Chess OTB regular rating should have taken into account both his FIDE and US Chess online regular rating in initializing his US Chess OTB regular rating but since his converted FIDE rating was incorrectly manually added to his US Chess OTB regular rating, his initial US Chess OTB regular rating was only initialized based on his converted FIDE rating.
Alex, the idea behind the blending process is to use all the information we have on someone at the time that they play their first event in a given ratings system, which means the 3 Online and 3 OTB ratings systems we maintain plus FIDE and Canadian. In rare cases if someone comes in with a rating from another foreign country where we have some reasonable data on how those ratings compare to US Chess ratings, those could be entered manually, but I don’t think I’ve seen a single case of that since we went to the blended system.
We do not currently have any interaction between the OTB/Online/FIDE/Canadian ratings information and correspondence chess ratings. We’ve had some discussions about redoing the correspondence ratings system, which still uses formulas that date back to at least the mid 70’s, but are a long way from even testing the impact it could have on CC ratings much less on OTB/Online ratings.
The office should be made aware not to insert the adjusted FIDE rating anymore since then the system doesn’t apply the blended formula to initialize a players rating like it should.
Note that as published the blending formula would use the player’s FIDE rating when initializing his ONL-Regular rating and then use it AGAIN when initializing his OTB-Regular rating.
That’s by design. The more recent the data, the higher it is weighted.
With the turnover the office had in the past two years, coming and going, there’s always some additional training that needs to be done.
I used to work for a guy who often said that it took six months for a new employee just to figure out where the bathrooms are. He wasn’t exaggerating much.
Right, and the more recent data (the post-event online-regular rating) would be weighted a bit higher (relatively) than the older FIDE data.
The blending formula seems complicated, but it is actually simpler than the hodge-podge of initialization formulas we used before. And if/when a new ratings system came along, it could be incorporated into the blending formula fairly easily.
OK, now I get it. Micah’s original typo made me think that he thought either that U.S. Chess could determine FIDE ratings, and/or that FIDE sometimes used U.S. Chess ratings to initialize its own.
I suppose you could say that what actually happened is that U.S. Chess re-initialized its copy of this player’s FIDE rating to match FIDE’s real version, then used this corrected copy to do a blend of the player’s FIDE and U.S. Chess ratings.
What do you mean by “U.S. Chess re-initialized its copy of this player’s FIDE rating to match FIDE’s real version”?
The issue here is that the system only used the players FIDE rating to initialize the players US Chess OTB regular rating but it should have used a combination of the players FIDE and US Chess online regular rating to determine the players initial US Chess OTB regular rating. The reason this happened is that his converted FIDE rating was manually added to his US Chess OTB regular rating.
Actually, what they (incorrectly) did was to post a ratings corrections entry to manually set the player’s regular OTB rating to 1700/5 as of 8/1/22, before his first event.
They need to void that entry, after which a rerate would fix both events.
From what I can tell, the on-line tournament should have been initialized at 1669/5 from the FIDE rating. That looks like that was done. The recent tournament should be a blend of the 1633 OL rating (game weight around 4) with the converted FIDE 1669 (game weight 5).
I’m not sure I understand the 1700/5 (not that it matters much). It’s not like 20+1.02 x FIDE is that hard a calculation.
1699.98/5 was the initialized rating (using just the FIDE rating) according to my log files for August of last year.
The event hasn’t been rerated yet, as it was submitted for rating after today’s rerate completed, so finding the initial rate is a bit more work, but I suspect the manual regular-OTB rating correction overrode the blended formula.