Universal Rating Converter

General discussions about ratings.
Post Reply
John Foley
Posts: 387
Joined: Thu Oct 15, 2009 9:58 am
Location: Kingston-upon-Thames
Contact:

Universal Rating Converter

Post by John Foley » Sat Sep 28, 2024 4:19 pm

From NoseKnowsAll Blog reporting on Lichess.
ECF has reported an easy way to compare ECF and FIDE ratings. Based on my results, I believe their measurement underestimates a player's FIDE rating for the low end and overestimates a player's FIDE rating for the high end. If you want an extremely simple estimator, I think a more accurate estimate is ECF = 1.4 * FIDE - 715 under 1940 FIDE, and ECF = FIDE + 60 for players over 1940 FIDE. If you want to flip the mapping around: FIDE = 1/(1.4) * ECF + 510 for players under 2000 ECF, and FIDE = ECF - 60 for players over 2000 FIDE. Hopefully, this more accurate mapping is clean enough for ECF members to prefer instead.
Who knows? He could be right.

Nick Burrows
Posts: 1946
Joined: Sat Aug 14, 2010 12:15 pm

Re: Universal Rating Converter

Post by Nick Burrows » Sat Sep 28, 2024 4:52 pm

It works almost exactly for me. 1980 Fide / 2042 ECF

Keith Arkell
Posts: 1083
Joined: Tue Dec 30, 2008 2:10 am

Re: Universal Rating Converter

Post by Keith Arkell » Sat Sep 28, 2024 9:32 pm

The USCF ratings used to be much too high. Have they become a bit more aligned in recent times?

Brian Valentine
Posts: 626
Joined: Fri Apr 03, 2009 1:30 pm

Re: Universal Rating Converter

Post by Brian Valentine » Sat Sep 28, 2024 10:09 pm

Hi Keith,
If you look at #65 in the discussion, he describes how he fitted the USCF/FIDE curve. It's an interesting way of doing things that would not have been available until quite recently.

NickFaulks
Posts: 9258
Joined: Sat Jan 02, 2010 1:28 pm

Re: Universal Rating Converter

Post by NickFaulks » Sun Sep 29, 2024 12:21 am

Keith Arkell wrote:
Sat Sep 28, 2024 9:32 pm
The USCF ratings used to be much too high.
Not too high, just different. They can only be too high if you believe there is any reason why the two should be closely related.
If you want a picture of the future, imagine a QR code stamped on a human face — forever.

Brian Beattie
Posts: 49
Joined: Tue Feb 20, 2024 2:02 pm

Re: Universal Rating Converter

Post by Brian Beattie » Sun Sep 29, 2024 9:17 am

I suspect the USCF will implement minor tweaks here and there such that the USCF 2200 rating will continue to be slightly easier to achieve than FIDE 2200. Simply to motivate more players to keep playing in search of an NM title. And as long as the disparity between the two isn't too great NM players can claim CM equivalency. When I eyeballed the data it appeared that only half the players that reached NM would have actually achieved CM.

Keith Arkell
Posts: 1083
Joined: Tue Dec 30, 2008 2:10 am

Re: Universal Rating Converter

Post by Keith Arkell » Wed Oct 02, 2024 6:30 pm

NickFaulks wrote:
Sun Sep 29, 2024 12:21 am
Keith Arkell wrote:
Sat Sep 28, 2024 9:32 pm
The USCF ratings used to be much too high.
Not too high, just different. They can only be too high if you believe there is any reason why the two should be closely related.
When, for example, a chess engine is assigned a rating, it is presumed that players on both sides of the Atlantic have a shared understanding of the level of play they can expect. I.E. This engine plays at 1700 on level 8’ ought to mean something both in the States and here.

NickFaulks
Posts: 9258
Joined: Sat Jan 02, 2010 1:28 pm

Re: Universal Rating Converter

Post by NickFaulks » Sat Oct 05, 2024 11:59 am

No doubt, but the mechanism to achieve that does not exist.
If you want a picture of the future, imagine a QR code stamped on a human face — forever.

Post Reply