uscf ratings over the years

Sort:
breaker90

Does anyone know if the USCF chess ratings decreased over the years? I've heard that in the 90s, USCF changed their rating system that effectively lowered everyone's rating by 100 or so points. Is this true and was there another time where USCF dramatically changed their rating system?

HungryChild

They do have some kind of rating adjustment thing...Not sure the purpose of it.  Its not drastic or anything.  My guess is its to adjust for 'local ratings'.

I've seen my rating jump by a few points when I've been inactive. 

For example, I play a  weekly event.  I usually lose there.  The strength of the average player is pretty astounding.  Then I go to a casual tournament on a weekend, and guys rated 200 points higher than me seem very weak.

So, my guess is the uscf tries to balance out stuff for local ratings like that.  In my case, its all in the same city, but different player bases.

No idea if what

breaker90

Yeah, I heard the USCF changed it in the 90s because the rating system was not close to FIDE's. BTW, it still isn't today. Apparently USCF is higher by 100 points or so.

HungryChild
breaker90 wrote:

Yeah, I heard the USCF changed it in the 90s because the rating system was not close to FIDE's. BTW, it still isn't today. Apparently USCF is higher by 100 points or so.

I was thinking they were like 'ok...guys in chicago, you get +30 points,...guys in south dakota, you get 0 points, your player base is average, etc'.

This is my imagine-ings, totally unfounded by any facts .

breaker90

@cookiemonster, no, the USCF rating system today is mathematically different from FIDE's. Something to do with the K factor and other things (http://en.wikipedia.org/wiki/Chess_rating_system). It's similar to why chess.com ratings are way off from uscf ratings too. Different equation.

leiph18

Also people may misunderstand that it's not a measurement like height or how fast you can run a mile. Ratings are arbitrary in that the numbers themselves mean nothing. It's the difference between them and other ratings in the pool that matters.

So e.g. cookiemonster is correct saying you can't compare across pools... this is true even if the formulas are the same. This is because they're measuring a player in relation to different sets of players.

Of course there is a positive correlation. Stronger players have higher ratings. So you can also get a rough estimate. But you can't say something like e.g. Stockfish (rated ~3300 in CCRL) is ~450 points higher than Carlsen (~2850 FIDE).

breaker90

My comment was about hearing about USCF changing their system to be more aligned with FIDE. The USCF rating was so much higher than FIDE players that it was a bit embarassing when a high rated USCF opponent would get regularly beaten by a lower rated FIDE opponent. So they made the change and everyone in the 90s ended up "losing" points because the elo equation had changed. This seems to match your earlier comment about talking to other older chess players. Perhaps your 2100 friends were only 2100 because the system gave more points? Just a thought. I was hoping to find an article about it but I don't think USCF was web-friendly two decades ago.

I know you shouldn't compare different rating systems to measure their strength. I brought up comparing the USCF and FIDE system because the USCF apparently wanted their rating system to be closer to FIDE. How do you not see that USCF provides more flucuating rating points than FIDE? Still is off by 100 or so points today.

teknikyobi77

Dream on dude, you are 2100? Records say you really have a hard time reaching 2000. Both uscf and fide.  

breaker90
teknikyobi77 wrote:

Dream on dude, you are 2100? Records say you really have a hard time reaching 2000. Both uscf and fide.  

Says the guy who will never play OTB, lol