Chess.com VS USCF Rating

There is no accuracy.

As mentioned prior, many factors determine.

1) Some rarely play over the board but play a ton of bullet (aka garbage chess) that they get a false reading on their skill.  You could be 2400 bullet and yet suck at chess aND are really 1300.

2) You get those like me who are 2091 over the board and don'the give 2 hoots what their online blitz rating is.  Someone on another post tried to say I was a 1973 player.  Must have been my current blitz rating.  Don't know, don't care.  I am roughly a 2100 player and have hovered between 2050 and 2150 for the majority of the last 5 years over the board.

3) Some players might, by coincidence, see their online and OTB be nearly the same.

Like I said, I fall in category 2, but others fall in 1 or 3.

If you really want to know the true skill of a player, over the board slow time control ratings tend to be the most accurate.  You do not get lag in OTB, and you do not get who the F cares players there either like you do online.

Hello, I am joining this old thread. I am interested to hear what more experienced players have to say about this. Some players, particularly younger players, play online a lot more than "over the board," because they are online a lot, because they can't drive, etc. So, knowing this correlation would be helpful. Otherwise, kids can get a potentially inflated impression of their skill level before they ever go to a tournament. Also, I have heard that some websites inflate the ratings as a marketing tactic?

I found an old threat that had some actual data on this:

https://www.chess.com/article/view/chesscom-rating-comparisons

I wonder if anyone has updated data with more data points?

Thanks! 

here we go again

My Chess.com is going down while my USCF is going up. I wouldn't draw any conclusions from this limited data.

While there is of course a loose correlation, the variance is so high that trying to make anything of it is pretty much impossible. Totally different pools of players, different way of playing, different time limits, the presence of cheating affecting the ratings, The presence of players here who are playing just any old junk on a lark and don't really care a hoot, differences in the way you are originally assigned a rating, and the list goes on and on.

Yes, I get the attraction of being able to convert one to the other. This is way more convenient. But at the end of the day, you really cannot predict your USCF rating with anything approaching usefulness by using your rating here.

CC ratings are generally 200 points higher than your USCF unless you sandbag or cheat. 

Thank you for the explanation. That is helpful.

Elo and Glicko are only meant to measure accuracy within a pool. Someone can do a statistical analysis of comparison between USCF and chess.com but it would mostly be a point in time. For the OP, I see that your rapid is closer to your USCF. perhaps you just suck at faster time controls.

One clear difference between chess.com and OTB is that people say things that they would never say in person!

Agree. Youtube chess hustlers ... that's mostly in good fun though

I haven't read all the previous posts to see if somebody has already suggested this, but the obvious way to compare ratings by different algorithms is by their corresponding percentiles.

I just started playing OTB in 12/2018 and have played 22 standard games. My USCF rating is 1454 which is similar to all my chess.com ratings +/- 100 elo as I fluctuate between 1300-1600 depending on the month. I tend to go on good runs and bad runs but if say over the last year on chess.com my average elo has gone from 1400 up to around 1475. I do wish chess.com had an annual report or stats on your long term trends. Regardless, my comparison between my Chess.com and USCF might be seriously flawed as I mainly okay blitz and bullet on chess.com where OTB I play standard long games. 

I am 1440 USCF and 1200 on Chess.com

No shame in that either!

It seems very strange to imply that 10 second bullet computer chess is particularly similar to OTB chess.

Really? How does that work?

It doesn’t, but you can play USCF rated games here. I played in at least one USCF event most weeks from early July to mid-December. One week, I played in four.

Ah, that's really cool. I'd like to try that. I assume/hope they have some solid measures in place to avoid cheating since it's online?

Plenty of anti-cheat detection on this site. Additional scrutiny of some sort gets applied to USCF rated events. Most events are rated one to two weeks after they are played.

https://www.chess.com/club/uschess-members-only

I just counted. I’ve played in 41 USCF rated online events in 2020. 38 were on chessdotcom. The others were on ICC and organized by the Continental Chess Association. Most on chessdotcom were Morning Membership events, but there were others.

The CCA events required Zoom so they could watch me play. Another layer of cheat detection.