Another Look at the Chess.com vs. USCF Ratings Debate

Americans think they have to be #1 at everything... Sorry, Europeans play more chess, that's just a fact. Indians play an awful lot too. The United States is not the greatest chess community.

Ratings irrelevant

That's not being done here because so few people play standard on chess.com.

So basically by the Ops logic I should have a 2450 fide and by drawmasters logic I should have a 2800 fide.  

Makikihustle is completely right.  In the U.S. there are not many fide rated tournament opportunities unless you are playing in the top sections. This explains why my fide is 60 points less than my uscf.  I highly doubt there is a good correlation between chess.com blitz and uscf rating at a uscf rating of 2000+.  Also I see all these random masters who have 1800 ratings explain that.

Your idea may be good, but you express it poorly.  Comparing the FIDE ratings is ridiculous.  Comparing number of titled players is much better.  If you had some averages for ratings in national federations that would be good too.

I shouldn't have to express it at all. Its blatantly obvious, for example, that Europeans are better at chess than Americans. It's a fact. Indeed, a high percentage of the ranking members of the USCF are not originally from the United States. America simply does not have the chess culture that exists overseas.

Hah, then save yourself the trouble of sounding foolish and don't speak at all.

I would also be an outlier in your charts

Bullet-Around 2050

Blitz-Around 1750

Standard-Around 1450

FIDE Standard-1599

Here is another example I just noticed:

http://en.wikipedia.org/wiki/Luciana_Morales_Mendoza
http://www.chess.com/members/view/ollanta

Her blitz rating here is more than a hundred points lower than her FIDE rating.

If you can't understand what I'm saying, then don't respond to me. ; o

Well, anyway, I agree with you.

My logic would not suggest you have a rating of 2800 FIDE ... the range of validity of that pseudo-correlation I generated can at most be pushed to 2000 USCF (the range of Glickman's adjustment correlation) ... and of course none of the data I collected as part of my little study could say anything meaningful about much higher rated players. Correlations have limitations, and are just that: correlations. Nothing more. Extrapolation from already tenuous correlations is pure speculation and is most often misleading - as it would be in this case. 

so someone who has a 2000 uscf or fide rating is likely to have an internet blitz rating within 200 points of this. didn't we already know this without having to employ statistical analysis? 

I think it's a function of how seriously people take internet blitz. there are many active titled players who have much much higher internet blitz ratings than standard/fide ratings. I think the relationship asserted here might start to break down when you look at internet blitz ratings above 2200. 

You're a student, you've beaten Sam Shankland in blitz, and you boast a top 0.1% blitz rating here. You're close to expert level OTB.

You probably have the potential to reach IM level if you go on like this. Why not ?

Expert level is so enourmously far from IM level, I am having trouble telling whether you are being sarcastic or not?

Blitz is significantly different than otb...Beating a gm in blitz is far easier than beating them otb.  I've played a few Gms otb already and have 0 wins and draws.  Also I play extremely solidly which allows me to win on time a lot in blitz, but that results in me drawing a lot otb.   an IM level player is going to be significantly better than me in otb games, but completely beatable in blitz.  

That doesn't change the underlying statistics.

Well, in OTB tournies, players don't rage quit, always play, and don't usually have to leave half way through the match.

I saw an expert drop his queen and flip over the table.

Excellent OP. good work :0)