Online correspondence doesn't compare well to OTB... and I know the online/offline ratings of a couple dozen players here. But if you can't make 1400 online, it's very likely you aren't going to make 1000 OTB because that means you still hang pieces.
Live Long & Live Long compare fairly well to OTB. But this assumes you play with a small increment, color=random, don't hand-pick opponents, etc.
Live Quick doesn't compare.
Of course, OTB and online are always going to be somewhat different. If you want to know your OTB rating for certain, you need to play OTB. 
Right now, over about 140 games, I'm right at 1700(Online), I just started playing chess in the last couple months so i feel Like i'm getting a.... :) feel for it. I started playing the 10, 0 blitz games but I've only played about 40 some-odd games, and I doubt it compares(1150). I only play with one person OTB and I really feel the difference which is why I ask. I'm sure your right though, gotta play OTB to know my OTB ranking. :) Thanks, and I hope more will reply with their view/knowledge on the subject.
I see you didn't get any more responses, so I crunched the numbers for the sixteen people where I knew their USCF, Blitz, Long, and Quick ratings to draw some conclusions. Blitz ratings were both the most common and the most predictive, surprisingly.
What the data tells me is:
Thank you VERY much sir, I really appreciate you taking the time. Perhaps from this, someone could write a very simple program that can draw loose assumptions about OTB ratings. Then again, I suppose it really doesn't matter to most what there "Projected OTB rating" would be. :) None the less, I will continue my growth from likely "Class D or lower" to ... better. Thanks(again, I still invite more on the subject obviously).
I'd say it's around 350-400 points different. For example I'm 2000something on here whereas OTB I'm 1600something. 1400's here are generally around 1000's OTB in general.
There are exceptions of course. I beat a Master here (aj green) rated under 2000 but he was playing over 100 games at once (and simply played too fast) so it all depends. One on one OTB he'd actually be ranked much higher than on here.
It also depends of course on how many games the player in question has under their belt.
Narz: USCF 1700, Blitz 1600, Long 1550.
AnthonyCG: USCF 1200, Blitz 1200, Long 1200.
Added to my spreadsheet.
AnthonyCG> I don't agree entirely. My online rating is low, but then I don't pay as much attention as in otb games though unless it's 30 minutes.
They won't match exactly, of course, but I have 18 data points showing some correlation between USCF and chess.com live blitz ratings.
There must be some formula that could give some idea as the translation... The formula would have to entail certain variables other than simply your online rating.
1.Online Rating
2. Attention payed to games(or we could assume 100% and without falling under that catagory, one would expect to see inaccurate results)
3. Number of Games, again, this would have to be a sufficiant number otherwise inaccurate results.
4. Avg. Opp.
5. OTB rating.
I dunno, I'm pretty math oriented. :)
I've just recently dropped to the low 1600's with online correspondence chess. However, I am a drinker at night, so when there is no games and my urge to play chess, (especially when drinking) is dispayed with live chess. I would say that I am about a 1350 live long game player if I dont drink, which makes sense since I like to subtract 200 points from online to figure what I would be as an OTB player.
I think chess.com's online chess is about 200 pts. below your OTB rating. Some people gain a lot of points from timeouts so that is factored in too.
*online chess 200 pts higher than OTB rating
WellRounded> There must be some formula...
For which rating do you mean?
For correspondence, there was a survey of 60+ ratings done a year or two ago and the correlation between them is very weak. There are USCF 1200s with 2200 chess.com ratings and USCF 2100s with 1800 chess.com ratings. Bill Wall has the data.
For blitz >1400, (OTB = Blitz) is the linear equation, with stddev=119. A quadratic equation would only do slightly better. For blitz <1400, there is no correlation.
WellRounded> 2. Attention paid
That's why blitz is a better predictor than correspondence--generally, both players spend the same amount of time looking at the game. Factors such as drinking, taking the game seriously, and experimenting are hard to factor in. I suspect most regular USCF players take online games far less seriously than rated tournament games.
WellRounded> 3. Number of Games
I considered only players with 8+ games.
WellRounded> 5.OTB rating
I considered only players who had played within the past 2 years.
WellRounded> 4.Avg Opp
Usually this doesn't matter, because Glicko/Elo systems already take this into account. The exception are players who attempt to inflate their ratings by playing weak opponents who cannot win. No such players are in the pool so far.
AnthonyCG> I actually haven't played an official USCF game yet.
Ahh, thanks. I wonder why "USCF: 1200" was filled in when you signed up.
If you decide to give OTB a try--and it is alot of fun--there are a few chess clubs listed in your area. And most clubs have skittles, meaning you can play without it being an official rated game and see how your like it and how well you do:
WASHINGTON. HOWARD UNIVERSITY CHESS CLUB. Affiliate Type: College. Contact: C/O ALI TORAIN.
WASHINGTON. OLYMPIC CHESS HOUSE WASH DC. Affiliate Type: Regular. BENNING RD P DC LIBRARY. TUES & THURS 5:30-7:30PM , 3935 BENNING RD NE , WASH DC 20019 . Contact: VAUGHN BENNETT. (202) 227 - 0102. Additional Information: FAX (202) 397-4439
WASHINGTON. COSMOS CLUB LIBRARY. Affiliate Type: Regular.
WASHINGTON. METRO CHESS. Affiliate Type: Regular. M-TH 6-9PM,SAT/SUN NOON-6PM. US CHESS CENTER , 1501 M ST NW . Contact: DAVID MEHLER. (202) 857 - 4922.
--- likesforests (Who is done with doggone statistics... tactics, tactics, tactics!)
OTB im around a 1500 consistantly, ive beaten many 1650's and a 1777 OTB, my live chess here runs 1550-1650 so its a little higher than my OTB but not much...CC is new to me but my online is running 1600+ right now but ive played top board on a few matches and thus played a 630, a 810 and a few 1000's so its scewed up too...id say you could take you Chess.com rating and subtract between 100 and 200 points to get a rough estimate of you OTB skills...
Its amazing how curious people are concerning how online ratings compare to OTB ratings and yet the same people dont seem to have much desire to simply play some rated otb chess and obtain an otb rating. I can only wonder why ? If you play 10 games of otb rated chess you will get an otb rating. This would require playing 2 maybe 3 tournaments. Your otb rating will be more reliable if you play many more games than 10, ofcourse. It has been mentioned that probably most otb players take their otb rated games far more seriously than any online chess and for me this is certainly the case. It is also the case for all the otb players I know.
some people live in areas devoid of chess culture...look to see how many tournies there are in Wv each year, we average like two, the state championships and maybe one other organized by a frustrated guy who took his TD test and made a tourny up for him and his friends. i would have to go to penn, virginia, or ohio for a tournament...after you factor gas, food, hotel, entry fee, uscf membership, smokes, and anyother expense that might pop up you better be winning the whole kitten cabudal to even come close just reimbursing your costs...i havent played a otb tournemant game since i was a high school senior in 1997...so Reb I would say the answer to your quandry is a mixture of finances and availability, not drive, ignorance, or desire.
My case: there is a club in my town, however without a car it takes me more than 2 hours to get to it. They meet from 7 to 11 pm, on Tuesday and Thursday. Well, I don't know of others, but after a day of work I really don't like to arrive home at about 1 in the night. And get up early to go to work the next day.
As for the question of the person who started this topic, you cannot compare the ratings from blitz, online and quick with OTB. There might be some correlation between your rating for long games and your OTB strength. However, there is a huge difference between the ratings here and those on ICC for instance (400-500 points in my case). So you cannot take them as a measure :)
Narz: USCF 1700, Blitz 1600, Long 1550.
You can't compare blitz & "long" (usually 15 minutes for me) online matches (with people IMing you on gmail & your roommates talking in the background & half-watching youtubes while your opponent thinks) to real OTB games. Generally OTB games bring out the best in me.
Coorespondence on the other hand you can wait until you at least have some degree of focus before moving. Ideally your level of coorespondence play should be much higher than your OTB rating but alot of players (myself included) occasionally blitz thru & end up playing an even worse game than they would in a 2 hour OTB game.
Its amazing how curious people are concerning how online ratings compare to OTB ratings and yet the same people dont seem to have much desire to simply play some rated otb chess and obtain an otb rating.
There is nothing quite real tournaments face to face with your opponent. I miss the cheap tournaments I used to attened weekly in California (well usually the same tourney 4-weeks in a row, G/110). While I like playing online it's just not the same level of intensity! Anyone who's lucky enough to have a weekly OTB tourney group (even an unoffical one) in their area should jump on the opportunity.
It is the case for that I just started playing a couple months ago, (before which, I only even knew how the pieces moved. :) ). So, not knowing ANYONE in the chess comunity in North Georgia seems like it makes it a bit more of a challenge. As well as I work amoung other responsibilities that I (will eventually) have to fit OTB chess in between.
Um... Title.
This question I am certain has been asked, but what are the typical translations of OTB ratings vs Online, and "Live" (Both Blitz and Long if it matters) I'd really like to see some examples if anyone plays all three on a regular basis.
EDIT: Also, what would everyone consider decent Sample size of games before each can be judged?(all three)