SAT I to Elo correlation equation / Levitt equation

Hello.
Using the IQ to SATI conversion chart found at: http://www.iqcomparisonsite.com/SATIQ.aspx

and the Levitt equation: Elo~ IQ*10+1000
One can calculate SAT I to ~Elo conversions.

Please keep in mind these are SATI's taken after May of 1995 to whenever as long as its on the 1600 scale/roof (Verbal + Math). That is do not add in your writing score if you took the SAT when it had a scale/roof of 2400!
IF you took the SAT test before May of 1995 good for you; your estimated Elo will be HIGHER than the one predicted by the formula!

Namely (using IQ 15 SD column the leftmost column...as most IQ tests use a standard deviation of 15) picking a pair of points and finding the line IQ= .0724*SATI + 36.33 
Elo ~ ((.0724*SATI+36.33)*10) +1000  = .724*SATI + 1363.3
Elo ~ .724*SATI +1363.3

The short and approximate version of the formula is take your SATI divide by 1.4 and add 1400.
Cheers,
JWILD

No offense, but this whole IQ to SAT1 conversion and chess rating conversion is total garbage. Linear regression is probably a terrible way of doing this. 

Agreed. Two noisy correlations which are both probably better regressed using at least a quadratic (am I really to believe that someone with an IQ score of 25 would score negative on the SAT (scores below 400 are not possible) but play chess at a 1250 Elo? No. That said, like (literally) all linear approximations, there is a range over which the approximation is a good fit (if else, restrict range further). In this case, I would guess that range covers from at least 1000 to 1400 SAT, 108 to 138 IQ, 2080 to 2380 Elo... a range covering about 27% of the population.
Also, this conversion suffers from problems that IQ to SAT I conversion and IQ to chess ratings suffer. In addition there's the possibility that even IF IQ and SAT I correlate and IQ and Chess Elo correlate, SAT I and Chess Elo could, in fact have zero or even negative correlation. For instance if IQ were composed of orthogonal elements (two elements of which could be SAT skill orthogonal to Chess skill) or negatively correlated elements (picture Elo and SAT negatively correlated with each other in some restricted amount (that is to say this amount has a limit whose value is determined by the magnitude of the positive correlations. For instance two elements, perfectly negatively correlated with each other, cannot also both be positively correlated with a common third element.)). So there are limitations on this idea which shrink as the correlation between Elo and Chess skill and/or the correlation between SATI and IQ grows.
SATI to IQ ~ R = .7 (http://www.psychologicalscience.org/pdf/ps/Frey.pdf?origin=publication_detail)
IQ ~ Elo R = ?      --> Small data set

The only thing that needed to be said.  We could have locked the forum at that.

I've done two IQ tests in my life. One I scored 132 and the other 129.  Those tests showed that I have decent problem solving skills, but they didn't show how bad my short term memory is, how little determination I have when it comes to completing tasks, or how easily my mind wanders while playing chess lol. I'll never master this game, no matter what my IQ says. I just don't have the stuff. I enjoy trying though 

If you meant "needed to be said" as in "should have been said" you violate your own warning by saying... anything. If you meant "needed to be said" as in "required by physical law" you are again in defiance, now to your supposed physical law. If you meant "needed to be said" as in "required by manmade law" then I remind you of the flag next to your name and mine, representing the United States of America, a nation which, counts as primary among its laws, freedom of speech, a notion opposite to your assertion under this supposition. You are then either self-contradictory, misconceived of the nature of physical law, or un-American.

I care not which it is or whether you believe me.

So do I go straight to FIDE with my SAT results and get a title? How much is the fee?

@SilentKnighte5 You do not go straight to FIDE, you go through me first. There is no fee on your end; instead, I owe you 1,000 Schrute Bucks. You may choose to convert to Stanley Nickels if you so desire, given you provide the correct conversion table.

I got 2240 on SAT (averages to 1493 on 1600 scale). It has nothing whatsoever to do with chess ability. So I don't think there's much to this idea

I mean, there's the gaping hole that tons of high SAT scorers have no clue how to play chess... so they would be 0's. 

@dpnorman  Your predicted Elo roof is 2444 = (1493*.724 + 1363.3). Your chess.com blitz is 2051 and USCF is 1932. So the formula predicts a max that is indeed above all your ratings, and yet off by 522 USCF and 393 Chess.com . I realize that 522 is a number and have a good sense of how much 522 is, so perhaps we can avoid arguments over how much or what 522 is.

1493 SAT would be about 99.85th %tile  http://www.iqcomparisonsite.com/SATIQ.aspx   and
2051 Elo is 99.4th%tile        https://www.chess.com/stats/live/dpnorman?type=blitz
So please please please, let's start arguing over  why the formula failed to predict  the last 0.45%tile points (because 99.85 - 99.4 still equals .045).
Let us too note that at the time of this writing your various chess ratings provided on chess.com, (excluding variants) range from a low of 1654 up to a high of 2141. The difference between you tactics rating and chess daily rating is therefore 487. If we include chess variants the difference grows. Now please enter argument of how chess skill is not related to chess skill or the space on the number line between 487 and 522 is where the magic happens.

^You do know my low rating in CC on here is because I timed out in about ten games in a row

And SAT scores can't...

I guarantee you that I don't have the potential to become a GM, or even an IM (nor do I want to do that). The "roof" is clearly flawed there. 

I know a person who scored perfect 1600, super high IQ, and is a total fail at chess. 

This whole system is just some speculative stuff to make people feel good lol

My GM coach told me that he thought if I worked very hard I had the potential to be IM. I think it's entirely possible that was just him patting me on the back and giving me false confidence though. Besides at 18 years old and not even 2000 yet that seems like a major stretch. But I don't doubt that FM or at least FIDE CM and USCF NM could be possible

Well I guess it's *possible* for me to get some FIDE master, but I'm happy enough to just get the NM title already and then just play for fun lol. 

Also, your USCF improvement is currently at the same speed as some NMs on this site like TheLancer and VanishingDragon, who both started younger, and would probably improve less quickly if they started later. NM definitely possible. 

Troll. Identical twins often have different Elo's too. In fact they often even have different pairs of shoes! 

Looks like a waste of time