chess.com ratings are deflated against USCF

Sort:
corrijean

Nice syllogism, rooperi. Laughing

mrguy888

I am a heron and my head is blue.

corrijean
mrguy888 wrote:

I am a heron and my head is blue.

You are an idiot and your example is clearly made up

Wink

rooperi

Actually, he's Marge Simpson

Expertise87

Well, we can't easily poll a random sample of chess.com players who are active in both blitz and USCF standard, but we could just make a thread where active blitzers/USCF tournament players post their ratings for both and someone can do the math to determine the relationship. If we got a good 100-200 samples we could probably get a decent linear regression going.

I suspect most people have blitz and USCF ratings that differ by no more than 300 points, but that this 300 points can be in any direction. My suspicion is based on the number of people I have observed in this thread and others who have USCF ratings and blitz ratings. Generally I have observed higher blitz ratings on here than USCF ratings, although the OP suggests the opposite is true by a wide margin.

While the OP has an interesting argument which he supports better than some poeple in this thread suggest, and his argument does have some historical basis in rating system comparisons (check out the formulas for converting ECF to FIDE to USCF - especially as most USCF events use FIDE+100 as the converted USCF rating for foreign players, this is an example very similar to that cited by the OP but obviously only valid at higher levels as FIDE didn't even until recently have players rated below 1800), I do not think his conclusion is accurate.

And I don't think taking the top fraction of a percent of USCF players who have chess.com memberships and comparing their blitz and USCF ratings is a valid comparison either. More benefit would be derived from taking our sample in the upper 50% of USCF members, who are more likely to know how the pieces move and do so quickly enough to not lose every blitz game trying to figure out where they can move ;)

Also, 'blitz' encompasses everything from etime=3 to etime=14, and USCF standard encompasses everything with etime>=30. So there are huge ranges for both in terms of time control, so we can just make a bunch of blanket assumptions if we are going to compare these two (relatively unrelated) rating categories as well.

So I propose that we make a thread where active USCF members post their USCF ratings and chess.com blitz ratings, compile a list, run a linear regression, and check our values. It would be much more interesting than the OP's claim that he has a sample and that it's easy to check rather than actually providing us with data and repeating himself every time somebody posts something he disagrees with, and claiming that he has convinced himself but still feels a need to post in this thread at all for some reason.

AdamRinkleff
FirebrandX wrote:

Now granted I never hit my peak in OTB rated play, but I seriously doubt I'm any stronger than my blitz rating already suggests.

One word: "average". I'm talking about the average person, and the average difference. For most people, the drop between USCF(standard) and chess.com(blitz) is 2-300. Obviously, there are exceptions. Some people are especially good at blitz, but that is not normal.

In your case, its been two years since you played in a USCF tournament, whereas I was only talking about people who are currently actively engaged at both USCF and chess.com. My guess, is if you come back to USCF, you will see your rating increase.

rooperi
AdamRinkleff wrote:
rooperi wrote:

My blitz here is 1513.

There is no way my USCF (f i had one) would be 1763.

Yah, I agree, if you would read what I have written a little more closely, you would know that I am talking specifically about people who actively play both blitz and standard at both chess.com and in the USCF. You are not such an individual and I was not talking about your rating.

But now, you tell Firebrand that his USCF will increase if he returns?  And mine wont be close to 1760. So, does that mean if I should play USCF my blitz here will drop?

AdamRinkleff
rooperi wrote:

[idiocy]

Quit trolling.

rooperi
AdamRinkleff wrote:
rooperi wrote:

[idiocy]

Quit trolling.

Thanks for clearing that up. You cant answer a question, and you resort to insults. Classy.

cferrel
AdamRinkleff wrote:
furtiveking wrote:

Sure you can extrapolate, but 10 examples isn't NEARLY enough in this case.

Oh, honey, I know I said that I have more than 10. I think I have plenty. If you want to calculate the average ratings of a 1000 people, I don't think your ultimate average will vary considerably. I think its pretty clear that chess.com blitz ratings are about 2-300 below uscf standard ratings. Every active uscf player I know who uses chess.com has agreed with me that this is the case.

Blitz and standard chess raitings do not belong in the same set. I saw you were trying to use set theory to validate your arguments however, use of definitions too loosly to make a valid argument. Thus, your argument is false since the conditions of the game are not the same based on the pool you gathered. 

I will finish my math degree in Applied Mathematics in 3 semesters so dont try to pull bs on me.

waffllemaster

So show of hands, you guys really believe he's sincere?  Surely this is a troll thread.

rooperi
_yiquan_ wrote:
ozzie_c_cobblepot wrote:

I've always thought that the underlying motivation behind people asking about how USCF or FIDE correlate with chess.com is one of:

people love patting themselves on the back lazy people who want to avoid working seriously on their game, looking for an excuse to call internet blitz/bullet "work"

On a related note, I think that there's a nice correlation between Tactics Trainer and ceiling potential.

I too think the work to fun ratio is strongly in favor of "fun" when it comes to internet blitz and people tend to tremendously exaggerate it's significance.

I am also wondering about the specifics in regards to: "a nice correlation between tactics trainer and ceiling potential". did you actually conduct a full fledged statistical analysis of some sort? or just hypothesizing based on what you've noticed around the site?

I think that really changes from person to person.

My  TT is 500 over my online rating here. I think for me it's because I KNOW there is a solution in TT, while in online I only hope there might be....

I think most other players have TT lower than online. (From what I remember of previous forum posts)

Tmb86

When I was a member I could get my TT rating flirting with 2000 most of the time... which is 400 above my online rating. Of course, there's absolutely no reason to think the two ratings should be around the same, playing a game of chess and practising TT are two completely different entities.

I think TT could very well give an indication of a 'ceiling rating' though. I never saw someone with a TT rating +2100 who had a lower online rating than me. 

zborg

If @Adam wasn't from North Korea, the "rat pack" might not have taken him to task with such verve.  Whatever.

But if anyone knew just a tiny bit of Myers-Briggs, you'd see this 300+ thread for what it is -- Intuitives fighthing with Sensors over the "meaning and implications" of Data.

And that argument ALWAYS degenerated into a Big Food Fight.

Knock yourselves out guys.   Maybe start another GG thread?

At least those posts are funny.  Smile

AdamRinkleff
zborg wrote:

But if anyone knew just a tiny bit of Myers-Briggs, you'd see this 300+ thread for what it is -- Intuitives fighthing with Sensors over the "meaning and implications" of Data.

INTP here. I can't stand SJs. Despise them. Meanwhile, I've come across two additional people, with verifiable active USCF and Chess.com ratings. In both cases, their Chess.com rating is about 250 points lower.

Abhishek2

I think a scatterplot with a best fit line like someone else before did, an average may not be the most effective.

AdamRinkleff
Abhishek2 wrote:

I think a scatterplot with a best fit line like someone else before did, an average may not be the most effective.

Of course. We could devise all kinds of complex equations in order to gain the most effective understanding of the subject. Is it worth it? An average is good enough. We don't need a better understanding. I certainly don't care. 2-300 points is close enough for me.

ozzie_c_cobblepot

Seems like the guy without the Maths degree is satisfied with the knowledge he has put forth, unwilling to engage in a constructive dialogue with those which do have the aforementioned. Further, as I hypothesized, the whole point of the thread is to validate a lack of work as substituting for the opposite. I cannot claim surprise at the correlation.

How many pyramid tiers did I paint with the above brush?

Abhishek2

why not play a longer time control? In fact, I only play in super long OTB games.

zborg
ozzie_c_cobblepot wrote:
 

Sorry Ozzie, but this pyramid leaves a lot to be desired--

The Rhetorical Tetrad (4 parts) includes, Facts, Logic, Metaphors/Models, and Stories.

But we still must take note of Speaker, Audience, historical-political context, and the medium of transmission, inter alia.

An Internet Food Fight (as witnessed above), with participants hiding their identities, AND baldly asserting the pyramid (above) is ONLY focused on the LOGIC element of 4-part Tetrad.

Moreover, participants in this discussion come in (at least) 4 flavors, (SP, SJ, NT, and NF) according to Myers-Briggs.  That adds to the complications and fosters more food fights.

So you are leaving lots out, in all its messy glory.  Smile

Bertrand Russell is NOT the last word.  Sorry.

And who said Chess players were not an ECCENTRIC lot?  Nuff said?