ACG rated #1 in grading consistency

CHESSNUTDOG

Found this in one of their sales....interesting."....ACCUGRADE WAS RATED #1 AS THE MOST CONSISTENT GRADING SERVICE BASED ON THE NOVEMBER 17, 2003 ISSUE OF COIN WORLD REGARDING THEIR RECENT SURVEY. SEE PAGES 62 THRU 66. PCGS WAS RATED LAST WITH THE MOST SCATTERED GRADING........Thats a quote....Mike

jdimmick

"Consistently inaccurate"

clw54

Does anybody ever resubmit coins to ACG, trying to get a higher number or something? That can be a source of inconsistency.

dollardude

This is a very good discussion. I think we all beat this issue to death yesterday. I believe the consensus was that this was not a valid experiment because of the statistically small sample size.

mintluster

Yeah right. Too bad nobody in the collecting world heard about that, or did but don't care.

razorface1027

"Consistently inaccurate" ...YEPPER! Like Marty's sig line states, "Friends don't let friends by ACG." That, in itself, says a helluva lot about ACG...

BillJones

<< <i>ACCUGRADE WAS RATED #1 AS THE MOST CONSISTENT GRADING SERVICE >>



Yea, I'd agree with that statement ... They are wrong at least 98% of the time, and I can't think of any other service, which is RIGHT or WRONG with THAT much consistency.

1jester

Consistently wrong. That's what I'd say too.

Bobk

Lies, damn lies, and statistics.

wingedliberty

As I predicted. I knew they would do this. This is comical. There is no limit to their dementia.
They actually think that people buy this nonsense.

roadrunner

Bill Jones,

You are totally wrong! ACG is not 98% accurate! They are 100% accurate in grading incorrectly! Of course if Coin World tossed in the 1-5 point variations in ACG grades below the real ones, they'd be inaccurate too! I guess I didn't get my 70 cents worth out of that issue.

Bet for just getting it wrong. They can't be beat. Well Done Coin
World. I just didn't know!

roadrunner

BillJones

No, I had evidence that I am right, but I destroyed it. A few weeks ago I purchased a Hard Times token in an ACG holder that they graded EF-40. Since this variety comes weakly struck, it was in fact an EF-40 although it would have looked like VF-30 to collectors who don’t know the series. ACG was right for ONCE.

However since I don’t allow ACG holders to “grace” my case, I cracked it out for resale.

dpoole

Facts are facts . They were consistent in the study.

But here are other facts:

1. the study only involved eleven coins, which is insufficient to draw valid. conclusions. Coin World and the author made this point clear.

2. the study only involved coins anonymously submitted. ACG may be able to grade when they decide to grade. This does not mean that they wouldn't grade a whizzed coin if it were submitted by a buddy or somebody paying for that specific "service."

I'm not surprised, either, that ACG jumped on these data for their ads. Typical.

Besides running a study with a large enough sample, a REAL test in ACG's case would be to crack out a large random sample of the stuff they've already graded, and see if you get the same grades/opinions back if you submitted those same coins anonymously and without direction//requests for grades. I'll bet you'd see some statistical inconsistency then!

RotatedRainbows

ACG rated #1 in grading accuracy

I do not remember anyone saying anything about ACG being the most accurate service just they were the most consistent using their own grading standards which unfortunately are nowhere near the industry accepted standards. It is not very hard to over grade everything the hard part is to accurately grade using the industries accepted standards which ACG does not use.

dpoole

RR,

For those particular 11 coins, ACG's opinions were consistently close to the consensus (average) grades derived from all the services. For those 11 items, then, they in effect did use industrial accepted standard, whether by skill or luck.

Barry

Here's one for the conspiracy theorists...

ACG said to Coin World: "We're really hurting since the Stuppler hearing and that survey of yours. If you want us to keep advertising with you, you'll have to do something to help us out."

So, Coin World came up with some ludicrous statistical analysis showing how accurate ACG is.

MrEureka

ACG may be able to grade when they decide to grade. This does not mean that they wouldn't grade a whizzed coin if it were submitted by a buddy or somebody paying for that specific "service."

Dpoole - If that were the case, it would explain a lot of things.

fcloud

I read the entire article with interest. And from my understanding they graded each service against each other. There was no set (agreed upon) grade before the coins were sent off for grading. I find this the most inacurrate part of the test. They determine the standard from the results? I find it ironic that the most accurate grading service turns out with the worst grade. All this article proves is that there are statistics and there are damn statistics. The market seems to have made its decision on the accuracy of grades and I will leave it there. The market has found the top service(s).

ms71

What is clear is that publication of the "study" represents a real disservice to the collecting community. Obviously, it won't influence anyone who has any knowledge of the marketplace, but it could well have unfortunate consequences for the unknowing novice collector who makes it the basis for any buying decision. An unfortunate situation at best.

BigE

ACG was allegedly the top for grading "consistency" NOT "accuracy". Big difference---------------BigE

sinin1

It would be interesting for someone to do a correlation between results and advertising dollars spent in that magazine.

Trime

Accuracy is a the ability to measure against a defined and true standard and get the correct result.
Precision is the ability to be consistent in the results you get.
Thus a testing system can be extremely precise (the same exact answer every time) and yet be very inaccurate.
Unfortunately there are no absolute ( ie gravimetrically determinable) standards available in coin grading although there have been aborted attempts to make the standards more objective using various not human assessment techniques. The addition of eye appeal and market grading has added confusion ( although popular support) to the already non scientific approach to grading. It is thus difficult to assume accuracy in a non scientific process. That said, what we have is better than what we had before grading services.

fcloud

Perception is reality especially for the novice.

Look at it as if a complete novice picks up the paper as just starting in collecting. They will believe that the best grading service is AGC, and that one of the worst is PCGS. Now, in time they will probably get a different view, but only after spending their hard earned cash for an expesive education.

A quick look at Heritage's auction archive shows PCGS coins consistantly bringing higher dollars than coins in any other holders. But that message is simply not in the article!

islemangu

typical coin mag sell out disinformation bunk
I stopped reading them years ago when all that cleaned raw choicebu/gembu crap was continously allowed to totally rip off their readers. From the sound of this thread it sounds like they havn't changed much. Sorry for the generalization of all coin mags into this as I'm an isolated islander. I do remember one nice one out of the bunch though...Coin Connoisseur

Shamika

Accuracy and consistency are not the same thing!

My free throws are "consistently" 10 inches to far to the left. As a result my "accuracy" is 0%.

pontiacinf

????????? for BillJones

is that the only evidence you've seen? I trust your judgement but you cant base a statement like that on one coin....

dollardude

IMO, ACG consistently overgrades coins on purpose. Some unethical dealers purposely carry ACG coins knowing that they are overgraded. Then, they sell these to uneducated coin collectors who think they are getting professionally graded coins. Until this cycle is broken, ACG will stay in business.

BillJones

<< <i>????????? for BillJones >>



Please note, someone said that ACG was wrong 100% of the time. I was simply pointing out that I had find one instance when they got it right. If they got it right once, that means that they are not wrong 100% of the time. They might get it wrong 99.999% of the time, but you now what? Evey dog has its day.

I also once owned an ACG silver three cent piece that they graded AU-58. It was a nice original AU-50. Could that be like horse shoes or hand granades? They get credit for getting close?