Test the TPG's Project

Cladiator

This would be fun and very informative if it could be done. The logistics and financing would suck and finding people to loan out the coins would prove less than easy...

Get together a group of U.S. gold coins ranging in grade, size and era and submit them to PCGS. Log down what they come back as, then crack them all out and send them to NGC. Log in their new NGC grades, crack 'em and ship them to ANACS then log in the new ANACS grade. Repeat this five times always submitting the coins raw. Put the resulting information into an Excel Spreadsheet and it could make a very interesting project.

This could be done with any coin(s).

goose3

end result would probably be that you'd never want to submit another coin again as long as you lived because you'll likely get many different grades from the same service for the same coins.

rsdoug81

I'm with Goose. Someone would spend a lot of money and discover what most of us already know. I think you'd probably find that coins are better left in an airtite labeled with your grade gleaned from the ANA guide to grading.

Goose...have you checked that thing for flu?

goose3

Oh it's sick for sure!!!

Boom

<< <i>you'd never want to submit another coin again as long as you live >>



I'm pretty much there now. Coins put thru their paces, carefully scrutinized under ideal conditions
placed side by side with known 6s and 7s and NONE graded what they deserved. NONE !

I just sold a Peace Medal nickel graded MS 66. The handshake side was pristine but Jefferson's face
looked like he had been in a knife fight and lost...THAT BAD ! If that was a 66 then I need to submit
all the rest sitting around here because they have that POC beat by a country mile. Makes you wonder.

Peaceman

This has been done on the boards.... I think about 1-2 years ago..... not sure how far the search goes back.

Deadhorse

<< <i>

<< <i>you'd never want to submit another coin again as long as you live >>



I'm pretty much there now. Coins put thru their paces, carefully scrutinized under ideal conditions
placed side by side with known 6s and 7s and NONE graded what they deserved. NONE !

I just sold a Peace Medal nickel graded MS 66. The handshake side was pristine but Jefferson's face
looked like he had been in a knife fight and lost...THAT BAD ! If that was a 66 then I need to submit
all the rest sitting around here because they have that POC beat by a country mile. Makes you wonder. >>



I used to submit 60 to 80 coins a year back around 5 years ago. Kept it up till a few egregious rips late last year and earlier this year. I've submitted exactly 3 in the last 4 months and have no intentions of sending in any more in the near future.

I suspect my average will fall to around 2 or 3 a year for the forseeable future. Maybe even less.

I don't wonder anymore. I'll buy them slabbed from one of the top two, but my days of submitting are nearly finished. It's just become a crap shoot anymore and it seems the odds are stacked against you.

keoj

Dxxxxx if you do and daxxxx if you don't.

So you either submit coins to a TPG and work the system protect and validate the value (sometimes in error) or you "go with your gut" but the value is no longer validated?

The current issue that many of us are facing. I have to protect the value of what I collect so I've opted for the former.....live with it from the top two.

keoj

Boom

It should NOT be a crap shoot. These are supposed to be professionals that should be unbiased and grade coins

consistently according to their merits, their properites, irregardless of who the submitter may be. This is supposed to

be the best third party grading service on the planet. Something has sure gone wrong.

flaminio

Not a gold coin, but I'll report my Washington Half Experiment from a couple years ago.

I think if many forum members did just one coin, the cost would not be prohibitive, and we'd have some real information.

relayer

You'd be surprised on how well ACG did when CoinWorld did their test

grip

I believe this has been done.It was in Coin World,2/3 yrs ago.

Al

lordmarcovan

It's been done by a number of folks. Thanks for reposting your Washington half, Bob.

Boom

I think if many forum members did just one coin, the cost would not be prohibitive, and we'd have some real information

I think I just may do that. Shoot, I'm halfway there on a couple coins already.

Conder101

<< <i>I think if many forum members did just one coin, the cost would not be prohibitive, and we'd have some real information. >>


To get good statistical data you might be surprised at how much it would cost. If you really want to get good data you would have to submit the same coin to each service at last five preferably ten times. You would want the group of coins to cover a wide range of types and grades, both mint state and circulated. (one interesting thing from the Coin World study, people often say that grading circulated coins is easy and you really need to concentrate on how well they grade the MS coins. But on the circulated pieces in the study the services were all over the map. If they can't be in agreement on the "easy" circulated grades, how much can you really trust that they can make the one point precision grades on the MS coins?) So I would envision 20 or more coins would be needed. you would want to continously be mixing up the contents of the packages shipped to the graders (you don't want them noticing that the same group of coins keeps coming in. and you want them coming from various people so the packages aren't always coming from the same place.

So you have 20 coins times at last five services times ten submissions. That's 1000 submissions Plus don't forget the postage and insurance fees, both coming and going, for 200 submission packages (five coins per package equals 4 packages per submission times 5 companies times ten submission trips. Five coins per package was chosen again to minimize recognition of repeated submissions.) Plus postage and insurance to sent the coins back and forth between the different submission locations. (This can be reduced by passing the coins between submitters at shows.)

A minimal test such as that outlined here could be undertaken for around $18,000. Of course the more coins used and the more repeat submissions the more accurate the results will be. Each additional coin added to the test adds another $750 to the cost plus another $750 for postage and insurance for every five coins that are added to the test. Figure a cost of roughly $4.500 for every five coins used. or $900 per coin in fees. Adding on another grading service adds $3,600 to the minimal test fee. or an extra $180 per coin cost. (5 companies $900 per coin, six companies $1,080 per coins, 7 companies $1,260 per coin etc.) So 25 coins to 7 companies each ten times would cost $31,500.

wayneherndon

It could be just as interesting to keep resubmitting to the same grading service.

WH

burfle23

Would be interesting on counterfeits...

ricko

An expensive exercise that likely would not be worth the resulting information. I have better use for my disposable income. Cheers, RickO

ZoidMeister

Just as intriguing as it was 17 years ago.

Z

Aercus

A gage R&R would be a good study to do here. Sadly, it requires a fair amount repetition, depending on how much precision your want. I believe it would tell us what we already know, which is most coins are graded accurately with good precision, along with the occasional outlier. This can already be seen as most well photographed "guess the grade" coins are nailed pretty well by the forum consensus.