Grading is in the hands of the beholder

Hi all,

Just read an article on Grading coins and was interested in a test performed where 7 notable experts were given a coin to value and grade. All gave a different grade and value. Just shows how difficult this process can be.

Although I believe it is good advise to get a professional assessment, at the end of the day, the beauty and grading of a coin lies in the hand of the beholder. Would you agree?

 

Log in or Sign up to hide this ad.

I'd like to know what the coin was and who the experts were. Such results would be incredibly rare when experts are involved. In a large majority of cases they would be within a point of each other, if not at the same grade.

 

I agree that sounds a bit questionable.

 

Yes, it would be of assistance to know what was being graded. Why not post the source of your article, a link, the name of the publication, etc. ?

 

I immediately thought the same thing. There's discrepancies but they aren't that bad. I agree with Mark. I think even I could grade within a one point difference of a panel of experts for most issues.

 

Im waiting for the day when a computer program can take the guess work out of it.
A highest grade stock picture (MS) is stored in the software, and the coin is scanned.
The software makes a comparison betwwen the scan and the MS stored file, and decides on the grade. Looks for nicks, dings, and wear on the high points. Other files within the application store the current trend value of the grades.
Amount of reflection of light during the scan determines lustre.
If it can be done, then people that feel that heavily toned coins deserve a higher premium are SOL.

 

I would like to see the coin that the 7 experts graded.

 

I agree that even with 7 notable experts there will be a difference in opinions however would think it would be in the upper ranges above MS65 this is where it gets sticky and when it starts to really affect overall value.

 

believe it. I have taken a TPG slab coin to Three dealers once. Guess what. It was a seated liberty dime in a SEGS holder. Now that makes 4 different grades and I can say , they were grades apart. I wound up selling the coin at Heritage, for $1,200.00

I paid $800.00 ......Go figure.

Larry Briggs is a personal friend and I purchased the coin from him .

 

Here is the link to the article:
http://www.thecoincollectingguide.com/grading-coins.html

Sorry, got it slightly wrong, coin sent to major grading services who graded a 7 point difference. Even still, not accurate. Like Dollar1948 idea of computerised grading although it would be still difficult with so many variables to take into account.

 

Unfortunately grading is not entirely objective.
For instance, eye-appeal can be a factor in grading, and the less experience one has, the more eye-appeal affects the grade.
In my case I'd possibly grade a rainbow toned coin a point lower than others because I don't find it appealing.

So personal preferences enter into the mix: subjectivity.
And that leaves computer grading in a weak position.
It is a rule-based system and can't reliably cope with subjectivity.

 

Your absolutely correct GD. Let me re-phrase that to , taken to dealers with whom I had the opinion of, they were far more expert than I . Once again, I was wrong. Suffice it to say, Larry Briggs was correct with his grading,and an expert in that issue of coins. Someone acknowledged that at auction when they coughed up the $1,200.00 I would assume.

P.S. Larry even offered to buy the coin back from me with an additional 10% on top of the price I paid him.

 

From what I'm gathering from these threads, there should be a one universal grading standard. Same issue arises with stamps.

 

Im waiting for the day when a computer program can take the guess work out of it.


Eye appeal enters in to it as said earlier, although a computer could get close I would think.

 

I realize that over the years, many people have advocated and/or predicted that we will have computer grading. But I don't even think it would be possible to do so without having many of the flaws and inconsistencies that we do now. Among the biggest two problems are 1) imperfect humans would still have to program it and 2) generally speaking, no two coins are identical.

So, for example, how do you program a computer to grade coin A a certain grade due to marks, but coin B the same grade, due to more, but less severe marks? Or fewer, but worse marks? Or more marks, but in a less conspicuous location? And so on?

 

Supposing a TPG scanned its coins into a computer database, much like fingerprints are scanned into computer databases, but with grading checkpoints, instead. I don't have all the details worked out, yet, D), but one such series of checkpoints might be on the high areas, in order to detect wear. The database, of course, would be the "market," much akin to a multiple listing service for real estate. Submitted coins would be rated on the checkpoints, and ultimately assigned a "market grade." I wonder...

 

I suspect someone might be able to get a computer to do the grunt work.

It could be as simple as having the computer note all the "irregularities" on a coin (marks, rubs, dents, dings, flat high points, scratches, etc.)
The computer would not make grading evaluation; it would just bring all this to the attention of the human grader(s).

Or it could be as complex and sophisticated enough to do technical grading.
The human grader(s) would then decide how well the computer did AND add in the the more subjective aspects of grading.

I can see the computer system starting out with the simple task above and grow over time into the more complex.
The technology exists.
Step 1 would be to get all the experts together to list all the general elements to be evaluated, then split them between the technical and the subjective.
Step 2 would be to create teams dedicated to specific issues (IHCs, Lincolns, SL Dimes, etc.) and refine the general elements into something "unique" to the specific issue.
Step 3 ... well, let's see if they can get 1 and 2 done. That should take several years.

 

I'd be interested in seeing uniform grading standards, at least on the technical side.
Reduce the guesswork to the subjective portion of the evaluation.
I don't believe we'll ever get everyone to arrive at the same grade for a coin.
But I do believe we can narrow the range of grades arrived at.
I can live with some professionals arriving at MS-64 for a coin, others at MS-65 for the same coin.
I can live with some professionals arriving at AU-53 for a coin, others at AU-55 for the same coin.
What causes me real grief are grades ranging from MS-63 to MS-66 for the same coin. Or AU-50 to AU-58.
Or worse.