Login

I. R. Bama's Blog

28 Feb 2021

The 100 Point Coin Grading Scale: Its the Mathmatical Underpinnings that Make It Great!

| I. R. Bama

Many of you read with interest and have encouraged me to pursue a CGS system that is mathematically sound. So what makes a model or grading system useful? Well, it has to have internal validity. What is that, you may ask? Its a fancy way of saying that it accurately measures what it purports to measure in a manner stable over time and multiple observations. So you get the same results over time, every time. the other construct to be considered is observational reliability. That means that you could in this case, show a coin to any grader and you will have almost perfect agreement in giving the coin a grade.


Multiple criticisms of Sheldon's scale exist, I am not the lone voice out there: Sheldon developed his system as a teenager to grade large cents. I will not repeat the criticisms in writing but instead refer you to an example of some of what has been written finding major flaws in Sheldon's grading system. "The Sheldon 70 point grading scale | Coin Talk" www,cointalk.com/threads/the- sheldon-70-point -grading scale.232527

In order to make this model a powerful tool, it has to be be statistically designed to ensure an equal value interval between points on the scale and fall under what is known as a normal distribution curve or a bell curve. A bell curve will predict how often a particular coin will appear relative to its position under the bell curve. 68 % of coins graded with fall with in two standard deviations from the average.


In a scientifically based grading system each point on the scale must reflect an increment of one. Each descriptor of a coin must be even in its position from another descriptor.


91-100 Mint State

81-90 Almost Uncirculated

71-80 Extra Fine

61-70 Very Fine

51-60 Fine

41-50 Nearly Fine

31-40 Extra Good

21-30 Very Good

11-20 Good

0-10 Nearly Good


For those of you who are experienced graders, I would invite you to try this system and grade some of your coins with it. Once one becomes used to it I believe it will allow for more accurate grading than Sheldon could ever hope for. Feedback of your results is most welcome! Enjoy this picture of Dr. Sheldon....

Comments

CoinHunter

Level 5

Sounds like a grading scale that should be accepted, it makes a lot more sense for mint state coins to be 91-100 instead of 60-70

Kepi

Level 6

I enjoyed this blog! Especially reading all the thoughts and ideas in the members comments. ; ) Thanks Bama!

I. R. Bama

Level 5

Hey Gary, I think using probability makes it easier to grade coins. We all use probability in decision making whether consciously or it's running in the back of my mind. You just look at a coin and by sight are running the probability whether this coin is average, or above or below that average. Then we get more precise from there. Again as with how we grade coins with Sheldon, we stick with 5 point grade assignments and reserve 91 to 100 for the MS coins.

I. R. Bama

Level 5

Mike you expressed concerns about intervals. You rightly identified my proposed model has what you call groups. You need to recognize that the Sheldon scale also has intervals or groups. The advantage of my scale that each interval that is assigned a written descriptor is equally 10 points apart. The failure of Sheldon lies in the logarithmic variety in relationship to several point intervals in his scale. In Sheldon's world, much on like Orwell's animal farm, "All points are equal, but some points are more equal than others". I can assure you that there is no call to grade circulated differently from T.W.AD.I. so if on a 100 point scale I expect I would see grades like 45 and 60. Mint State coins: I'm not sure you can argue that there would be magnitudes more of outraged TPG consumers because both scales grade point by point for mint state coins. Actually since it's a science based grading system, it should be more accurate grade between observers too.

Mike

Level 7

It's not about pricing but that's what it all comes.down to. You think someone will send a coin back if the higher grade did not mean more money. Sometimes one grade could mean thousands.

I. R. Bama

Level 5

You are contradicting yourself: "it's not about the price but that's what it comes down to". I respect your opinions, but your thinking is circular in this instance. And doesn't have to come down to with whether we can find a better grading system? People will quibble with TPGs. That is entirely independent from what scale you choose to use. People who want to argue over the grade of a coin have a bone to pick with the grader, not Mr. Sheldon nor with my proposed scale

I. R. Bama

Level 5

That has nothing to do with what scale you use. That's about a coin owner who feels that they were treated unfairly by a TPG. I don't know anyone like that, but I'm sure you have. It's an argument that I don't think is valid

I. R. Bama

Level 5

Mike, as far as my scale goes, or Sheldon's for that matter.... Neither scale makes assumptions on how graders assign a coin grade. I imagine graders would grade circulated coins by increments of 5 with 10 points for MS coins.

Mike

Level 7

I think my friend you misunderstood me. Ask yourself why does the regular collectors send there coin in? I send my tokens in to improve the quality of the rare set I have. I have had tokens that were a 63 seen a better looking token . Seen not held. I just knew it was better than what I had. Sure enough I received it back it was a 65. That's a reason to send them in . Your right for the quality of the coin the right grade. However most send a coin in for value only. But when you increase the quality of your set the value also goes up. I'm taking your advice and reading them over. .Thanks that was another good idea!!

Longstrider

Level 6

I believe we are talking about Technical Grading here. Most comments here seem to be referring to Market Grading.

I. R. Bama

Level 5

Right, this is all about technical grading. Grading really should be the driver of prices, not fitting grades to market value.

Mokie

Level 6

I think a 100 point scale as described is brilliant. From the very first coin produced by a brand new set of dies (100)_ to a coin so basal that it can barely be identified (1), the 100 point scale works. The only thing preventing this logical step is the fierce resistance of the entrenched grading companies and the old timers that cling to Sheldon. This can work but it will take a heavyweight in the hobby to start up a new company utilizing the 100 point scale and the backing of someone like a Bowers writing the 100 point grading guide. If that company can be formed, the reslabbing alone will be worth millions. I think this would also be the chance to begin the process of using AI to grade coins, it was tried before but the technology has advanced tremendously since then. I am all in and hope it comes to fruition.

Longstrider

Level 6

I feel the same for Technical Grading. To much human subjectivity now.

TheNumisMaster

Level 5

I resemble all of the above, Mokie!!

Golfer

Level 5

I would accept either grading system. A 100 point system is great. But I wouldn't want to be a professional grader. I was pretty good in my youth grading Lincoln cents, but I did Good,verygood,fine, etc. I didnt assign numbers. How do you determine a 34, 35, or 36 grade for a coin.? With a 100 point system, i would like to see those CAC stickers ! I don't understand the CAC stickers. Buy the coin, not the holder, like in the good old days.

Longstrider

Level 6

Now you went a done it. Sorry to most but I am interested in hearing more. I know this idea has been talked about for years but I wasn't involved. I have a science background in school and understand what Bama is trying to get across. If you think this is all hogwash I invite you to look into how CAC does their thing. Very much the bell curve thing. It's just an idea to be looked into. It's to easy to come up with hypothetical problems for this or the 70 point scale. To all the friends I just lost I say Sorry but it's just my opinion. I could be wrong. Happy Sunday.

Mike

Level 7

Well I wrote my comment and hit the wrong button. I give you credit. But it will never fly . You cannot grade coins in groups. What is the price between a 100 And a coin 95 . I don't see it. One coin one grade. When who ever made us he made us imperfect. Therefore the scale will also be imperfect. Think of a grader going thru a bad time. That will reflect in his work. I'm sorry I looked at this from every angle. Is my coin a high 90's or the middle And pricing would be crazy. They fight about one grade lower. They think there coin is a 70. They will fight you tooth and nail for that one grade. Did you read N.G.C. way of grading. The customer service reps here most of the c.omplaints. Your not the first to try this. There is a history with people and the perfect grade and pricing. . I have never seen grouping before. Why don't we say your coin is between a 100 And a A/U.. It doesn't make sence. What price would a 91 get? As compared to a 86? Tell me. There has to be prices with every grade grouping does not work. Auctions would colapse. I'm sorry I know you put work into this. But Sheldon was imperfect . I'm imperfect and so is 100 point grouping of coins and prices. I looked at ever angle like the others.. As long as people are grading you cannot grade like this. You would also need one hundred coins . So you can match them up. Sorry my friend The system we have is not perfect but it's the best we have without ruining the hobby. There were also many people in Sheldon corner. That's why his system was accepetd. . Sometimes You Tube has the 100 point graders. Didn't take there. You have to show the difference between a 88 And an 87. People want to know why there coin is not a 70. Never mind in a group. I also have researched this. I have read articles pro and con. In would suggest people do there research before commenting. I know you would agree to that.. It makes for a more informed collector. Instead of just saying great to for it. They don't even know the 70 point scale. I think more should research there own information first . That's only logical. They would be better informed to make a decision. That's important it may change there their attitude towards 70 or 100.

Stumpy

Level 5

Interesting idea, unfortunately we are dealing with an entrenched group whose mantra is "we've always done it this way", seems very accurate, but even I would prefer a sit down over a cuppa to go over this in greater detail. Probability is real, I clearly understand the probability of my spousal unit kicking my buttocks if I forget to take out the trash, however, the 100 point grading scale enters into the realm of logic and I'm not sure if any of us are prepared for that! Take care and interesting idea.

coinsbygary

Level 5

While the Sheldon scale has its problems, such as the next higher grade being twice as good as the previous grade, I have questions about this. Primarily, I don't understand it. I don't understand how you can predict what coins will grade by probability. I do like how your circulated grades below VF are more precise. Still, coin grading has always been and will be for the foreseeable future, somewhat subjective to a human opinion. An interesting way of grading could be done by artificial intelligence to be totally impartial. But then there are so many variables to be considered when grading coins that it will always require a human element. That said, something should be done, especially with VF that can grade 20,25,30 and 35. Or AU, that can grade 50,53,55,58. I often can't tell the difference between a 30 and a 35 or a 53 and a 55.

TheNumisMaster

Level 5

FIrst off, I LOVE how nerdy we all are about this!! I mean, who else, besides numismatists, would incorporate hours of time to find a grading system that is mathematically solid?! I love it so much! Secondly, I love this idea. Even without trying it yet, it seems a lot simpler than the Sheldon scale we all use. Catch ya later!

We use cookies to provide users the best experience on our website. If you continue without changing your cookie settings, we'll assume that you agree to receive all cookies on money.org. You may disable cookies at any time using your internet browser configuration. By continuing to use this website, you agree to our privacy policy and terms of use. To learn more about how we use cookies and to review our privacy policy, click here.