Let's TRY to have a **serious discussion** about the philosophical aspects of coin grading.
Insider2
Posts: 14,452 ✭✭✭✭✭
I'd like to discuss the truths, principals, and basic concepts involved in ESTABLISHING a basic grading system from scratch. This can only be a mental exercise because we already have a commercial grading system that was encumbered with problems right from the beginning. So, just as an exercise for interested members, what would we need to make a brand new system: An ideal grading system?
I've done this exercise decades ago. I don't think my ideas are stale. I want to read what CU members think are the major requirements for a grading system IF NONE EXISTED.
Coins have always been graded in some way. Take yourselves way back in time when there was no ANA Grading System and no TPGS. Now, you personally have been asked to devise a universal grading system for american numismatists. What would it need to do? Please don't get hung up on specific grades or the requirements to meet those grades.
Hopefully, those who have nothing to add except for ridicule and jokes will refrain from commenting here. Perhaps you can start a parallel discussion and have at it with the jokes. Thanks.
I'll start. EDIT: Nope, you all can go first, otherwise there is no point in this discussion. I want to know your ideas. 
Comments
Not sure what you are looking for here??
Well, if you go way back to when there was no grading scale, then rather than using math and numbers to grade, you would likely be going by the good, better, best method.
As long as we don't use "Gemmy" I am good with whatever we come up with.
WS
How about no details/body-bag in the new system.
Cleaning, scratches, wheel marks & AT are all factored into the "grade". (whatever it is)
My Saint Set
Unc. & BU
I always like the descriptive terms when I was a kid because I understood them.
For example saying a coin was very good or fine meant something, Even almost uncirculated held a firm concept of the coins's state or grade. Now when numbers are brought in you put another tier into each of those descriptive monikers.
I could see a novice collector wanting to see a descriptive moniker as well as a number that defines it a little better, say calling something a Fine1, Fine 2 or Fine 3 with Fine 3 being a prime example. This might work on collector coins but when you start talking high value pieces or very high grade pieces it can get a little too simple.
I would have to put a lot more thought into this but that would be my starting point.
Is it possible for this new system to be completely objective? Any subjective criteria will be argued, and will not allow others to arrive at the same grade consistently.
If not. The subjective points could/should be separated out into another descriptor.
ACG had this concept going for a while.
CAC
All comments reflect the opinion of the author, even when irrefutably accurate.
Great answers so far, and they hint at what I was going to offer:
EAC-style grading, but without the numbers, and with a well-managed condition census (CC).
Circulated/Uncirculated Details, Below Average/Average/Above Average/Choice Surfaces
Any problem coin is SCUDZY.
So for a proof coin, the best superb gem proof deep cameo is CC#1, best superb gem proof cameo is CC#2, best superb gem proof is CC#3, and working down from there.
The logistical issue would be having each coin seen by two or more experts, to reach the condition census standing.
But, what I do like, in the end, is that the expert reviewed CC#1 coin is now both top-pop & registry #1.
Classic coins are best served, but moderns would have many ties in the CC rankings.
I'd still stick with a number format, but with a much, much smaller range. 1-70 is overkill IMO.
When was the last time you saw a VF28?
01-21 for circulated coins
21-30 for uncirculated/proof coins
Have fun -
Dave
All coins get ranked
Start with #1
My Saint Set
Philosophical discussion The Love of Wisdom comes through the coins themselves.
The grading system has never changed since the advent of the Sheldon Scale 1 to 70, no matter who was grading the coins,
the problem was and still lies with the grading services.
A basic grading system should provide a structure that would allow accurate and consistent grades to be assigned based on precise measured standards. Establishing methods and procedures to build precise measured standards is not trivial.
Not sure if this answers any of your questions.
We should move to pure market grading as that is primarily what the cost index is based on. The farce that technical grades superceeds eye appeal is a dealer mechanism that gets collectors in trouble when the confuse value with liquidity.
11.5$ Southern Dollars, The little “Big Easy” set
No extra points for rarity, toning, popularity, luster or any other factor. Compare apples to apples period.
Grading raw and slabbed collector coins in a truly consistent manner will never happen. Too many different eyes with different tastes looking at the coins. Most of the members have seen this in higher graded coins, hence the breakout game. I don’t know if these inconsistencies are in say, XF and lower graded seated coins, but I am seeing much too much of this in the circulated state colonial area and Massachusetts silver, which for the most part are not the easiest coins to grade accurately especially for a novice but it can be done with a lot of practice. I know that I am being very stubborn but there is no way that I will buy a slabbed XF coin at an XF price when I can clearly tell that the coin is a VF.
Form is emptiness, emptiness is form.
Or for the slow learners: Buddha buy bulk bullion.
I think the problem with the current system is that it boils everything down to one number. And all coins with that number are NOT equal!
So, an ideal grading system would have multiple grades. One for wear/marks...one for luster, or surface quality...one for eye appeal....one for strike (where appropriate). Maybe more.
Problem: You can't make a coherent "price guide" if there are half a dozen columns for "grade".
In reality, no grading system will be perfect. You STILL need to SEE the coin to judge it for yourself. It's really just a starting point for the potential buyer to make their own decision.
(And let's not fool ourselves: Without a buy/sell transaction, there is no NEED for a grading system! Nobody cares if it's a F-12, or AU-50 if it just sits on a shelf. Most grading discussions seem to think there is some magic to grades and grading. There really isn't.....).
If I were creating a grading system, I would start by identifying the attributes that should be judged - wear, marks strike, luster, eye appeal and whatever else would be important. Then grade the coin on a numerical scale on each attribute separately. The total points would be the grade. I’m waffling as to whether each attribute should have the same number of potential points or whether there are some attributes more important. Perhaps the best scoring attribute should be identified by a letter and the score for that attribute followed by the total score. That way, each person could pick by the attribute he considers most important.
I kindda think the way I am seeing ancients graded would be a great start.
1-Give the Details Grade
2-Rate the Strike
3-Rate the originality
4- Possibly (giving the subjectivity) give an opinion on eye appeal
5-Note any problems or damage....probably should be number 2 on the list!
I concur: "It's really just a starting point for the potential buyer to make their own decision."
First, establish three fixed points of reference: 1) identifiable by type, 2) visible abrasion [aka 'AU'], 3) as struck. These are not moveable. 2 & 3 are physical definitions; 1 is opinion but founded on un-readability of legends.
Second, divide degrees of abrasion between 1 and 2 into subsets that distinguish significant parts of design that are visible or not visible. These would be all the "circulated" grades.
Third, do the same between no circulation abrasion and 3 using surface disturbance as defining characteristics, with 3 being free of all non-die/press damage. These would be all the "uncirculated" conditions.
My understanding is that market grading is some undefined combination of technical grading and assigning a value based on several other characteristics. If that is correct then grade variations due to the subjective nature of the undefined characteristics and there weighting are part of the grading system. In addition, these grading variations are not consistently repeatable.
I would not say "that technical grades supersede eye appeal" rather that they can be more easily accurately measured and serve as repeatable precise standards.
...didn’t you strike out bunting, throw the bat and quit the forum just yesterday?
How about the PSA 10 point scale?
As @TommyType mentioned, even under the current grading scale, coins with the same grade are not equal. You can have common date MS65 Morgan Dollars sell for $100, $200, $500 and well over $1,000 or even $10,000 if they are pretty enough... but all of those are the same grade in the current system. Not very helpful to a collector or prospective buyer as the grading system doesn't fully factor in eye appeal.
This leads graders to try to market grade coins by giving the really pretty coins extra credit on the grading scale. Just look at the top pop examples in a common series like classic commemoratives. Nearly all of the very highest graded examples are monster toners that look awesome. Certainly there are untoned examples out there that have the same technical grade as those monsters when looking at luster and marks on the surfaces, but they will rarely be granted those super high grades like those eye appealing toners.
So, what it really comes down to is that every coin is unique and if the graders are going to market grade certain coins in clear defiance of the grading scales, then why not go all the way. Maybe we should abandon the 70 point scale all together and just have grading companies assign a different number to the coin. If the point of a grading scale is to help someone figure out what their coins are worth, why not just have the graders assign the coin a dollar amount the grader thinks the coin is worth at that time.
Well i guess that has problems too as prices change over time... but that would help encourage more resubmissions for higher numbers... great for the graders and those who are shareholders. But then again, when prices go down not many would turn in their coins for lower values to be placed on them.
I guess that's why I still have Hannes/Tulving slabs with crazy dollar values printed on them from the 1980s!
Michael Kittle Rare Coins --- 1908-S Indian Head Cent Grading Set --- No. 1 1909 Mint Set --- Kittlecoins on Facebook --- Long Beach Table 448
"If the point of a grading scale is to help someone figure out what their coins are worth, why not just have the graders assign the coin a dollar amount the grader thinks the coin is worth at that time."
This implies that graders are experts in determining the value of coins. I've never heard of coin graders being tested and evaluated on their ability to accurately determine the value of coins. However I know that they are tested and evaluated on their ability to accurately and consistently grade coins.
The "grade" should indicate tbe Quality relative to other examples of the issue, and the Quality should lead directly to the Value. IMO, a higher "Graded" coin should not be worth less than a lower Graded one. If it is, one of the "Grades" is inaccurate.
Liberty: Parent of Science & Industry
That sounds so simple, but you did not define how you measure the "quality". Your definition and preferences may be very different from others.
Of course, that's the crux of it.
Think about trying to "Grade" a nice old car, say pre-1970
Obviously, one with 6 miles on the odo, all original, just off the lot and drained, lubed, wrapped in velvet, and stored in some collector's climate garage is best.
But which of these has higher Quality:
One with 55,000 miles, original everything but has worn seats, faded paint, and normal repairs, or one with 255,000 miles, mostly restored to pretty good condition with upgraded engine, brakes, suspension, and electronics, and custom seats and paint?
What about all the other permutations? How do you Grade all the variables as a third party, such that the attribute descriptions convey from seller to buyer/seller to next buyer? and so on..
A. Have a multi-variant hypercontinuum in which the composite "Grade" is supported by the market in which a pool of dealers would "buy it as a...[grade xx]
Liberty: Parent of Science & Industry
I would retain the Sheldon grading scale as it primarily deals with wear issues, marks, dings and spots. The Sheldon grade would be the coin's primary grade. Then that grade would be adjusted for a number of factors that relate to attractive toning, questionable toning, cleaning, corrosion, and most of the other details items that body bag coins. An example might be as follows:
Toning & eye appeal...... -2 to +2
Questionable Color........ -5
Cleaning.......................... -5 to -10
Alteration....................... -20
This allows for all problem coins to be net graded instead of body bagged, which is essential if the grading system is to be reflective of market value. Importantly, it will upgrade an attractive coin two points, and it will downgrade an unattractive coin by two points, again moving the grade closer to a market value grade.
Perhaps the grade should be stated as the technical grade with the adjustment. Like the coin is MS65 +2, or the coin is 65 - 5(QC), or the coin is PR66+2. The adjustment to technical grade will be an adjustment to true market value,
OINK
The original idea of the TPG was to stop all of this "grading nonsense" and "level the playing field" it seems to have resulted in arguments over "loosening and tightening" of the services and another third party service to stamp approval on the TPGs
So what did the first party pay the TPG for in the first place? Ans. Subjective opinions that would make the coin acceptable in the marketplace and then subjected to yeah or nay prices. All the rest of the above listed comments amount to blatherskiting is obvious to anyone playing in the coin market.
The final answer will be computer grading. With AI being applied in many areas now, the mountain of work required for programming will become more manageable. Specific standards for marks and wear can be established and will be repeatable (eliminating human opinion). Eye appeal must be left to the market. Beyond technical, physical issues, people are attracted to different appearances (much like art, cars and the opposite sex). This will lead to a slow, but massive change in the hobby.... but one that will benefit everyone in the future. Cheers, RickO
This is an interesting topic. We (where I work) spend a great deal of our time addressing complex problems like this and seeking solutions (albeit not numismatic-related problems). I wish I had a few hours to work through this.
The major challenge I see with grading coins, whether it be using the approach of the late 19th century (four qualitative grades) and the early 21st century system (over 70 quantitative/qualitative grades) is the marketplace will trump an assigned grade nearly every time.
In other words, not all coins graded the same will be valued the same. The perceived value is what drives the coin's price, not the grade. Coins are not commodities.
See http://www.doubledimes.com for a free online reference for US twenty-cent pieces
Well said: "In other words, not all coins graded the same will be valued the same. The perceived value is what drives the coin's price, not the grade. Coins are not commodities."
I don't think anyone can accurately assign values to exceptional toners like the 1958 Franklin graded MS-67+ by PCGS which realized $129,250. The same problem exists on the opposite end for coins with problems and or negative eye-appeal.
The free market does a good job of assigning values to coins. A coin is worth what someone is willing to pay for it, not what a grader thinks it worth.
Assuming sufficient interest/reward among all parties:
Official graders are nominated by their peers and considered the 'best' for a given coin type
Each coin is separately reviewed by 20 type graders with their grades recorded
The final encapsulated grade is the average of the 20 grades
RE: "The free market does a good job of assigning values to coins. A coin is worth what someone is willing to pay for it, not what a grader thinks it worth."
Correct - this is generally the best determinate of "value."
The only task of "grading" is a completely objective assessment of the surfaces. If built on fixed "way-points," clear open standards, and consistency, the hobby would have meaningful grading and free market valuation.
I see the biggest problem with grading is the human . If you took 100 coins and had the same person grade them say 5 years apart could they grade the same?
How many times has a MS64 coin value $6000 cracked and resubmitted for a MS65 value $50,000?
it would not matter if you had 1-70 or 1-100 same problems.
It’s up to the smart collector to judge a coin on its own merits don’t look at the grade, the bean, or the name on the holder.
The old saying “buy the coin not the plastic”
https://www.pcgs.com/setregistry/quarters/washington-quarters-major-sets/washington-quarters-date-set-circulation-strikes-1932-present/publishedset/209923
https://www.pcgs.com/setregistry/quarters/washington-quarters-major-sets/washington-quarters-date-set-circulation-strikes-1932-present/album/209923
It would be dependent on CAC agreeing to approve the best 2/3ds of each new grade.
I personally don't care for bumping a grade for nice eye appeal. An AU58 is an AU58, not a 61 because it's pretty, likewise adding a point for pretty toning. I don't care for that, just grade it what it really is...
My YouTube Channel
Rarity should have no effect upon the grade. A better date should not get extra grading points because it's scarce or rare.
People complain about net grading, but the Early Copper dealers and collectors do it all the time. How do you rank coins for a condition census or place a value on an item that has an issue? To say you can't net grade anything is avoiding an obvious question. What do you do with those items if you can't adjust the grade for them some how? Do you throw up your hands and toss them in a junk box or melting pot? If you did that, a lot of early coins would be unmarketable and uncollectable.
Fundamentally "grade (condition)" and "value" are completely separate concepts. Just like buying a new car and selling (trading) an old one are separate transactions. The collector (consumer) suffers whenever these are mixed.
Though I have completely switched over to coins, I started out collecting sports cards so my input relates to the TPG's of sports cards. Some have a 10 point scale, some have a 100 point scale.
As it relates to this thread I think an overall numerical grade that is based off of multiple sub categories determines that final grade. What those sub categories are? Well, by comparing what one sports card TPG utilized; corners, surface, edges, and centering how about an equivalent for coins?
The overall numerical grade supported by a numerical grade of obverse strike, reverse strike, rims, eye appeal and add on of any additional considerations such as off center, details, etc.
But I think @ricko hit it as the future will most certainly adopt AI computer grading. Maybe at first to determine a numerical grade based off of surface criteria and a human eye provides the "eye appeal" qualification but eventually this seems like the logical process given the advances in technology and benefits for companies to lower cost through automation. Given the sums of money that flows through the coin world there appears to be opportunity for such a system to develop and be implemented.
Interesting thread.
Interesting thread.
I am now so contaminated with "knowledge" about grading that I have developed/ acquired over the last 50 years, I honestly don't think I could erase it all and get back to square one.
Subjectivity is a tough factor to eliminate.
I know what I like and look for, but also, I have allowed the TPGs to change my preferences to what they like.
And why have I "allowed" them to do this?
Because I sell coins and I want to be competitive and make money by doing so.
What the hell, we keep circling back to this "money thing".😉
What problem are you trying to solve? Developing a way to describe the appearance of coins to someone who is not able to view them? Figuring out how to price coins for sale? Ranking coins for inclusion in competitive collections? Something else?
@EagleEye has some good ideas as far as grading that I can agree with. I haven't looked lately but he had a description on his website.
"A dog breaks your heart only one time and that is when they pass on". Unknown
I agree with the grading of Anciets in many ways. Have categories for:
I also like European grading where there is no such thing as About Uncirculated (like being about pregnant).
If a coin is uncirculated, but weakly struck, say it is "VF, as struck", if that describes the details. Think of a 1924-S Buffalo nickel that is uncirculated, but with no horn.
Creating a grading system for coins? Comparing to school, we would have A Grades, B Grades, C Grades....coins testing on a 100% scale...and would you want coins to follow with a 4.0 grade average? Perhaps not.
How about a zero to one scale with one being perfect? You could have excellent coins grading .999 or mid-range .500 or in really poor shape grading .001 . The quality of the coin would determine the higher number, the usual attributes would be considered....detail, wear, dings, marks, strike... each coin would be judged on its weight too...if a coin is not at original weight, it would lose points on the scale...due to natural wear.
People renovating their houses have a distinct advantage over people building from scratch — they know what they like and want to preserve, and they know what they don’t like and want to change. Moving to a new system would bring all sorts of new problems. The current system may not be perfect, but many of us get the main points and most of the nuances, so I’d avoid an overhaul.