What is the definition of 'consistency' when it comes to coin grading?

bidask · July 23, 2008 12:33PM

The term 'consistency' is often used in description of the worthiness ( or not ) of TPG companies.

Some believe TPG's are consistent, others believe not.

I think it would help to have a 100 point scale to get more consistent.

What's your thoughts on 'consistency' relative to grading?

Can it ever be defined more universally?

Or is 'consistency' ultimately in the eye of the beholder and it never will be defined?

jfoot13 · July 23, 2008 12:38PM

consistency to me is if I send a coin in today and it's a AU55, if I crack it out and send it next week it will still be a AU55

fc · July 23, 2008 12:39PM

<< consistency to me is if I send a coin in today and it's a AU55, if I crack it out and send it next week it will still be a AU55 >>

next week? how about five years from now.
grading standards should not change when the wind decides to blow
a different direction in the market.

mrpotatoheadd · July 23, 2008 12:39PM

<< I think it would help to have a 100 point scale to get more consistent. >>

Assuming a larger range of numbers for MS grades (otherwise, why bother?), I'm not sure how that would work. A wider numerical range would allow for less, not more, consistency. At least, that's the way it would seem to turn out, but I could be wrong...

edix2001 · July 23, 2008 12:44PM

consistency = subjectivity

jfoot13 · July 23, 2008 12:47PM

<< 

<< consistency to me is if I send a coin in today and it's a AU55, if I crack it out and send it next week it will still be a AU55 >>

next week? how about five years from now.
grading standards should not change when the wind decides to blow
a different direction in the market. >>

baby steps...

Placid · July 23, 2008 12:48PM

<< 

<< I think it would help to have a 100 point scale to get more consistent. >>

Assuming a larger range of numbers for MS grades (otherwise, why bother?), I'm not sure how that would work. A wider numerical range would allow for less, not more, consistency. At least, that's the way it would seem to turn out, but I could be wrong... >>

Is it a 82 or a 83

dizzyfoxx · July 23, 2008 12:49PM

"Get it right"...at least most of the time.

SirAnthonyAC · July 23, 2008 12:56PM

Although there are many variables, consistency to me is grading like this from Barber Coins.org

I think there should be a page like this for every coin type

Grading Barber Coins

Edited to add a few photo examples, please click the link above to see more examples and all the details

Very Good (VG 8). On the front, any three letters of LIBERTY on the headband will be present (usually LI and Y). On the back, the rim will be full, and a few feathers will begin to show

Fine (F 12). On the front, all of the letters of LIBERTY will be visible, but the bottoms of a couple letters may be very weak or missing. Laurel wreath has more detail. On the back, about half the feathers and most or all of E PLURIBUS UNUM will be visible.

Very Fine (VF 20). All of the letters of LIBERTY will be easily visible. Still more detail in laurel wreath. On many dates, the band under LIBERTY (see arrow) will be complete on VF30 coins (VF-EF). On the back, most of the feathers will be visible.

Cladiator · July 23, 2008 1:01PM

A 100 point system would in no way make more consistency in my opinion. My guess is it would make consistency at TPG's worse. My idea of consistency when I talk about it is the idea of sending in one coin and getting a grade. Then sending that same coin in 4 more times and each time it comes back with the same grade. Or if it gets body bagged the first time I'd expect a body bag each subsequent time.

golfer72 · July 23, 2008 1:01PM

Most of the inconsistency comes in when grading uncirc coins so I dont see how that Barber grading site helps that much. Looks like Photograde to me. Nothing wrong with it though.

pheh · July 23, 2008 1:01PM

100% consistency is impossible even when dealing with just yourself as the grader. Will you ever see every Morgan ever minted? Will you ever see every MS 63/4/5 "Hotlips"? Can you possibly build a reliable and richly faceted enough scale to accommodate something that, in the very end, is subjective? I believe not. That fact that a slabbed coin itself can in-slab alter in ways which could cloud a grade years down the line isn't something which, to my knowledge, could ever be accounted for.

100% consistency in a TPG really shouldn't even be considered. Employee turnover. Personnel having bad days. And exactly what is described above all play a part.

None of this is to say TPGs can't and don't serve a very useful purpose. But different purposes to different sections of the community.

- To many a dealer a, reliable and *mostly* consistent, TPG is a way of saying "Hey, don't take my word for it - Look at what XYZ says about my coin on this scale of 70."

- To anyone who collects but doesn't have the time or money to invest to become a relative expert in the many facets of grading various types of coins, they can serve as a reliable sounding board for proving out/or disproving what the collector subjectively believed to be worthy of purchase.

However... no TPG is or ever will be the end-all-be-all of authority, and nor should they be.

I'll leave it with this... Could a 1000 point scale fix the fact that all 64 of a given type aren't equal to most eyes. 65s? 70s?

Cladiator · July 23, 2008 1:03PM

<< Most of the inconsistency comes in when grading uncirc coins >>

I see great inconsistancies in the top TPG's grading of circulated coinage.

EagleEye · July 23, 2008 1:11PM

I see a great reduction of grading standards for circulated coins in the last ANA Grading guide. It's vagueness let the grading services interpret grades much different that the previous standards.

For Indian Cents we are starting Fly-In Club Standards, similar to EAC standards.

cameoproofcoins · July 23, 2008 1:15PM

I would hope that a coin is not UNC one time and altered the next time. I would hope a coin is not graded and then graded as one of the finest known. I would hope that a coin is not regraded as finest known. I understand one or two point differences and catching a prior error, but they should at least get it right that it is altered or not.

cheezhed · July 23, 2008 1:18PM

Academic Grading would be more consistent versus Market Grading imo, just for the simple fact that the market by nature is not consistent.

RedTiger · July 23, 2008 4:31PM

100 points helps how? The graders, the dealers and collectors have learned the 70 point scale. It doesn't matter where the 30 points are added, it would just make more tweener coins.

The opposite is true, more consistency in grading would come if they went back to three grades for all coins, Good, Fine, Unc like the old, old Redbooks. Of course, this won't float because it is not good for business.

The grading companies floated the trial balloon of 100 point grading, and the response was overwhelmingly negative from collectors and dealers. It adds nothing, and creates confusion. The bottom line was that it was seen as bad for business.

Computerized grading could be consistent. However, the big two grading companies depend on resubmits for a large percentage of revenue, so probably don't want something that is 100% consistent. 95% consistent or 80% consistent is good enough and generates more revenue via the resubmits. It is more likely that a small start up, or a company from another industry will be what brings consistent computerized grading to the hobby on a wide scale. The tech is here today, even five years ago. The problem is that coin grading is a tiny market in terms of revenue, so it isn't worth the investment needed at the moment.

Zoins · July 23, 2008 6:26PM

<< For Indian Cents we are starting Fly-In Club Standards, similar to EAC standards. >>

Interesting development. I wonder if there will be an additive effect with FIC standards increasing adoption and use of EAC standards as well.

RTS · July 23, 2008 8:53PM

I think it would help to have a 100 point scale to get more consistent....if it's more grading consistency you wish then you only need seven points so-to-speak: G, VG, F, VF, EF, AU and MS or for even more consistency use my grading scale: not-so-nice, nice and very nice...now many may scoff at my grading system but it has served me well.

BAJJERFAN · July 23, 2008 9:30PM

<< 

<< consistency to me is if I send a coin in today and it's a AU55, if I crack it out and send it next week it will still be a AU55 >>

next week? how about five years from now.
grading standards should not change when the wind decides to blow
a different direction in the market. >>

What makes you think its really the standards changing and not merely the fact that the graders are different? Might be the reason that one time a coin is a 63 and next month its a 64. Would also help if the coins were consistent too. I don't think the standards change at all; maybe how they are applied or interpreted. Just because PCGS might overgrade a ms64 Morgan and give it a 65 doesn't mean that they have changed their standards; more likely they erred or thats just the way those graders saw the coin on that day.

MrEureka · July 23, 2008 10:03PM

On a more practical level, what can the TPGs do to become more consistent?

BAJJERFAN · July 23, 2008 10:12PM

<< On a more practical level, what can the TPGs do to become more consistent? >>

Offer jobs to some of the experts here.

dorkkarl · July 23, 2008 10:42PM

<< What is the definition of 'consistency' when it comes to coin grading? >>

if you like the coin, if you really REALLY like it, then the grade (price) does not matter

you stick to that advice w. 100% honesty, & consistency takes care of itself EVERY TIME

K S

RedTiger · July 23, 2008 11:59PM

<< On a more practical level, what can the TPGs do to become more consistent? >>

For high profile, low pop coins they can do a detailed computerized image of the coin and put it in a database. Then when a coin of that same date/mintmark comes in they can have expert system software, plus human eyes scan that database for a match. This isn't time efficient for common coins in common grades, but for certain dates and/or certain grades it can be. This would translate into less doctoring, less dipping and stripping, less accelerated toning. It also means less resubmits, so keep in mind that there is a point where more consistency hurts the current business model and the companies will not implement those changes.

doh · July 24, 2008 12:08AM

<< I think it would help to have a 100 point scale to get more consistent. >>

No... The more numbers in the grading scale the LESS consistent grading would become.

<< Or is 'consistency' ultimately in the eye of the beholder and it never will be defined? >>

Bingo.

19Lyds · July 24, 2008 12:24AM

<< I think it would help to have a 100 point scale to get more consistent.

What's your thoughts on 'consistency' relative to grading? >>

Personally, my observations have been that regardless of what system is used, there are always going to be folks grousing about it!

MrEureka · July 24, 2008 12:26AM

<< Or is 'consistency' ultimately in the eye of the beholder and it never will be defined? >>

Bingo.

Not really. It's "accuracy" that cannot be defined. Consistency is easy to measure. Just look at the results for coins that are submitted multiple times.

ricko · July 24, 2008 12:40AM

Consistency.. or more to the point, accuracy (as Andy stated) will be achieved with the advent of computer grading. Cheers, RickO

relicsncoins · July 24, 2008 12:56AM

It will probably happen at some point. As revenue starts to dry up at the TPG's for newly submitted coins, presto the new 100 point grading scale with a special deal to get your coins in the "new" holders.

JJ

TwoSides2aCoin · July 24, 2008 1:04AM

On a scale of 1 to 10
consistency is a 10, theoretically.

CoinRaritiesOnline · July 24, 2008 1:20AM

<< On a more practical level, what can the TPGs do to become more consistent? >>

Well, a grading set for each major series would help. AND -- that grading set has to be used by the graders, not stored away in some vault.

I know, I know, the top 2 already HAVE a grading set.

I remember seeing the grading set of Standing Liberty 25c that was displayed by one of the top 2 grading firms several years ago.

I gotta tell you -- those were some beautiful coins! The 62 looked like a 64+, the 64 looked like a lock 66, etc. I would have paid a huge premium over the "market" levels to buy that set.

I believe at least PCGS (and maybe NGC too-- I'm not certain) have expanded on the basic grading set idea in more recent times, i.e. they have several examples per grade, including an MS-63 but with a weaker strike, etc. If these sets are used, they will go a long way towards maintaining a consistent product.

Unfortunately, comparing submitted coins to a grading set takes time. And if you are grading 80,000 to 100,000 coins each month, the time spent per coin must be managed very carefully.

BAJJERFAN · July 24, 2008 1:28AM

<< 

<< On a more practical level, what can the TPGs do to become more consistent? >>

Well, a grading set for each major series would help. AND -- that grading set has to be used by the graders, not stored away in some vault.

I know, I know, the top 2 already HAVE a grading set.

I remember seeing the grading set of Standing Liberty 25c that was displayed by one of the top 2 grading firms several years ago.

I gotta tell you -- those were some beautiful coins! The 62 looked like a 64+, the 64 looked like a lock 66, etc. I would have paid a huge premium over the "market" levels to buy that set.

I believe at least PCGS (and maybe NGC too-- I'm not certain) have expanded on the basic grading set idea in more recent times, i.e. they have several examples per grade, including an MS-63 but with a weaker strike, etc. If these sets are used, they will go a long way towards maintaining a consistent product.

Unfortunately, comparing submitted coins to a grading set takes time. And if you are grading 80,000 to 100,000 coins each month, the time spent per coin must be managed very carefully. >>

So do the graders refer to the set when they have a coin for which its a tough call or do they use it more as a refresher? Or both?

ColonialCoinUnion · July 24, 2008 1:28AM

<< I think it would help to have a 100 point scale to get more consistent. >>

Is this a joke?

BAJJERFAN · July 24, 2008 1:30AM

Consistency would be all 15 graders at PCGS ALL giving the same coin the same grade, not just 2 of the 15.

roadrunner · July 24, 2008 7:48AM

Consistency could be improved by allowing more grading time per coin and improving the skill of each graders. I felt that TPG grading consistency was at 80-85% in the later 1980's in the MS/PF 19th century type coin arena. Today I feel that this area has slacked off some in order to cover moderns as well. Today we're at 60-70% at best imo. I felt 80% was quite reasonable at the time. And I think 90% is achievable if PCGS/NGC really wanted to go there.

The consistency of 3 graders should be better than any of the individual graders. Yet today it seems we have individual graders who can hit them at 70-85% on their own, yet with 3 of them together the rate drops! Go figure. Theoretically, 3 graders at 70% gets you 98% accuracy. I'd be thrilled with 90%.

roadrunner

Aegis3 · July 24, 2008 7:53AM

<< On a more practical level, what can the TPGs do to become more consistent? >>

Reduce the number of grades. That really is the only way to increase consistency. Of course, there's the small issue that it's not going to happen.

bidask · July 24, 2008 8:06AM

<< 

<< I think it would help to have a 100 point scale to get more consistent. >>

Is this a joke? >>

No its my opinion. I'm not sure but I don't think the Europeans think its a joke with their grading system.

So, Mr Colonial Coin Union, how do you define consistency when it comes to coin grading?

roadrunner · July 24, 2008 8:09AM

Reducing the number of grades will not help one bit. We had that back in the late 70's: BU, Choice BU, Gem BU, Superb Gem Bu or loosely 60, 63, 65, and 65+.

Yet there were many saavy dealers and collectors who split those grades up into 60, 60+, 60++, 60+++, 65, 65+, 65++ (or in today's numbers 60, 62, 63, 64, 65, 66, 67). Some of them also used qualifiers such as A, B, C for strike, eye appeal, marks. This was long before the official TPG's. So you can reduce the grades as you like, but the market will dissect them further and charge accordingly. I know I had to pay big premiums for 65+ and 65++ coins back in 1975-1977. They were still called gem MS65 but you couldn't buy them at auction for that kind of money. Quality ruled and in finer gradations that were formally published on the price guides. No different today where we have 11 MS grades and people have broken those up into 40 or more grades.

roadrunner

derryb · July 24, 2008 8:23AM

SGS?
Consistency

charley · July 24, 2008 8:41AM

Would it not be true that prior to establishing consistency, in the context being asked, there must first be an established basis of agreement of a grade/condition/color/strike/luster/wear by definition for every coin type, that is a universal definition accepted by all involved, be they TPG Graders, collectors, or dealers? Then we would have to establish a vision requirement that will be the standard for consistency and finally a testing process that would allow verification of the ability to be consistent. Until then, 100% consistency is not possible, because we are human and different. The only consistent ability will be the ability to offer an opinion. Increasing the scale to 100, 200, etc. doesn't do anything to establish consistency. An individual may train him/herself to be consistent within their own opinion, but this doesn't mean there is universal consistency, even if three or 100 other individuals have the same personal consistency.Respectfully, John Curlis

Aegis3 · July 24, 2008 12:01PM

<< Reducing the number of grades will not help one bit. We had that back in the late 70's: BU, Choice BU, Gem BU, Superb Gem Bu or loosely 60, 63, 65, and 65+.

Yet there were many saavy dealers and collectors who split those grades up into 60, 60+, 60++, 60+++, 65, 65+, 65++ (or in today's numbers 60, 62, 63, 64, 65, 66, 67). Some of them also used qualifiers such as A, B, C for strike, eye appeal, marks. This was long before the official TPG's. So you can reduce the grades as you like, but the market will dissect them further and charge accordingly. I know I had to pay big premiums for 65+ and 65++ coins back in 1975-1977. They were still called gem MS65 but you couldn't buy them at auction for that kind of money. Quality ruled and in finer gradations that were formally published on the price guides. No different today where we have 11 MS grades and people have broken those up into 40 or more grades. >>

Isn't then just the issue that we should reject the false equation grade=price? Once we accept that coins of the same grade can have different prices, why pretend that we need a different grade or sub-grade or condition or strike qualifier or whatnot to justify price differences?

JSsson · July 24, 2008 4:28PM

I think consistency would be great with all the grading companies. I define consistency in grading when a grading service can grade the same coin multiple different times and reach the same grade every time. That doesn't seem totally possible since grading is subjective.

What is the definition of 'consistency' when it comes to coin grading?

Comments

Leave a Comment