How accurate do you expect PCGS grading to be?

DHeath

When you submit, is it your expectation that PCGS will be accurate enough to assign the same grade to a given coin, say 5 times in a row? How about a different grader looking at the same coin in 5 years? Do you expect a borderline coin to produce a mixture of two grades out of 5 submissions? If your borderline coin is sandwiched in the submission by lock higher grade coins, do you think the grader is influenced either way by the coins they've just looked at? How many coins out of say 100 do you expect consensus from the submitter on? Just food for thought. How accurate can grading be? I don't mean the softballs. Almost anyone can grade an xf coin. I'm thinking particularly about AU58-MS68. How accurate can a human be, and is that our expectation?

kranky

With respect to AU58-MS68, I think it's folly to expect every coin to be graded the same every time. It cannot be done. The graders change, and the differences between MS grades are very, very slight.

A scale of 60/63/65/67 made it possible to grade coins the same time after time. Now we split hairs so fine, the top experts do not always agree on a coin's grade.

What would be interesting is if PCGS had a "secret shopper" concept. Take a set of coins, and make arrangements to have them sent in periodically by different people, so PCGS could track repeatability over time.

TheNumish

If I sent in 100 coins today I'd expect 30 to be undergraded, 60 to be graded correctly and 10 to be overgraded. Then there are packages I just get whacked on where over half of the coins are undergraded. Had one package that I sent 10 coins back for regrade and 8 of the ten went up a point. I wonder how many more coins from that submission would have gone up.

The crazy part is that most of what I deal in is modern coins so no other service has the liquidity or price levels PCGS has. I'm pretty much dependent on them for a living so I have to live with it. I live pretty good so I'm a pretty happy person.

KlectorKid

I expect PCGS to me more accurate than a dealer no more no less.(I buy coins that are graded so no dealers try to screw me out of money)

ursabear

Mr. Don,

Now here is a thoughtful question. I had to read it a couple of times to think it through.

When I send a batch to PCGS, I expect them to grade them fairly (relative, I know). When the coin/coins is/are a series that I am still learning about, I expect them to be more expert than me, and to give a grade that not only confirms the coin's state, but helps me learn. When the coin/coins is/are a series that I understand very well (like MS Sacagaweas or proof Kennedys), I actually expect them to grade them as I see them (a "confirmation", I guess).

What really happens isn't very far from what I expect. What has actually happened in the hundred or so coins I have submitted has mirrored (for the most part) my expectations - for example: Each batch has come back largely as I expected, but with one coin very undergraded to my way of thinking. Every other batch actually comes with a coin that grades better than I expected.

My percentages are about 10% under what I expected, 80% what I expected, 10% more than I expected.

To return to answering the question, I expect them to keep in this 10/80/10 range.

FC57Coins

Don:

Over the past three months I've sent over 50 coins to PCGS and I'd say the majority of them - 90%, were graded correctly, with maybe 10% falling into the category of should've gone one way or the other. I think too much is being made of the issue that PCGS is grading conservatively. Every time I hear this, I wonder what people would be saying if they were to grade liberally! - No, I take that back, I do know - just look up any posting having to do with ACG. Anyway, I think sometimes we study our coins to death because they're "our coins" and we feel a need to know everything there is to know about it. PCGS looks at our coins for a grand total of 10 seconds or so to assess their grade. Maybe we should be asking ourselves if we're being a little too anal retentive with respect to what our expectations of PCGS and NGC should be.

Frank

fcloud

There are always going to be fringe coins. Can you really detect a coin of say 64.9 vs 65.1? These coins may go back and forth if you were to send them in five times. However, a solid 65.5 should come back the same every time. As far as cam dcam. I have seen coins with great mirrored surfaces and white cam graded without designation, and coins that look the same as cam. It is subjective, and if you are buying the coin and not the holder, you should be happy when you can pick up a premium coin in which may go up a grade if resubmitted. On the other hand, you should be able to buy an overgraded coin for less than the going rate of said coin. If you feel a coin does not make the grade on the holder, skip it. The collector is the final grader. It either makes the grade or it doesn't.

nwcs

Well, what you're asking for is consistency. I expect that all the time. However, I don't expect precision because it's not possible. Everything can influence a person's perceptions from the jerk who cut in front on the drive to work to the indigestion from a bologna sandwich. Or even positive things in a life can affect the grades like a nice tax return or finding $5 on the sidewalk during lunch. And even the coins in the stack can influence it a little. After a load of circulateds, the mint state one can either look lower graded or higher graded than it actually is.

coinguy1

Hi Don,

In answer to your questions:

There are certain coins which I think we should be able to reasonably expect NGC or PCGS to grade the same, five times in a row. Some coins, in my opinion, obviously belong at one particular numerical grade, end of story.

Other coins, are liners, where many experts would be split between two different possible grades. Of course, if those coins don't receive the higher of those two realistic grade possibilities the first time, they are often resubmitted, for as long as it takes, until they do.

There is a third category of coins which are extremely difficult to grade and over which there can be a wider variation of grading opinions, even by experts. One such area would be pre - 1808 bust coinage. Many of those coins possess "cabinet friction" or "rub", which are distinguished (by some people at least) from "wear". I have seen experts debate whether such coins should grade AU58, MS60, 61, 62 or even 63. Those are toughies!

Just last week in Baltimore I viewed a Draped Bust half dollar in a PCGS MS64 holder, which NGC had graded MS62 the previous week. It was highly lustrous, well struck and had a bit of "friction" on it. It looked to me, to be a 63/64 liner.

There are other coins which, for one reason or another, don't always neatly/easily fall within a 2 point grade range, but I believe the overall percentage of such coins is quite low.

I do think that graders can be and are influenced by the overall "look" of a group of coins submitted on the same invoice. It is human nature to think things such as "nice group of coins here" or "well, that's 3 dogs in a row I've just looked at", etc.






When you submit, is it your expectation that PCGS will be accurate enough to assign the same grade to a given coin, say 5 times in a row? How about a different grader looking at the same coin in 5 years? Do you expect a borderline coin to produce a mixture of two grades out of 5 submissions? If your borderline coin is sandwiched in the submission by lock higher grade coins, do you think the grader is influenced either way by the coins they've just looked at? How many coins out of say 100 do you expect consensus from the submitter on? Just food for thought. How accurate can grading be? I don't mean the softballs. Almost anyone can grade an xf coin. I'm thinking particularly about AU58-MS68. How accurate can a human be, and is that our expectation?

moosesr

I am giving PCGS sort of a grading test. I just sent in 14 1972P Eisenhower dollars that I cracked out of PCGS MS64 holders. These coins were graded about a year ago. I don't have any idea what they will grade this time around, but I covered the grades and mixed two 1972P PCGS MS65 coins in with the group and was only able to pick one of the MS65 coins out of the group. The coins all look good enough to me to be MS65 grade. I have know idea what they will grade this time around, but I am hoping for at least one or two MS65 grades. What do you think they will grade this time?

Charlie

supercoin

A scale of 60/63/65/67 made it possible to grade coins the same time after time.

Certainly fewer points would make it easier to get the same grade consistently. But you still have the "in between" coins. And, with fewer scales, the price jumps between grades would be even more extreme than they are now.

And the price jumps, of course, are the reason why anyone cares about accuracy.

So you'd have fewer but bigger problems. I'm not sure that's an improvement.


To answer the original question... I think +/- 1 point is about the best you can hope for with MS coins. Though the market doesn't price them that way once they're in the holders.

EVillageProwler

As Mark Feld said, there are three types of coins: the solid-for-grade ones; the liners; and, the ones that are super tough to grade due to technology or design.

For the first, I expect a high degree of consistency. Something like 90% consistency seems reasonable.

For the second, I expect that 90% of the grades would be one of the two grades between which the liner falls.

For the third, that's hard to say because the specific issue type matters. Still, I'd expect a 70% consistency rate between two grades. Example: an EF-ish Early Dollar should grade either VF35 or EF40 about 70% of the time. Or, those super-slider Seated coins, for which I'd expect the variation to be with AU58 and MS62 or MS63.

EVP

Russ

<< <i>I am giving PCGS sort of a grading test. I just sent in 14 1972P Eisenhower dollars that I cracked out of PCGS MS64 holders. >>



moosesr,

If you would, could you post the results of the experiment when completed? I'd be very interested, and I'd guess others would also be.

Russ, NCNE

fcloud

Mooser,

Now that you posted you are sending these 17 in for a test, they will all come back as 64s.

leothelyon

Let's say every coin was accurately graded.........where do we go from there?

Leo

DHeath

Thanks for the thought provoking posts all. Since I have been hoping for more consistency in grading, I really wondered if what I hoped for was unrealistic. Obviously we all have differing grading skills, and collect different series. I believe I am competent in a few series, but many of you have experience that is much more vast, and it is always educational to hear your thoughts. I certainly agree that some coins are almost instantly identified as x, and feel sure the grading services won't miss those once out of 10 tries. Sliders, tweeners, rub/no rub, cam/dcam, hairlines seem to be the trouble areas. One last question please, what in your estimation would best improve the results the current mass production grading produces?


OK- I more question. Is 99% accuracy achievable?

IrishMike

Don, less submissions of same.

DHeath

OK Mike....I should have seen that one coming.....LMAO

michael

the numish summed it up really well!!!!!!!!!!!!!!!!! the best i have ever heard

he said i think

"" i am dependent on them for my living and i live pretty good""

that tells me that most all
BUT

not all coins in plastic slabs the plastic makes most of the value of the coin
they maximise the value of the coins and for some/many coins the slabs really make all the difference as the plastic holds a huge value!!!!!!!!!! in the value of many coins in holders

do not break these coins out of their holders!!!!!!!!!!!!!!!

buy the coin not the holder

i think there will be a day of recokening for many slabbed coins when the music stops playing there will be few chairs and many will be left standing holding the bag

sincerely michael

michael

for me i think for many many coins but not for all coins

the grading services are pricing services NOT grading services at all

especially so for the current market leader slab where the plastic makes the biggest value of a coin encapsulated as such

and that is okie nothing wrong with that just make sure and understand what you are getting yourself into
because when the music stops many will be left holding the bag and all those so called dealers you buoght these coins from will not even want to buy these coins back even at 10-25 cents on the dollar if that but that is also okie but that also tells you alot about the coins you are buying from them in their respective plastic holders!

just let the buyer beware

sincerely michael

jom

<< <i>And the price jumps, of course, are the reason why anyone cares about accuracy. So you'd have fewer but bigger problems. I'm not sure that's an improvement. >>



Superman: I wholehartedly....disagree. Not that the price jump wouldn't be there but because you would let the MARKET decide the in-between grades. Uh..oh..look out! We are all gonna have to learn to grade. God help us!

Everyone learning to grade is definately an IMPROVEMENT.

jom

DHeath

Michael,

I understand the argument you present, and think there is some validity. As primarily a submitter, rather than a slabbed coin purchaser, I pay between $10 and $100 for encapsulation. I know what the price of the plastic is every time, so I rarely pay a premium for it. My primary area of concern is protection and guarantee of coins that I already own. My raw collection is perhaps 2000 coins, and my slabbed collection is perhaps 400 coins. I have been collecting since I was a kid (a long time ago). I submit the coins I believe have enough value as to benefit from encapsulation on the off chance I should ever have to liquidate my collection, or my family should have to. Every dollar I spend on grading reduces the amount of money I spend on coins. Thus far, I've spent perhaps $8000 on grading fees. I imagine/hope those fees have increased the marketability of my collection. Obviously I am interested in the grading service practicing due diligency and assigning the correct grade the first time I submit. When I began submitting, I preferred PCGSs grading of modern proofs. My choice of grading services is series specific, and I submit to NGC and occasionally ANACS for series I believe they do a better job with. It is from that point of view that I ask the question that is the topic of this thread. I'd love to hear your thoughts. May I end this babel by saying, fluctuating standards at the services tend to render them far less useful to me as a collector, because even though I buy the coin, and not the holder, when it's time to sell, my family probably won't be nearly as savvy about their value, and I certainly don't feel like paying multiple grading fees to achieve my goal. Coincidentally, I dislike overgrades as much as undergrades. When you own many examples of a single coin, grading disparity is more obvious. That was the intent of my question. It is an easy question to dismiss if it is co-mingled with issues such as greed/profit, modern/classic etc. My concern is specific to my collection. That's the real reason for my interest in PCGSs consistency. Honestly, if it weren't for the possibility my wife of childred might have to eventually sell my collection, I would submit them all to a service such as NCS that would attribute them without assigning a grade.

IrishMike

Don, what you say makes sense, the other wild card will always be the continued economic viability of grading companies and which particular service will be more in favor at a point in time. That is why I keep harping that it is vital that at least two companies remain economically sound and consistent in their grading standards. Just look at what is happening now, you don't think especailly with the ecnomic problems at CLCT, they are starting to realize that they have to address customer issues and grading consistency. My biggest fear with PCGS is that Hall reads these boards agrees with the vocal minority who think everything is OK and does nothing. We might be looking at our PCGS slabs someday wishing for the good old days when the company was either around or in the top three of services. Then I wonder the value of submission fees. At any rate the future is always interesting.

DHeath

Your right Mike, and I won't pretend I have no interest in how the market perceives the grading services. When my collection is sold, the credibility/market acceptance of the holder will certainly impact the price the coins receive. As I listen to the boards, tons of advice is given about which holdered coin to buy, how plastic is overpriced, why you shouldn't buy valuable coins raw, slab prices are silly....etc,etc,etc. My point is simply that all of these diatribes completely ignore the real core of the grading companies support, the collector/submitter. None of us represent the largest part of the market for the services at any point in time, but most of us are lifers, and over the long haul, it is our support that will determine the success or failure of any particular coin related service. The collector is the end user. BTW - The services all know that, and gladly accept submissions from collectors.


BTW - Multiple successful services = better service for the community.

RotatedRainbows

If the coins in PCGS slabs are accurately graded as so many people keep trying to claim (or undergraded according to some) then if PCGS went out of business tomorrow it would not affect the value of coins in their holders one bit. Because in the end it is the same coin isn't it?? Unless of course people are paying for plastic without regard to the coin inside, I know all my slabbed coins have value because they are nice desirable coins not because they have a special plastic holder around them.

DHeath

RR,

I agree with what you are saying, but you miss my point. Do you believe your heirs will be able to sell your holdered collection for what it is worth if it is misgraded by the grading service, or if it is sold raw? You could probably properly liquidate your collection, but could someone else? There is value in an independent third-party opinion only if it is accurate. IMO

IrishMike

Anyone who is either collecting coins in volume or expensive coins in PCGS slabs should be concerned about the reputation of the holder. Look at ICG in their early years compared to the value of their slabbed coins now. Whether we like to admit it or not, the slab adds value to many collectors. Crack out you PCGS coins and send them to NCG and post them for sale in the registry forum. As leader of the comapny David Hall needs to bring in people who will maintain their prominent position.

mdwoods

I would have to agree with Mark Feld and EVP on this one. Overall I would expect to have about a 90% accuracy. The more I learn about grading (the more coins I see), the more I can differentiate between the higher MS grades. And the more I realize how hard it is to be objective if you are grading your own prized coins. mdwoods