Population Reports Accuracy?

NGC & PCGS, to name a couple, have population reports on their site for different grades. My question is this: how accurate are those?

This example is classic and I know talked about often; I wanted to start this thread to see more feedback and thoughts. If you have a coin that is let's say PCGS MS-64, but the price difference is huge from 64 to 65, so you crack it out and resubmit. At first PCGS shows 100 in MS-64 and 25 in MS-65 (all for example) It comes back MS-64 again. You repeat this two more times before deciding to give up. Now will it show 104 in MS-64 and still 25 in MS-65?

Does this dilute the price of MS-64 by giving the impression a lot more exist than actually do? Does PCGS ever "correct" these numbers or is it rolling? So if somebody cracks a coin out for their set then later slab it again some years later, will it account for 2 of the coins shown in the population report?

Lastly, and perhaps a bit different than my previous questions: is there a way to guesstimate a particular coins 'real' population? For example, how many 1893-S Morgan dollars actually exist in all conditions?

 

Log in or Sign up to hide this ad.

The population reports are what they are, which at this time means they simply are another indicator of the number of times each issue has been sent into any TPG and how many grading events each has received. It's not perfect by any standard, but it is still information that can be worked with and interpreted.

If your hypothetical PCGS MS64 is sent back to PCGS as a raw coin then it would indeed count again in the population report. If, however, it is sent back in its holder then the old holder data would be removed from the population report. Additionally, if used certification tags are sent back then they are removed from the database and population report, too.

Guesstimating extant populations has been done to some degree on the PCGS CoinFacts website, but specialist groups have for many years done more detailed analyses regarding how many coins they believe have survived. These rarity ratings are supplied for many type coins from groups such as the Bust Half Nut Club.

 

LikeTom said, its a data point. You are very right about crackouts distorting these numbers, but since they are the only hard numbers anyone has, they try to glean as much as they can from them concerning relative rarity. Doug has convinced me I am one of the few holdouts nowadays who put away nice coins years ago and have never had them slabbed. When people like David Bowers estimates 85% of all coins worth grading have already been graded, I guess I am the weirdo holding nice coins not in slabs. I wonder if my kids sell mine in 40 years if people will even be willing to accept that some nice coins have survived all of these years NOT slabbed, lol.

 

I am going to use the term bottleneck grades. That is grades that are right below a major threshold in price. Typically these are grades such as MS-64. If in MS-65 there are 500 reported on PCGS, but 4,500 in MS-64, how much should be discounted as reslabs? I've often said to myself that there realistically is only 1,000 and the other 3,500 are just resubmits. What are the thoughts on this rationale?

 

How are you coming up with that ratio? It seem pretty high.

 

A dealer like Tom would have a better opinion, but I am not sure its that high. There is a reason there is such a large jump, usually having to do with how they were handled en mass after leaving the mint. I know exactly where you are going, but simply do not know how large an issue it is. Always remember that a lot of those 64's are already overgraded, so why even attempt to resubmit those, right? If I had to guess in your scenario, I would guess more like 1500-2000 of them are resubmits, instead of your 3500 guess. Usually only upper end 64's are resubmitted hoping to upgrade, so I am not so sure its as large as you think.

 

To be honest, complete and wild guess. No facts behind it whatsoever. What do you mean by high? Think I'm discounting too many or too few?

 

What Doug said.

Including his question

 

Fair enough. Realistically I was trying to figure out just "how rare" something really was. With coins that there are only total population reports of 2-3K, I was wondering of these how many actually exist. The reason for this was to see how many I could buy to hold a sizable amount of them. For example (complete fictitious numbers) if there are 2,000 total 1893-S Morgan Dollars in the population reports, how many do I actually need to buy to hold 10% of them in the market. I may only need to buy 150 of them to hold 10% because of that 2,000 reported some are resubmits, etc.

 

One can already determine this by getting a hard copy of the TPG population report from any given date and then comparing this report to the current population figures. I have population reports going back twenty years that allow me to see how many coins were added to specific populations in given time intervals.

 

I'd also be interested in obtaining said copy.

 

Here's a thought exercise. Suppose you have a coin for which there is a huge jump in price between grades (say $200 to $2000). You are convinced that cracking and re-submitting will eventually get you the bump in grade. How many times would you re-submit before you give up (assuming you never get the bump)? The answer to this question may provide insight into what the ratio of the two grades might be. Notice I said insight, not a hard and fast answer.

 

PCGS sells the population report-

http://www.pcgs.com/popreport.html

 

If you wanted to think about it this way, then I would say it depends on the probability I believe I have of upgrading. If I have a 20% chance of upgrading for a value increase of $1800, then the expected payoff is $360, quite a bit more than the $30 submit cost so I could continuously submit until my belief in probability changed.

 

You are assuming it's random chance like the lottery. It's not, or at least shouldn't be. Trust me, finding a high end 64 in such a situation will be tough, since everyone and their brother has already been looking for these.

 

Actually, I really was not thinking of random, but rather, a situation that apparently occurs frequently - resubmission until the desired grade is acquired, and when to give up. One thing I am assuming is that the greater the price jump, the longer people will keep trying. Eventually, it reaches a point where the cost of further submission becomes counter-productive. My bet is that the coins and grades with the greatest price bump, will have less accurate accurate figures on the low end. In other words, if a 64 to 65 bump is large, the 65 population numbers should be pretty accurate, but the 64 population values will be skewed toward the high end. Take the 32D quarter. PCGS lists a 64 at $3850 for a "+" coin and $11,500 for a regular 65. Assuming $50 a pop for re-submission, one could resubmit the coin 100 times and still make a $2600 profit if it bumped on the 100th try. Given the inconsistencies in the grading that have been discussed here frequently, I would think that many people would do at least 10 resubmissions before giving up. Recall the number of posts describing upgrades, sometimes after three or four tries, of various sorts in posts over the past year. I don't think the supply of candidates for resubmission has dried up yet.