Beckett / BGS4657

Originally Posted by shrewbeer
Quote:

Originally Posted by jrs
In all seriousness, how can the census -- even a majority of it -- be accurate when anyone has the ability to crack open a slab and resubmit the book for grading (a practice we all know occurs, especially since pressing has become so prevalent)? So while it's a useful tool (because there's nothing else like it), I'm not sure I'd qualify it as an accurate one.

It's definitely useful in certain analysis, but you're correct;not accurate. To Bob's point, deceptive enough to novices that are led on to "trust" it and be taken advantage of.

To the extreme, this reminds me of the guy on ebay selling a .5 GI joe #1 for crazzzy money with the logic (and selling point) of "RARE! ONLY ONE IN CENSUS!)

Originally Posted by jrs
In all seriousness, how can the census -- even a majority of it -- be accurate when anyone has the ability to crack open a slab and resubmit the book for grading (a practice we all know occurs, especially since pressing has become so prevalent)? So while it's a useful tool (because there's nothing else like it), I'm not sure I'd qualify it as an accurate one.

Originally Posted by Lonestar
@DocBrown I'm sure it's not the same, but your comments made me laugh the same way as that scene. All good!

Originally Posted by DocBrown
Quote:

Originally Posted by jrs
In all seriousness, how can the census -- even a majority of it -- be accurate when anyone has the ability to crack open a slab and resubmit the book for grading (a practice we all know occurs, especially since pressing has become so prevalent)? So while it's a useful tool (because there's nothing else like it), I'm not sure I'd qualify it as an accurate one.

That's a fair question, and the answer is this:

1. The vast majority of books are "unimprovable" to begin with. CPR isn't as big a business as some think it is. (By "unimprovable", I don't mean pressing won't make them look better...simply that they have technical defects which prevent them from attaining a higher grade.)

2. A good chunk (this didn't use to be true!) of the books that ARE improvable, are being improved PRIOR TO being slabbed.

3. The vast, vast majority of the books represented on the census are "first time" and a lot of those are "only time", because they are subbed and/or bought by collectors who have no interest in CPR. While no one can ever know completely accurately, the truth is, for the vast majority of books, people are NOT cracking them out, resubbing them, and then not turning in the labels (thus skewing the census.)

How do we know that? CGC keeps a photographic database of "the rare books", which they can compare if another copy comes in, so they know if it's a resubmission, or a new book to the service. Also, auction houses like Heritage keep photograpic records of these books, so the market has a really good idea if a particular book was resubbed, or is a new copy in a new slab.

If a book like All American #16 comes up for sale, many, many people compare that copy to photos of previous copies sold, to see if this is a resub, or a new book in a new slab. And, for the most part, the majority of new slabs have NOT been resubmitted books.

And, the more copies of a particular issue that get submitted, the more the census tends to (that's the key phrase, there) be accurate. For example: there are 104 Hulk #181 9.8s, and 10 SS 9.8s. Now, the odds are fair that some of those 9.8 Universals ended up becoming some of those 9.8 SS. But clearly not the majority. And, since 9.8s are not going to be cracked out to be CPR'd, it's probably a good bet to say that there are 90-95% of those copies still slabbed, so the census for that book is very close to being accurate.

Now, there are 258 9.6s. Now, certainly, some of those 9.6s (and we have no way of knowing how many) ended up in "not 9.6" slabs...whether higher or lower...and those 9.6s labels didn't get turned in. Also, there's a small percentage that stayed the same, and a few that turned into SS copies.

However...again, we're looking at...and keep in mind, this is a guess, still...probably 80-90% of the 9.6 copies still existing in those 9.6 slabs.

That said, however, the lower you get on the grading scale, the more you run into "unfixable" defects, and the less likely books are to be CPRd in the first place. Less LIKELY, not that it doesn't happen.

In other words, it's not 10%, or 25%, and the rest are resubmissions. You see that on VERY, VERY rare items...census numbers represented in single digits, for example...but not over time, with common or low value items.

4. CGC used to give a credit for people who turned in old labels. This tended to keep the census more accurate than not.

5. Even though CGC doesn't give credit anymore, there are still quite a few people who, in the interest of keeping the census accurate, turn their labels in. The motive for the census being inaccurate...that is, the census cannot be artificially LOWERED (well, theoretically, people can make counterfeit labels to turn in, but that's getting REALLY out there), it can only be artificially RAISED...is weaker than the motive for the census to BE accurate. After all...if you're using the census as a selling point, you want it to be as close to accurate as possible...that is, LOW...rather than higher. There's no "upside" to having a HIGHER census number as a seller. It gives the appearance that a book might be less rare than it actually is (unless you get really conspiratorial and argue that a seller might want to hurt competition...if so, that's quite an elaborate scheme to pull off!)

Because of all these factors, the tendency is that the census is more accurate than not.

"But how do you KNOW that? No one REALLY knows, so how can you say?"

It's true, no one knows for sure, unless you have a book with a census population of, say, 3, and you own all 3 slabs. But, because we can see patterns of behavior over a long period of time...15+ years now...we can predict reasonable outcomes based on what has been observed. And since A. the majority of books cannot be "improved"; either from not having pressable defects at all, or because they were pressed prior to certification, B. the majority of buyers aren't looking to "improve" them, and C. those books that ARE CPR'd have tended to have their labels turned in, then we can say, with reasonable certainty, that the census is pretty accurate.

Yes, there's a LOT of conjecture, here...no doubt. But we can know what tends to happen, so we can make reliable predictions about how accurate the census really is, even if we can never know with absolute certainty beyond a few specific cases.

100% accurate? No, not at all. But accurate enough to be of statistical value? Absolutely.

Here is data on the census over the years, by Valiantman, aka Greg Holland:

http://www.cgcdata.com/cgc/totals/

There, you can see census numbers go DOWN, as labels got turned in.