Here's One for ya!
jdimmick
Posts: 9,783 ✭✭✭✭✭
So an auction company just put up a list of items coming up. Low an behold there is a scarcer variety large cent in the group that is pcgs graded and cac'd. They note that they cannont seem to find a provenance for this particular coin, and no past auction appearances. The provenance I understand because I know where this coin came from, but past auctions I cant, as they have not only sold it once, but twice in their own auctions the past year and half or so. I have to admit that I get ticked off everytime I see this coin because I brokered this coin for the family of where it originated. I tried it at pcgs, they wouldn't holder it, came back xf details scratched, ngc said the same thing, we finally gave up on the coin, and I sold it to a national coin company (not the same firm who is conducting the auction) on their behalf for a mere peanuts compared to what it sold for the first time in auction 6 months later after it was not only holder at a significantly higher grade in a regular holder, but also cac'd. Probably sell for strong amount again. This coin was put away since the 1950's originally by the grandfather of the person who was liquidating the coins years later.
Comments
I had a friend who had a chain cent that kept getting details graded. He tried it multiple times. He finally gave up and sold it at the Long Beach show. The dealer he sold it to stuck in his face before the end of the show in a problem free Fine holder after he submitted it at the show. Believe me... I know where you are coming from.
I hope this thread get addressed instead of deleted. You’ve raised a serious concern. It deserves a serious answer.
Both incidents are completely out of line. I'm sure it happens more often than we hear about here.
Great transactions with oih82w8, JasonGaming, Moose1913.
From what you can tell, from the time you tried the coin to the time it got holdered, was it still the exact same? Or, do you think it had some work done on it?
I've been told I tolerate fools poorly...that may explain things if I have a problem with you. Current ebay items - Nothing at the moment
When shopping for colonials and early type I've seen numerous examples, based on photos, that go both ways. I just try to avoid anything that has distracting scratches or porosity. Of course, I'm not after rare varieties or the like, so I can be picky.
It’s been my experience that this thread will be closed and/or deleted.
From a strictly customer service standpoint the issue begs a response. When you consider we are all quasi shareholders with an interest in maintaining the fidelity/integrity of the product we hold, the case only gets stronger. It’d be a mistake to view the concern as just some customer gripe.
The way it appears to me is that we don’t pay a premium for a plastic case, some hologram, or chip. We’re buying the good name and integrity of the service. If somebody (right or wrong) challenges that, it should be addressed as fervently as counterfeiting because it’s essentially the same thing; a debasing of confidence in the service. Continual, proactive measures are enacted to address counterfeits and the same should be the case for potential conflicts of interests or seeming disparities in the service provided. An independent, external look/audit of practices is already a norm for a publicly traded company with respect to finances. Expanding that to cover potential conflicts of interest within practices at the more tactical level is not going to kill anybody.
We suffer from a lack of defined, measurable standards. Cheers, RickO
I don’t think any TPG can address the concerns listed above. The concerns lack specific verifiable information. It amounts to hearsay.
I think this is a lost cause. Not that any of the grading services are doing anything dishonest, it is that I doubt there is any way to prove in any kind of legal way that this is the same coin that was rejected before, that it was held to some kind of different standard the time it got graded, or that standards change for certain customers.
I think that all this can do is open the door to slander or libel accusations and it is not wise to follow this path openly on a forum. It would be more prudent to contact customer service or an officer of the said grading service with your thoughts.
But I am a legalist or lawyer or anything like that so I am likely all wet
It could have been worked on per say, I have no idea.
there were other great coins from this deal , some that holdered the first time around others did not.
there is a high details grade 1804 Large Cent that I couldn't get into a straight graded holder either , but neither has anybody else because it is still in a net graded holder today as it is still on the market.
But there were some that did: there was an 1855 Half Cent in 65Red with full original red color, it was sold in auction and later upgraded to a Plus and cac'd (I didn't get the upgrade, but it was well deserving I might add however)
there was a 1794 large cent that was absolutely smokin (I know who still owns that, but aint saying)
there was also a real nice 1800 half cent that was graded 63 RedBrown, that had a lot of original red that came from this same collection, although I did not handle this as it was brokered by the coin firm that was helping the family before they contacted me. (ive seen it in auction a few times over the past years)
all these coins were in original envelopes similar to how a lot of the newman coins were stored.
As it should be adressed
This.
The mantra for years has been that grading is an "opinion." Yet when stories like this emerge (I have a couple of my own), we complain about the human factor. I submit that this human factor plays an unavoidable role in some grading decisions, despite what I am sure are conscientious efforts by policy and procedure to minimize it.
May we take this opportunity to dust off the proposal to use AI and scanning for the objective technical analysis of submitted coins?
Here's a warning parable for coin collectors...
I always have to laugh at the "AI fallacy" when someone brings it up. AI is merely highly sophisticated computer programming based on algorithms written by humans. Different humans at different grading services will develop different algorithms. More importantly, those algorithms will change over time as they are "perfected". Throw in variability of whatever cameras, lighting, and other sensors are used to capture data fed to the computer and you end up with a similar situation to what you have now.
Not entirely true. You can remove to the extent possible human variation; surface reflectivity, color, depth of scratches/dings, contrast as a measure of cameo, etc. Another potential is for it to be used as a BS caller when it sees disparities that require a more thorough review. Authentication aid is yet another possibility.
AI may not currently be a 100% solution but it has the potential to be and in the interim may be able to reduce costs enough to allow a more thorough, consistent grading process. To just scoff at it as too complex or problematic because it requires an initial human input to code would mean the whole concept is flawed and that is just laughable.
While I appreciate your argument, I cannot agree with it. The crux of the problem is that computers are fantastic at solving problems where there is an objectively correct answer. Coin grading doesn't remotely fall into that category. Beyond that, every single coin is unique. I doubt even the most expert graders could write a comprehensive set of universally applicable algorithms corresponding to how they assign coin grades.
If all you're going to use computers for is to generate a data sheet with reflectivity, depth of marks, etc. to pass along to a human grader, I'm not sure the added overhead justifies the marginally incremental utility. The human still has the ultimate responsibility of assigning the grade in that scenario.
While I wouldn't rule out computer coin grading at some point in the future, I doubt it'll have real world viability in my lifetime (I'm sixty). YMMV.
>
The components of the “correct answer” can be subdivided into those parts which are subjective and objective or alternatively use objective indicators to support the subjective overall answer. Either way, this isn’t about AI vs graders, it’s about the need for integrity and consistency in the overall process. That’s not just my take.
“As a result, our continued success is heavily dependent on our maintaining that reputation among collectibles dealers and collectors. Failures or errors in authentication or grading processes, such as inconsistent application of grading standards or incidents that put the integrity of those processes into question, could significantly impair our reputation in the marketplace which, in turn, could lead to a loss of customer confidence and a decrease in the demand for our services and, therefore, could have a material adverse effect on our business, financial condition and results of operations.”
Any error, inconsistency, or instance that even appears to undermine the integrity of the process needs to be addressed. I trust it will be.
If there isn't an objectively correct answer, no amount of subdividing will allow arrival at something that doesn't exist. And if all AI promises is arrival at a "subjective overall" answer, what has been gained? Ultimately, humans have to decide how the program will weight the myriad inputs, and humans have to verify (at some stage) that the machine is producing the "correct" answer. And we're right back to square one with the questions of which humans are making those decisions. This doesn't even begin to address potential issues with faulty sensing equipment and other technological challenges that would arise in practice.
I don't disagree that better consistency in grading would be desirable from the collector's point of view. However, lack of (absolute) consistency has been endemic to TPGs since their inception and has obviously not hampered widespread adoption and acceptance of their product. Most collectors who've submitted coins over the years have both benefited from and been disadvantaged by it, by turns.
I'm done on this topic. Cheers.
Yes, humans clearly have to determine what constitutes the objective standard for a particular grade. But humans are certainly capable of said determination, and preparing a computer scan to verify whether that standard exists on a particular coin or not.
Which humans, you say? Well, I guess the same humans who have always been there, i.e., the ones at the right place, at the right time, with the right credibility. As per when the 70-point standard was adopted for the large cents. Serendipity.
The point is that once the objective standard has been applied and programmed in, there it will be, and there it will objectively remain.
Then there well be the immutable standard for grading. Thereafter, all that will remain will be individual tastes.
Here's a warning parable for coin collectors...
If AI became a universal practice and acceptance throughout, who would we chastise if the computer "got it wrong", in our eyes? The programmer, the device itself? There is usually someone/something has to be blamed for what one does not see eye to eye with.
Whatever system is in place, all previous grades/appearances of the coin (at both services) should be linked to the coin's current grade. Without that, you only know one grading event....while the grading "history" of the coin IS the grade. My "favorite" 4 time submission was a $10 Liberty gold cracked out of a TPG MS64 holder. Following 3 submissions were No Grade (PCGS), No Grade (NGC), MS66. All occurred within 9 months. The "price history" of that coin was $2400, $1500 (raw offer - passed), $1500 (raw offer - passed), $5500. I dumped it raw into a major auction to get rid of it and hope someone would see it "had potential." The auction house resubmitted on their own account and got the 66 grade.
I agree that with the human factor involved in grading a coin there will be differences in opinion. The problem I see with his coin others is that when submitted by a major customer the coin all of a sudden gets in a slab. We all know that if we have a customer giving us a lot of business they are going to get some breaks. The little man normally gets the raw end of deal. Not saying favoritism is going on in this case but it does look like it from my seat.
We have all heard and made the excuses about some wear being market acceptable for MS graded coins. Do we really want an AI to tell us that thousands of market acceptable MS coins really aren't? I'm in favor of it, but I'm not paying the bill to create the systems. Do we want to program it to lie to us? If we do, we are in the same place we are now, so whatever we spend doesn't accomplish much. Maybe we could limit AI to doctoring or counterfeiting, but I'm not sure we even agree on that.
Submitter bias?
Market acceptable is highly variable when it comes to old copper and TPGs, as any experienced collector will tell you.
This is surely not a new concept for you, jdimmick, as your eyes are more than experienced.
But I hear you loud and clear, and I'm a bit surprised that it got a sticker.
I'd love to see the coin in question, preferably in hand, but a picture would do as well...
...but I'd trust your eye and opinion more than any of them.