@breakdown said:
I think there are a few categories of coins that fail to sticker and some people get these categories mixed up. C coins should end up in straight grade holders. Coins that are deemed over graded by CAC should also. The third category -
problem coins - get detail holders.
In other words, a C coin is not a problem coin. That’s my understanding anyway.
That’s the way John Albanese has explained it for years. I think I’ll go with his analysis, rather than that of others who predict that most C coins will receive detail grades.
CAC and CACG are using the A,B,C system. They don't mention a 4th category, so where do straight graded PCGS/NGC coins end up when they get details graded by CACG. They had to be "C" coins. There's no other explanation unless there's a secret 4th category that CAC/CACG doesn't publicly acknowledge. It can't be both ways.
@breakdown said:
I think there are a few categories of coins that fail to sticker and some people get these categories mixed up. C coins should end up in straight grade holders. Coins that are deemed over graded by CAC should also. The third category -
problem coins - get detail holders.
In other words, a C coin is not a problem coin. That’s my understanding anyway.
That’s the way John Albanese has explained it for years. I think I’ll go with his analysis, rather than that of others who predict that most C coins will receive detail grades.
CAC and CACG are using the A,B,C system. They don't mention a 4th category, so where do straight graded PCGS/NGC coins end up when they get details graded by CACG. They had to be "C" coins. There's no other explanation unless there's a secret 4th category that CAC/CACG doesn't publicly acknowledge. It can't be both ways.
It’s not very complicated or mysterious. Formerly straight graded NGC and PCGS that failed to sticker at CAC were in their opinion, either:
1) straight-grade worthy but of C quality
2) straight-grade worthy but overgraded (as in deserving of a lower numerical grade)
3) problem coins that should have received detail grades. This category would be your “4th category”.
So coins that receive detail-grades at CACG need not have been C quality coins in the eyes of CAC. They would have fallen into group 3 above, not group 1. Problem coins aren’t a subset of the C quality coin group, as you’ve mistakenly categorized them.
Again, according to John Albanese, most C quality coins will be straight graded by CACG, but at the next numerical grade down with a plus.
We're probably saying the same thing for the most part. My contention is that CAC/CACG are telling us there's only A,B, and C coins, so by default, straight graded problem coins have to be in the "C" category. It is complicated and mysterious in the sense that CAC/CACG don't publicly have a 4th category. If the above info (i.e JA explaining it) is their policy, they should publish it on their website.
Until the human element can be eliminated from grading there will always be inaccuracy.
What about the use of AI in grading? If an algorithm could be developed that would assign a grade, then have a committee of say three to five people evaluate that grade assignment, would that suffice?
Perhaps in the future AI will reach a point that it might be useful, I can especially see it being used for new modern mint releases where every coin is almost identical to the next like the ASE coins. But older coinage is different, it takes more than just counting the number of marks which is what AI can do. Adding a committee to review the AI results just adds back the subjective human element and thus negates using any sort of computer grading.
And in fact there have been several attempts to use computers to grade coins, one was showcased here on this forum just in the last year or two. But so far while interesting experiments those attempts have fallen short and have not been able to achieve any meaningful grading success, at least not yet.
You’ve only seen the beginning, AI in its infancy. Give it 5-10 years and I’d bet that we won’t believe our eyes. When someone integrates depth mapping with a computer that can rotate the coin under different light sources, it’s game over. It will take a lot of data and a lot of money, but it will be done before our lifetime.
The technology will for sure exist but here's the question. Given the Witter Box thread we all acknowledge, happily or not, that our TPG Private Equity overlords make the vast amount of their money from moderns. A human would probably still be required to handle the coins, even moderns.
Why would they spend the money to implement this system? Where is the incremental profit?
If Modern grading could be done by 1 computer with an initial investment of 1-2 million dollars, and replace a large number of humans who need to make at least 60k a year, I believe it could be very profitable long term.
The thing is, if an independent startup comes up with the technology, the big TPGs would either have to buy them or squash them. The first company that introduces AI grading can promote and market it as "The final and accurate grade", so possibly all previously graded moderns and classic coins would have potential for re grading. I expect PCGS or NGC to implement the technology first, considering they have substantial resources to buy out the competition. OR they might decide that it would NOT be profitable long term, and they would buy the entity just to dissolve it and keep the technology from being used.
It's inevitable. It will be done in our life time and a lot sooner than most may anticipate.
As I stated earlier, the way I envision it is AI assigning a grade after analysis of 1 clock to 12 o'clock photos of both obverse and reverse. Then 3 or 5 very senior human graders vote the grade assignment up or down.
This makes the most sense to my thinking.
I don't think the grading portion is really the issue. I'm confident that if coin grading was a priority for AI developers they would have a very workable model by the end of the year.
The actual physical handling of the coins must be done by humans. Under no circumstances is any coin grading firm paying for surgical robots to handle coins. Especially for moderns, which is the overwhelming majority of subs and revenue, the grading is fairly trivial and since a person is handling the coin anyway can't they just as easily say 69 or 70?
And then for the classics, is the AI grading by photos? Who is taking the photos, loading the coin into the device, removing the coin, etc? There are so many other fixed costs besides the graders' salaries to consider. I don't think it's anywhere near as easy as buying some software or a competitor and plugging it in.
EDIT: And this is why we've seen poor customer service, slow turnaround times, bad photos, extra slow servers that are down all the time, etc. The fixed infrastructure can't be removed, so where else to generate extra profit? The top line is nearly maxed out (which is one of the reasons for mystery boxes, etc) so they sharpen their pencil on the "expendables"
Yea of course, the handling will likely continue to be by humans, and the humans would need to feed the machine. But for photos, PCGS system is almost entirely automated by this point, I presume. Which would explain the cookie cutter light placement and poor quality.
But try to think about this from a non-corporate perspective. Everyone in the hobby realizes that human grading will always be subjective, and highly fallible. We all realize the potential of AI, and how it could change the landscape. Several young entrepreneurs have tried to develop an accurate coin grading app/system already. There will be numerous additional attempts to make AI grading functional. If I were to do it, I would build some type of rotating table with lighting, and the computer would analyze a video or live feed of the coin, being tilted and rotated under light. It would be able to measure the luster and contact marks, and with integrated depth mapping and a precise scale, it could determine the level of wear.
I don't expect PCGS or NGC to be the ones that initially develop this technology, I just expect that they're the ones who will end up with it.
If Modern grading could be done by 1 computer with an initial investment of 1-2 million dollars, and replace a large number of humans who need to make at least 60k a year, I believe it could be very profitable long term.
The thing is, if an independent startup comes up with the technology, the big TPGs would either have to buy them or squash them. The first company that introduces AI grading can promote and market it as "The final and accurate grade", so possibly all previously graded moderns and classic coins would have potential for re grading. I expect PCGS or NGC to implement the technology first, considering they have substantial resources to buy out the competition. OR they might decide that it would NOT be profitable long term, and they would buy the entity just to dissolve it and keep the technology from being used.
I don’t disagree with you, Dan. But I hope you’re wrong. There’s something to be said for having fellow humans exercise the nuances of grading a coin. There are also all kinds of ramifications and moral quandaries regarding the ever expanding role of AI. Graders and support personnel are all humans after all, and they need to live and likely have families to raise and support.
I’ll resist until there’s only the option but to use a TPG utilizing AI. I’d rather have the hint of uncertainty and disagreement over a point any day, than know I’m helping put people out of work who are otherwise capable and competent. I’m no communist, I just like humanity.
Having fun while switching things up and focusing on a next level PCGS slabbed 1950+ type set, while still looking for great examples for the 7070.
It's remarkable that a collector just uploaded written ANA grading standards and a couple pics and got a close grade back with a reasoned explanation from AI. It means that the tech probably already exists or will very soon to replace the grading room with your phone and replace the whole concept of TPGS. Imagine being able to identify every individual coin in the world that people put in front of their phones by the tiny individual variations of each coin. Every coin has a fingerprint so to speak. Coins could be instantly imaged/graded with a consistent standard, and it could be done so cheaply that it could be monetized without charging for the service. You could point your cam at a coin, identify, describe in detail, video/image and grade it instantly, see what it's worth in real time, and see who in the world wants to buy or sell it. It could replace TPGS, auctions and shows. Everyone could verify title on every individual coin so stolen coins wouldn't be as valuable. Those are just my initial thoughts after 5 minutes, there's got to be a million other applications for AI in coins.
If Modern grading could be done by 1 computer with an initial investment of 1-2 million dollars, and replace a large number of humans who need to make at least 60k a year, I believe it could be very profitable long term.
The thing is, if an independent startup comes up with the technology, the big TPGs would either have to buy them or squash them. The first company that introduces AI grading can promote and market it as "The final and accurate grade", so possibly all previously graded moderns and classic coins would have potential for re grading. I expect PCGS or NGC to implement the technology first, considering they have substantial resources to buy out the competition. OR they might decide that it would NOT be profitable long term, and they would buy the entity just to dissolve it and keep the technology from being used.
I don’t disagree with you, Dan. But I hope you’re wrong. There’s something to be said for having fellow humans exercise the nuances of grading a coin. There are also all kinds of ramifications and moral quandaries regarding the ever expanding role of AI. Graders and support personnel are all humans after all, and they need to live and likely have families to raise and support.
I’ll resist until there’s only the option but to use a TPG utilizing AI. I’d rather have the hint of uncertainty and disagreement over a point any day, than know I’m helping put people out of work who are otherwise capable and competent. I’m no communist, I just like humanity.
What if the AI just assigned the initial grade, but human finalizers/QA specialists were still part of the assembly line? I think AI has the potential to level the playing field even more than TPGs and CAC, and would ultimately benefit the hobby in a positive way in the long term. In the short term, as a coin dealer with selfish motives, I'd rather have some inconsistency so I can use arbitrage and exploit the weaknesses of human grading, which is an inherently flawed system.
If Modern grading could be done by 1 computer with an initial investment of 1-2 million dollars, and replace a large number of humans who need to make at least 60k a year, I believe it could be very profitable long term.
The thing is, if an independent startup comes up with the technology, the big TPGs would either have to buy them or squash them. The first company that introduces AI grading can promote and market it as "The final and accurate grade", so possibly all previously graded moderns and classic coins would have potential for re grading. I expect PCGS or NGC to implement the technology first, considering they have substantial resources to buy out the competition. OR they might decide that it would NOT be profitable long term, and they would buy the entity just to dissolve it and keep the technology from being used.
I don’t disagree with you, Dan. But I hope you’re wrong. There’s something to be said for having fellow humans exercise the nuances of grading a coin. There are also all kinds of ramifications and moral quandaries regarding the ever expanding role of AI. Graders and support personnel are all humans after all, and they need to live and likely have families to raise and support.
I’ll resist until there’s only the option but to use a TPG utilizing AI. I’d rather have the hint of uncertainty and disagreement over a point any day, than know I’m helping put people out of work who are otherwise capable and competent. I’m no communist, I just like humanity.
What if the AI just assigned the initial grade, but human finalizers/QA specialists were still part of the assembly line? I think AI has the potential to level the playing field even more than TPGs and CAC, and would ultimately benefit the hobby in a positive way in the long term. In the short term, as a coin dealer with selfish motives, I'd rather have some inconsistency so I can use arbitrage and exploit the weaknesses of human grading, which is an inherently flawed system.
This is a compromise I could certainly live with. I just believe after a certain point of AI refinement a company, whose goal is to rightly seek profit, would feel the reviewers are no longer necessary. Kind of a fine line that’s going to have to be reconciled where the free market is going to have to come to terms with humanity. Certainly some interesting times ahead….
Having fun while switching things up and focusing on a next level PCGS slabbed 1950+ type set, while still looking for great examples for the 7070.
Until the human element can be eliminated from grading there will always be inaccuracy.
What about the use of AI in grading? If an algorithm could be developed that would assign a grade, then have a committee of say three to five people evaluate that grade assignment, would that suffice?
Perhaps in the future AI will reach a point that it might be useful, I can especially see it being used for new modern mint releases where every coin is almost identical to the next like the ASE coins. But older coinage is different, it takes more than just counting the number of marks which is what AI can do. Adding a committee to review the AI results just adds back the subjective human element and thus negates using any sort of computer grading.
And in fact there have been several attempts to use computers to grade coins, one was showcased here on this forum just in the last year or two. But so far while interesting experiments those attempts have fallen short and have not been able to achieve any meaningful grading success, at least not yet.
You’ve only seen the beginning, AI in its infancy. Give it 5-10 years and I’d bet that we won’t believe our eyes. When someone integrates depth mapping with a computer that can rotate the coin under different light sources, it’s game over. It will take a lot of data and a lot of money, but it will be done before our lifetime.
The technology will for sure exist but here's the question. Given the Witter Box thread we all acknowledge, happily or not, that our TPG Private Equity overlords make the vast amount of their money from moderns. A human would probably still be required to handle the coins, even moderns.
Why would they spend the money to implement this system? Where is the incremental profit?
If Modern grading could be done by 1 computer with an initial investment of 1-2 million dollars, and replace a large number of humans who need to make at least 60k a year, I believe it could be very profitable long term.
The thing is, if an independent startup comes up with the technology, the big TPGs would either have to buy them or squash them. The first company that introduces AI grading can promote and market it as "The final and accurate grade", so possibly all previously graded moderns and classic coins would have potential for re grading. I expect PCGS or NGC to implement the technology first, considering they have substantial resources to buy out the competition. OR they might decide that it would NOT be profitable long term, and they would buy the entity just to dissolve it and keep the technology from being used.
It's inevitable. It will be done in our life time and a lot sooner than most may anticipate.
As I stated earlier, the way I envision it is AI assigning a grade after analysis of 1 clock to 12 o'clock photos of both obverse and reverse. Then 3 or 5 very senior human graders vote the grade assignment up or down.
This makes the most sense to my thinking.
I don't think the grading portion is really the issue. I'm confident that if coin grading was a priority for AI developers they would have a very workable model by the end of the year.
The actual physical handling of the coins must be done by humans. Under no circumstances is any coin grading firm paying for surgical robots to handle coins. Especially for moderns, which is the overwhelming majority of subs and revenue, the grading is fairly trivial and since a person is handling the coin anyway can't they just as easily say 69 or 70?
And then for the classics, is the AI grading by photos? Who is taking the photos, loading the coin into the device, removing the coin, etc? There are so many other fixed costs besides the graders' salaries to consider. I don't think it's anywhere near as easy as buying some software or a competitor and plugging it in.
EDIT: And this is why we've seen poor customer service, slow turnaround times, bad photos, extra slow servers that are down all the time, etc. The fixed infrastructure can't be removed, so where else to generate extra profit? The top line is nearly maxed out (which is one of the reasons for mystery boxes, etc) so they sharpen their pencil on the "expendables"
Yea of course, the handling will likely continue to be by humans, and the humans would need to feed the machine. But for photos, PCGS system is almost entirely automated by this point, I presume. Which would explain the cookie cutter light placement and poor quality.
But try to think about this from a non-corporate perspective. Everyone in the hobby realizes that human grading will always be subjective, and highly fallible. We all realize the potential of AI, and how it could change the landscape. Several young entrepreneurs have tried to develop an accurate coin grading app/system already. There will be numerous additional attempts to make AI grading functional. If I were to do it, I would build some type of rotating table with lighting, and the computer would analyze a video or live feed of the coin, being tilted and rotated under light. It would be able to measure the luster and contact marks, and with integrated depth mapping and a precise scale, it could determine the level of wear.
I don't expect PCGS or NGC to be the ones that initially develop this technology, I just expect that they're the ones who will end up with it.
They're private companies so we don't have access to their financials so everything here is a supposition. But I think that one of the key factors here is the difference between classics and moderns. We are of course predominantly interested in classics because that's what we like and it's almost completely what we focus on. But we're also the minority of the collecting community. Just like this forum is not an accurate representation of the entire population of collectors.
But there's a huge population of people who don't know or care about the grading or company. I'm not sure how much you've bumped into them in the shows yet but they're there, that's why those TV hucksters have done so well for so long.
I think you're absolutely right about the technology and implementation. I just don't think there is enough money in grading/reholdering/crossing classic coins for such a large capital infusion from the private equity companies that own P & N.
@breakdown said:
I think there are a few categories of coins that fail to sticker and some people get these categories mixed up. C coins should end up in straight grade holders. Coins that are deemed over graded by CAC should also. The third category -
problem coins - get detail holders.
In other words, a C coin is not a problem coin. That’s my understanding anyway.
That’s the way John Albanese has explained it for years. I think I’ll go with his analysis, rather than that of others who predict that most C coins will receive detail grades.
CAC and CACG are using the A,B,C system. They don't mention a 4th category, so where do straight graded PCGS/NGC coins end up when they get details graded by CACG. They had to be "C" coins. There's no other explanation unless there's a secret 4th category that CAC/CACG doesn't publicly acknowledge. It can't be both ways.
@breakdown said:
I think there are a few categories of coins that fail to sticker and some people get these categories mixed up. C coins should end up in straight grade holders. Coins that are deemed over graded by CAC should also. The third category -
problem coins - get detail holders.
In other words, a C coin is not a problem coin. That’s my understanding anyway.
That’s the way John Albanese has explained it for years. I think I’ll go with his analysis, rather than that of others who predict that most C coins will receive detail grades.
CAC and CACG are using the A,B,C system. They don't mention a 4th category, so where do straight graded PCGS/NGC coins end up when they get details graded by CACG. They had to be "C" coins. There's no other explanation unless there's a secret 4th category that CAC/CACG doesn't publicly acknowledge. It can't be both ways.
It’s not very complicated or mysterious. Formerly straight graded NGC and PCGS that failed to sticker at CAC were in their opinion, either:
1) straight-grade worthy but of C quality
2) straight-grade worthy but overgraded (as in deserving of a lower numerical grade)
3) problem coins that should have received detail grades. This category would be your “4th category”.
So coins that receive detail-grades at CACG need not have been C quality coins in the eyes of CAC. They would have fallen into group 3 above, not group 1. Problem coins aren’t a subset of the C quality coin group, as you’ve mistakenly categorized them.
Again, according to John Albanese, most C quality coins will be straight graded by CACG, but at the next numerical grade down with a plus.
We're probably saying the same thing for the most part. My contention is that CAC/CACG are telling us there's only A,B, and C coins, so by default, straight graded problem coins have to be in the "C" category. It is complicated and mysterious in the sense that CAC/CACG don't publicly have a 4th category. If the above info (i.e JA explaining it) is their policy, they should publish it on their website.
They never said there were only A,B, and C coins. They said there were those three categories (all straight grade worthy) AND problem/detail grade coins, as well.
They also said they don’t want C coins in their holders, so most C coins will be lowered a grade and given a plus.
Coins in straight grade NGC and PCGS holders that CAC/CACG see as detail grade/problem coins AREN’T C coins.
None of this is mysterious or new and it’s been discussed openly for a long time.
Mark Feld* of Heritage Auctions*Unless otherwise noted, my posts here represent my personal opinions.
@tcollects said:
It's remarkable that a collector just uploaded written ANA grading standards and a couple pics and got a close grade back with a reasoned explanation from AI. It means that the tech probably already exists or will very soon to replace the grading room with your phone and replace the whole concept of TPGS. Imagine being able to identify every individual coin in the world that people put in front of their phones by the tiny individual variations of each coin. Every coin has a fingerprint so to speak. Coins could be instantly imaged/graded with a consistent standard, and it could be done so cheaply that it could be monetized without charging for the service. You could point your cam at a coin, identify, describe in detail, video/image and grade it instantly, see what it's worth in real time, and see who in the world wants to buy or sell it. It could replace TPGS, auctions and shows. Everyone could verify title on every individual coin so stolen coins wouldn't be as valuable. Those are just my initial thoughts after 5 minutes, there's got to be a million other applications for AI in coins.
Imagine a system where you place the coin in a receptacle, a rotating camera takes thousands of images under multiple magnifications, and AI renders a verdict- all in less that 30 seconds. Again, then have a three to five CACG level graders pass it for encapsulation.
No emotion. No bias. No vested interest in a particular outcome. No financial interest. Just cold, hard, analytical determinations.
@breakdown said:
I think there are a few categories of coins that fail to sticker and some people get these categories mixed up. C coins should end up in straight grade holders. Coins that are deemed over graded by CAC should also. The third category -
problem coins - get detail holders.
In other words, a C coin is not a problem coin. That’s my understanding anyway.
That’s the way John Albanese has explained it for years. I think I’ll go with his analysis, rather than that of others who predict that most C coins will receive detail grades.
CAC and CACG are using the A,B,C system. They don't mention a 4th category, so where do straight graded PCGS/NGC coins end up when they get details graded by CACG. They had to be "C" coins. There's no other explanation unless there's a secret 4th category that CAC/CACG doesn't publicly acknowledge. It can't be both ways.
@breakdown said:
I think there are a few categories of coins that fail to sticker and some people get these categories mixed up. C coins should end up in straight grade holders. Coins that are deemed over graded by CAC should also. The third category -
problem coins - get detail holders.
In other words, a C coin is not a problem coin. That’s my understanding anyway.
That’s the way John Albanese has explained it for years. I think I’ll go with his analysis, rather than that of others who predict that most C coins will receive detail grades.
CAC and CACG are using the A,B,C system. They don't mention a 4th category, so where do straight graded PCGS/NGC coins end up when they get details graded by CACG. They had to be "C" coins. There's no other explanation unless there's a secret 4th category that CAC/CACG doesn't publicly acknowledge. It can't be both ways.
It’s not very complicated or mysterious. Formerly straight graded NGC and PCGS that failed to sticker at CAC were in their opinion, either:
1) straight-grade worthy but of C quality
2) straight-grade worthy but overgraded (as in deserving of a lower numerical grade)
3) problem coins that should have received detail grades. This category would be your “4th category”.
So coins that receive detail-grades at CACG need not have been C quality coins in the eyes of CAC. They would have fallen into group 3 above, not group 1. Problem coins aren’t a subset of the C quality coin group, as you’ve mistakenly categorized them.
Again, according to John Albanese, most C quality coins will be straight graded by CACG, but at the next numerical grade down with a plus.
We're probably saying the same thing for the most part. My contention is that CAC/CACG are telling us there's only A,B, and C coins, so by default, straight graded problem coins have to be in the "C" category. It is complicated and mysterious in the sense that CAC/CACG don't publicly have a 4th category. If the above info (i.e JA explaining it) is their policy, they should publish it on their website.
They never said there were only A,B, and C coins. They said there were those three categories (all straight grade worthy) AND problem/detail grade coins, as well.
They also said they don’t want C coins in their holders, so most C coins will be lowered a grade and given a plus.
Coins in straight grade NGC and PCGS holders that CAC/CACG see as detail grade/problem coins AREN’T C coins.
None of this is mysterious or new and it’s been discussed openly for a long time.
At this point, we're either nit picking, or talking past each other to some degree. That's ok, I'll continue a bit longer because we're having a technical conversation and technicalities matter.
First, I'm talking about coins already straight graded PCGS/NGC. If you're talking about raw coins submitted to CACG, then ok. My point is about straight graded PCGS/NGC coins.
CAC absolutely is saying they only have 3 categories, A, B, and C in regard to straight graded PCGS/NGC coins.
There's no mention of a "details" category or that some/many "C" coins should be details graded.
That being said, many "C" coins will receive a details grade from CAC, not necessarily a downgrade. We know this for a fact based on submissions posted here in the forum.
So again, my contention is that CACG will details grade many straight graded PCGS/NGC coins that the CAC stickering service regards as "C" coins in straight graded PCGS/NGC holders. If they have a mysterious "D" category for straight graded PCGS/NGC coins that should be in a details holder, then they need to say that on their web site. Since they don't, it's logical to assume that those coins are a subset of "C" coins.
I think my position is logically consistent with CAC website information and actual CACG grading practices.
@breakdown said:
I think there are a few categories of coins that fail to sticker and some people get these categories mixed up. C coins should end up in straight grade holders. Coins that are deemed over graded by CAC should also. The third category -
problem coins - get detail holders.
In other words, a C coin is not a problem coin. That’s my understanding anyway.
That’s the way John Albanese has explained it for years. I think I’ll go with his analysis, rather than that of others who predict that most C coins will receive detail grades.
CAC and CACG are using the A,B,C system. They don't mention a 4th category, so where do straight graded PCGS/NGC coins end up when they get details graded by CACG. They had to be "C" coins. There's no other explanation unless there's a secret 4th category that CAC/CACG doesn't publicly acknowledge. It can't be both ways.
@breakdown said:
I think there are a few categories of coins that fail to sticker and some people get these categories mixed up. C coins should end up in straight grade holders. Coins that are deemed over graded by CAC should also. The third category -
problem coins - get detail holders.
In other words, a C coin is not a problem coin. That’s my understanding anyway.
That’s the way John Albanese has explained it for years. I think I’ll go with his analysis, rather than that of others who predict that most C coins will receive detail grades.
CAC and CACG are using the A,B,C system. They don't mention a 4th category, so where do straight graded PCGS/NGC coins end up when they get details graded by CACG. They had to be "C" coins. There's no other explanation unless there's a secret 4th category that CAC/CACG doesn't publicly acknowledge. It can't be both ways.
It’s not very complicated or mysterious. Formerly straight graded NGC and PCGS that failed to sticker at CAC were in their opinion, either:
1) straight-grade worthy but of C quality
2) straight-grade worthy but overgraded (as in deserving of a lower numerical grade)
3) problem coins that should have received detail grades. This category would be your “4th category”.
So coins that receive detail-grades at CACG need not have been C quality coins in the eyes of CAC. They would have fallen into group 3 above, not group 1. Problem coins aren’t a subset of the C quality coin group, as you’ve mistakenly categorized them.
Again, according to John Albanese, most C quality coins will be straight graded by CACG, but at the next numerical grade down with a plus.
We're probably saying the same thing for the most part. My contention is that CAC/CACG are telling us there's only A,B, and C coins, so by default, straight graded problem coins have to be in the "C" category. It is complicated and mysterious in the sense that CAC/CACG don't publicly have a 4th category. If the above info (i.e JA explaining it) is their policy, they should publish it on their website.
They never said there were only A,B, and C coins. They said there were those three categories (all straight grade worthy) AND problem/detail grade coins, as well.
They also said they don’t want C coins in their holders, so most C coins will be lowered a grade and given a plus.
Coins in straight grade NGC and PCGS holders that CAC/CACG see as detail grade/problem coins AREN’T C coins.
None of this is mysterious or new and it’s been discussed openly for a long time.
At this point, we're either nit picking, or talking past each other to some degree. That's ok, I'll continue a bit longer because we're having a technical conversation and technicalities matter.
First, I'm talking about coins already straight graded PCGS/NGC. If you're talking about raw coins submitted to CACG, then ok. My point is about straight graded PCGS/NGC coins.
CAC absolutely is saying they only have 3 categories, A, B, and C in regard to straight graded PCGS/NGC coins.
There's no mention of a "details" category or that some/many "C" coins should be details graded.
That being said, many "C" coins will receive a details grade from CAC, not necessarily a downgrade. We know this for a fact based on submissions posted here in the forum.
So again, my contention is that CACG will details grade many straight graded PCGS/NGC coins that the CAC stickering service regards as "C" coins in straight graded PCGS/NGC holders. If they have a mysterious "D" category for straight graded PCGS/NGC coins that should be in a details holder, then they need to say that on their web site. Since they don't, it's logical to assume that those coins are a subset of "C" coins.
I think my position is logically consistent with CAC website information and actual CACG grading practices.
1) “CAC absolutely is saying they only have 3 categories, A, B, and C in regard to straight graded PCGS/NGC coins.”
I never said otherwise. And in fact, I’ve repeated it multiple times.
2) “There's no mention of a "details" category or that some/many "C" coins should be details graded.”
I see no reason why a grading company would feel any need to mention that problem coins would receive details, rather than straight grades. That seems quite obvious.
How does one know how many C coins are receiving detail grades? As far as I know, C coins aren’t marked with a “C” or other clue in order to categorize them as such.
3) “That being said, many "C" coins will receive a details grade from CAC, not necessarily a downgrade. We know this for a fact based on submissions posted here in the forum.”
How do we know this for a fact? Because a submitter said so and based on images, it looks like he’s correct?
And again, there’s no “mysterious "D" category for straight graded PCGS/NGC coins that should be in a details holder”. Coins in straight grade holders that CAC/CACG deems problem/detail grade coins receive detail grades. I’m not talking about straight grade C coins.
I give up trying to explain things that have been posted numerous times since before CACG even started taking submissions. Disagree as much as you like.
Mark Feld* of Heritage Auctions*Unless otherwise noted, my posts here represent my personal opinions.
Sorry @Manifest_Destiny, you're one of my favorite posters here but I agree with Mark. Though it may not be explicitly stated on the website, JA has stated on numerous occasions that many coins that fail to sticker are technically not "C" coins. Essentially, there are "D", "E", and "F" coins but JA has been very delicate with his phrasing over the years, in an effort to avoid disparaging the other services. I would say that a minority of CAC rejects are C coins, the rest being problem coins or coins that are severely overgraded.
Until the human element can be eliminated from grading there will always be inaccuracy.
What about the use of AI in grading? If an algorithm could be developed that would assign a grade, then have a committee of say three to five people evaluate that grade assignment, would that suffice?
Perhaps in the future AI will reach a point that it might be useful, I can especially see it being used for new modern mint releases where every coin is almost identical to the next like the ASE coins. But older coinage is different, it takes more than just counting the number of marks which is what AI can do. Adding a committee to review the AI results just adds back the subjective human element and thus negates using any sort of computer grading.
And in fact there have been several attempts to use computers to grade coins, one was showcased here on this forum just in the last year or two. But so far while interesting experiments those attempts have fallen short and have not been able to achieve any meaningful grading success, at least not yet.
You’ve only seen the beginning, AI in its infancy. Give it 5-10 years and I’d bet that we won’t believe our eyes. When someone integrates depth mapping with a computer that can rotate the coin under different light sources, it’s game over. It will take a lot of data and a lot of money, but it will be done before our lifetime.
The technology will for sure exist but here's the question. Given the Witter Box thread we all acknowledge, happily or not, that our TPG Private Equity overlords make the vast amount of their money from moderns. A human would probably still be required to handle the coins, even moderns.
Why would they spend the money to implement this system? Where is the incremental profit?
If Modern grading could be done by 1 computer with an initial investment of 1-2 million dollars, and replace a large number of humans who need to make at least 60k a year, I believe it could be very profitable long term.
The thing is, if an independent startup comes up with the technology, the big TPGs would either have to buy them or squash them. The first company that introduces AI grading can promote and market it as "The final and accurate grade", so possibly all previously graded moderns and classic coins would have potential for re grading. I expect PCGS or NGC to implement the technology first, considering they have substantial resources to buy out the competition. OR they might decide that it would NOT be profitable long term, and they would buy the entity just to dissolve it and keep the technology from being used.
It's inevitable. It will be done in our life time and a lot sooner than most may anticipate.
As I stated earlier, the way I envision it is AI assigning a grade after analysis of 1 clock to 12 o'clock photos of both obverse and reverse. Then 3 or 5 very senior human graders vote the grade assignment up or down.
This makes the most sense to my thinking.
I don't think the grading portion is really the issue. I'm confident that if coin grading was a priority for AI developers they would have a very workable model by the end of the year.
The actual physical handling of the coins must be done by humans. Under no circumstances is any coin grading firm paying for surgical robots to handle coins. Especially for moderns, which is the overwhelming majority of subs and revenue, the grading is fairly trivial and since a person is handling the coin anyway can't they just as easily say 69 or 70?
And then for the classics, is the AI grading by photos? Who is taking the photos, loading the coin into the device, removing the coin, etc? There are so many other fixed costs besides the graders' salaries to consider. I don't think it's anywhere near as easy as buying some software or a competitor and plugging it in.
EDIT: And this is why we've seen poor customer service, slow turnaround times, bad photos, extra slow servers that are down all the time, etc. The fixed infrastructure can't be removed, so where else to generate extra profit? The top line is nearly maxed out (which is one of the reasons for mystery boxes, etc) so they sharpen their pencil on the "expendables"
Yea of course, the handling will likely continue to be by humans, and the humans would need to feed the machine. But for photos, PCGS system is almost entirely automated by this point, I presume. Which would explain the cookie cutter light placement and poor quality.
But try to think about this from a non-corporate perspective. Everyone in the hobby realizes that human grading will always be subjective, and highly fallible. We all realize the potential of AI, and how it could change the landscape. Several young entrepreneurs have tried to develop an accurate coin grading app/system already. There will be numerous additional attempts to make AI grading functional. If I were to do it, I would build some type of rotating table with lighting, and the computer would analyze a video or live feed of the coin, being tilted and rotated under light. It would be able to measure the luster and contact marks, and with integrated depth mapping and a precise scale, it could determine the level of wear.
I don't expect PCGS or NGC to be the ones that initially develop this technology, I just expect that they're the ones who will end up with it.
Fully agree with everything you are saying re: ai with the exception that using a scale to determine wear won’t be accurate. There is too much variation in weight from the mint for at least classic series.
I have been in some brainstorming sessions with people regarding developing an AI grading service. I am a founder of a tech startup and as I plan on exiting in the next few years, I’m trying to figure out what venture I want to explore next.
I think AI grading is a huge opportunity - like you said, if a service can be developed that provided hyper accurate (or more importantly to the collector, consistent) grading, then market forces will eventually push all existing coins (classic, at least) to be graded by this service. Especially if it can be done quickly and affordably (something like $20 with a one week turnaround).
Building the physical system itself is simple, all you need is a high quality camera, good lights, and a rotating base to simulate what a human does when analyzing luster.
The software part is what would be the trickiest because there would be tons of data needed to train the AI on all of the nuances of grading. Is it hairlines or polish lines? Is it die crack or PMD? Is it wear or a weak strike? Etc etc
In fact in our discussions the biggest barrier to doing this is not necessarily the cost of the physical system or the AI engineers to code it, but the huge amounts of $$ that would need to be spent in order to gather enough examples to train the AI. Sure they could be resold, but it would suck up capital and there is value risk since they would have to be cracked out for the most accurate training data.
If Modern grading could be done by 1 computer with an initial investment of 1-2 million dollars, and replace a large number of humans who need to make at least 60k a year, I believe it could be very profitable long term.
The thing is, if an independent startup comes up with the technology, the big TPGs would either have to buy them or squash them. The first company that introduces AI grading can promote and market it as "The final and accurate grade", so possibly all previously graded moderns and classic coins would have potential for re grading. I expect PCGS or NGC to implement the technology first, considering they have substantial resources to buy out the competition. OR they might decide that it would NOT be profitable long term, and they would buy the entity just to dissolve it and keep the technology from being used.
I don’t disagree with you, Dan. But I hope you’re wrong. There’s something to be said for having fellow humans exercise the nuances of grading a coin. There are also all kinds of ramifications and moral quandaries regarding the ever expanding role of AI. Graders and support personnel are all humans after all, and they need to live and likely have families to raise and support.
I’ll resist until there’s only the option but to use a TPG utilizing AI. I’d rather have the hint of uncertainty and disagreement over a point any day, than know I’m helping put people out of work who are otherwise capable and competent. I’m no communist, I just like humanity.
I understand the sentiment but you can make the same argument regarding people losing jobs for every single technological advance humans have made. Society will adapt with AI just as it did with the internet and with the Industrial Revolution and everything that happened before.
The thing I don’t really agree with is grading accurately graded “C” coins a grade lower with a +.
If the “C” coin is an accurately graded coin in Fine condition, I don’t get grading it VG+. Why not grade it Fine and grade the “A” and “B” coins F+? Otherwise you are changing the grading standards or is that the ultimate goal?
@breakdown said:
I think there are a few categories of coins that fail to sticker and some people get these categories mixed up. C coins should end up in straight grade holders. Coins that are deemed over graded by CAC should also. The third category -
problem coins - get detail holders.
In other words, a C coin is not a problem coin. That’s my understanding anyway.
That’s the way John Albanese has explained it for years. I think I’ll go with his analysis, rather than that of others who predict that most C coins will receive detail grades.
CAC and CACG are using the A,B,C system. They don't mention a 4th category, so where do straight graded PCGS/NGC coins end up when they get details graded by CACG. They had to be "C" coins. There's no other explanation unless there's a secret 4th category that CAC/CACG doesn't publicly acknowledge. It can't be both ways.
@breakdown said:
I think there are a few categories of coins that fail to sticker and some people get these categories mixed up. C coins should end up in straight grade holders. Coins that are deemed over graded by CAC should also. The third category -
problem coins - get detail holders.
In other words, a C coin is not a problem coin. That’s my understanding anyway.
That’s the way John Albanese has explained it for years. I think I’ll go with his analysis, rather than that of others who predict that most C coins will receive detail grades.
CAC and CACG are using the A,B,C system. They don't mention a 4th category, so where do straight graded PCGS/NGC coins end up when they get details graded by CACG. They had to be "C" coins. There's no other explanation unless there's a secret 4th category that CAC/CACG doesn't publicly acknowledge. It can't be both ways.
It’s not very complicated or mysterious. Formerly straight graded NGC and PCGS that failed to sticker at CAC were in their opinion, either:
1) straight-grade worthy but of C quality
2) straight-grade worthy but overgraded (as in deserving of a lower numerical grade)
3) problem coins that should have received detail grades. This category would be your “4th category”.
So coins that receive detail-grades at CACG need not have been C quality coins in the eyes of CAC. They would have fallen into group 3 above, not group 1. Problem coins aren’t a subset of the C quality coin group, as you’ve mistakenly categorized them.
Again, according to John Albanese, most C quality coins will be straight graded by CACG, but at the next numerical grade down with a plus.
We're probably saying the same thing for the most part. My contention is that CAC/CACG are telling us there's only A,B, and C coins, so by default, straight graded problem coins have to be in the "C" category. It is complicated and mysterious in the sense that CAC/CACG don't publicly have a 4th category. If the above info (i.e JA explaining it) is their policy, they should publish it on their website.
No, by default straight graded problem coins do not have to be in the "C" category. Problem coins span the whole ABC spectrum. For example, there are plenty of coins on all parts of the grading (A, B, and C) spectrum with PVC. An MS65+ with PVC is not a "C" coin. CAC will slab the coin in an MS Details slab. Successfully remove the PVC and it theoretically will cross into a CAC 65 or 65+ slab.
@tcollects said:
It's remarkable that a collector just uploaded written ANA grading standards and a couple pics and got a close grade back with a reasoned explanation from AI. It means that the tech probably already exists or will very soon to replace the grading room with your phone and replace the whole concept of TPGS. Imagine being able to identify every individual coin in the world that people put in front of their phones by the tiny individual variations of each coin. Every coin has a fingerprint so to speak. Coins could be instantly imaged/graded with a consistent standard, and it could be done so cheaply that it could be monetized without charging for the service. You could point your cam at a coin, identify, describe in detail, video/image and grade it instantly, see what it's worth in real time, and see who in the world wants to buy or sell it. It could replace TPGS, auctions and shows. Everyone could verify title on every individual coin so stolen coins wouldn't be as valuable. Those are just my initial thoughts after 5 minutes, there's got to be a million other applications for AI in coins.
Imagine a system where you place the coin in a receptacle, a rotating camera takes thousands of images under multiple magnifications, and AI renders a verdict- all in less that 30 seconds. Again, then have a three to five CACG level graders pass it for encapsulation.
No emotion. No bias. No vested interest in a particular outcome. No financial interest. Just cold, hard, analytical determinations.
And dealers and collectors will still vehemently disagree with the determinations.
For anyone who's dealt with automation, 80% of the job is typically straightforward and easy. It's the last 20% that's the bugger. Getting a car to drive itself down a road and stop at a red light or stop sign or not hit a person is easy. The hard part is programming the car to wait in line at a school to pick up a kid or to find a parking space in a parking garage. That is the problem with grading and to make the problem worse is that all dealers or collectors will assume that their coin is in the small percentage that AI can't accurately grade and use that to insist that AI got the coin wrong and you should pay more because the coin is better than the machine says it is. So are we really gaining anything?
And for those who suggest that a machine grades a coin and then humans finalize it... what's the point? Might as well just have the humans grade the coin too.
@breakdown said:
I think there are a few categories of coins that fail to sticker and some people get these categories mixed up. C coins should end up in straight grade holders. Coins that are deemed over graded by CAC should also. The third category -
problem coins - get detail holders.
In other words, a C coin is not a problem coin. That’s my understanding anyway.
That’s the way John Albanese has explained it for years. I think I’ll go with his analysis, rather than that of others who predict that most C coins will receive detail grades.
CAC and CACG are using the A,B,C system. They don't mention a 4th category, so where do straight graded PCGS/NGC coins end up when they get details graded by CACG. They had to be "C" coins. There's no other explanation unless there's a secret 4th category that CAC/CACG doesn't publicly acknowledge. It can't be both ways.
@breakdown said:
I think there are a few categories of coins that fail to sticker and some people get these categories mixed up. C coins should end up in straight grade holders. Coins that are deemed over graded by CAC should also. The third category -
problem coins - get detail holders.
In other words, a C coin is not a problem coin. That’s my understanding anyway.
That’s the way John Albanese has explained it for years. I think I’ll go with his analysis, rather than that of others who predict that most C coins will receive detail grades.
CAC and CACG are using the A,B,C system. They don't mention a 4th category, so where do straight graded PCGS/NGC coins end up when they get details graded by CACG. They had to be "C" coins. There's no other explanation unless there's a secret 4th category that CAC/CACG doesn't publicly acknowledge. It can't be both ways.
It’s not very complicated or mysterious. Formerly straight graded NGC and PCGS that failed to sticker at CAC were in their opinion, either:
1) straight-grade worthy but of C quality
2) straight-grade worthy but overgraded (as in deserving of a lower numerical grade)
3) problem coins that should have received detail grades. This category would be your “4th category”.
So coins that receive detail-grades at CACG need not have been C quality coins in the eyes of CAC. They would have fallen into group 3 above, not group 1. Problem coins aren’t a subset of the C quality coin group, as you’ve mistakenly categorized them.
Again, according to John Albanese, most C quality coins will be straight graded by CACG, but at the next numerical grade down with a plus.
We're probably saying the same thing for the most part. My contention is that CAC/CACG are telling us there's only A,B, and C coins, so by default, straight graded problem coins have to be in the "C" category. It is complicated and mysterious in the sense that CAC/CACG don't publicly have a 4th category. If the above info (i.e JA explaining it) is their policy, they should publish it on their website.
No, by default straight graded problem coins do not have to be in the "C" category. Problem coins span the whole ABC spectrum. For example, there are plenty of coins on all parts of the grading (A, B, and C) spectrum with PVC. An MS65+ with PVC is not a "C" coin. CAC will slab the coin in an MS Details slab. Successfully remove the PVC and it theoretically will cross into a CAC 65 or 65+ slab.
That doesn't make sense if CAC stickers A and B coins. I doubt they're stickering A coins with PVC.
@DeplorableDan said:
Sorry @Manifest_Destiny, you're one of my favorite posters here but I agree with Mark. Though it may not be explicitly stated on the website, JA has stated on numerous occasions that many coins that fail to sticker are technically not "C" coins. Essentially, there are "D", "E", and "F" coins but JA has been very delicate with his phrasing over the years, in an effort to avoid disparaging the other services. I would say that a minority of CAC rejects are C coins, the rest being problem coins or coins that are severely overgraded.
That's a valid explanation, but then we're back to my point that if they aren't C coins, then CACG has one (or more) categories that aren't publicly stated.
I feel like we're all essentially saying the same thing. Either there are details C coins, or there are other unstated categories (D,E, etc.) as you're pointing out above.
I feel like the counter argument to my point is: "There are no details C coins" AND "CAC only has A,B,C categories". My point is that both can't be true. CACG either grades C coins with details, or they have more (unstated) categories beyond A,B, and C which they don't publicly state.
I think a coin with obvious PVC green gunk worse category than C lol. What shall it be called - ruined, cull, no go, destroyed, not marketable? What pct can even be salvaged?
In reality / Not all of them can be salvaged. It may have been an A coin b4 the pvc,etc but once it’s got pvc what the heck does it matter what it was b4?
anything I am keeping for myself goes to PCGS (that would higher dollar stuff, better stuff)
for the business:
Nicer early type, better date seated, bust material, better mid grade & up copper and higher better date dollars PCGS.
Generic gold ,lower to mid grade key dates, Modern to NGC (except like a 95-W that I decided to keep for myself as it was darn nice sent to PCGS.
World coins NGC by far.
Paper Money pretty much 95% PMG (I have used PCGS Banknote on some , but only as a convience becuase they were at a show I was attending .
I like CAC stickering, but have not sent anything to the grading service. From what I have seen I like their grading, but i really am not fond of the holders.
@DeplorableDan said:
Sorry @Manifest_Destiny, you're one of my favorite posters here but I agree with Mark. Though it may not be explicitly stated on the website, JA has stated on numerous occasions that many coins that fail to sticker are technically not "C" coins. Essentially, there are "D", "E", and "F" coins but JA has been very delicate with his phrasing over the years, in an effort to avoid disparaging the other services. I would say that a minority of CAC rejects are C coins, the rest being problem coins or coins that are severely overgraded.
That's a valid explanation, but then we're back to my point that if they aren't C coins, then CACG has one (or more) categories that aren't publicly stated.
I feel like we're all essentially saying the same thing. Either there are details C coins, or there are other unstated categories (D,E, etc.) as you're pointing out above.
I feel like the counter argument to my point is: "There are no details C coins" AND "CAC only has A,B,C categories". My point is that both can't be true. CACG either grades C coins with details, or they have more (unstated) categories beyond A,B, and C which they don't publicly state.
What you keep referring to as “other unstated categories” beyond A, B, and C, are simply coins that CAC/CACG thinks are either 1) are straight grade material, but over-graded (for example, a 63 coin in a 64 holder) or 2) problem/detail grade coins that NGC or PCGS straight graded, but which deserve detail grades. This has been discussed at length, publicly, regardless of whether you’ve seen it.
Mark Feld* of Heritage Auctions*Unless otherwise noted, my posts here represent my personal opinions.
That doesn't make sense if CAC stickers A and B coins. I doubt they're stickering A coins with PVC.
Coins that otherwise would be considered A or B coins that have PVC do not pass at CAC. At CACG, raw coins that would normally be A, B, or C coins are body bagged.
@DeplorableDan said:
Sorry @Manifest_Destiny, you're one of my favorite posters here but I agree with Mark. Though it may not be explicitly stated on the website, JA has stated on numerous occasions that many coins that fail to sticker are technically not "C" coins. Essentially, there are "D", "E", and "F" coins but JA has been very delicate with his phrasing over the years, in an effort to avoid disparaging the other services. I would say that a minority of CAC rejects are C coins, the rest being problem coins or coins that are severely overgraded.
That's a valid explanation, but then we're back to my point that if they aren't C coins, then CACG has one (or more) categories that aren't publicly stated.
Earlier it was posted:
Q: What percentage of coins CAC reviews sticker? A: As of April 1, 2023, CAC has stickered approximately
43% of the coins received.
So 57% of coins submitted to CAC do not pass. The last thing that coin professionals want to say is most coins are "details." body bagged, or not accurately graded. So JA chose to simply say they are not A or B coins for the good of the hobby. Most often when showing a coin to a coin professional , they will say, "Nice coin." Unless you are selling it.
@Manifest_Destiny said:
We're probably saying the same thing for the most part.
You keep saying this and then go on to explain why you disagree with what everyone else is saying. That doesn't sound to me like you are saying the same thing that everyone else is.
@Manifest_Destiny said:
We're probably saying the same thing for the most part.
You keep saying this and then go on to explain why you disagree with what everyone else is saying. That doesn't sound to me like you are saying the same thing that everyone else is.
We're all saying CACG details grades PCGS/NGC straight graded coins. I think they either do that to C category coins or they have other categories that they don't publicly acknowledge which is what Dan was saying with his "D", "E" and "F" comment.
@Manifest_Destiny said:
We're probably saying the same thing for the most part.
You keep saying this and then go on to explain why you disagree with what everyone else is saying. That doesn't sound to me like you are saying the same thing that everyone else is.
We're all saying CACG details grades PCGS/NGC straight graded coins. I think they either do that to C category coins or they have other categories that they don't publicly acknowledge which is what Dan was saying with his "D", "E" and "F" comment.
Hopefully that helps you understand better.
I think it's you that doesn't understand regardless of how many times it has been explained by multiple members in this thread. I understand perfectly how CAC is grading.
@breakdown said:
I think there are a few categories of coins that fail to sticker and some people get these categories mixed up. C coins should end up in straight grade holders. Coins that are deemed over graded by CAC should also. The third category -
problem coins - get detail holders.
In other words, a C coin is not a problem coin. That’s my understanding anyway.
That’s the way John Albanese has explained it for years. I think I’ll go with his analysis, rather than that of others who predict that most C coins will receive detail grades.
CAC and CACG are using the A,B,C system. They don't mention a 4th category, so where do straight graded PCGS/NGC coins end up when they get details graded by CACG. They had to be "C" coins. There's no other explanation unless there's a secret 4th category that CAC/CACG doesn't publicly acknowledge. It can't be both ways.
@breakdown said:
I think there are a few categories of coins that fail to sticker and some people get these categories mixed up. C coins should end up in straight grade holders. Coins that are deemed over graded by CAC should also. The third category -
problem coins - get detail holders.
In other words, a C coin is not a problem coin. That’s my understanding anyway.
That’s the way John Albanese has explained it for years. I think I’ll go with his analysis, rather than that of others who predict that most C coins will receive detail grades.
CAC and CACG are using the A,B,C system. They don't mention a 4th category, so where do straight graded PCGS/NGC coins end up when they get details graded by CACG. They had to be "C" coins. There's no other explanation unless there's a secret 4th category that CAC/CACG doesn't publicly acknowledge. It can't be both ways.
It’s not very complicated or mysterious. Formerly straight graded NGC and PCGS that failed to sticker at CAC were in their opinion, either:
1) straight-grade worthy but of C quality
2) straight-grade worthy but overgraded (as in deserving of a lower numerical grade)
3) problem coins that should have received detail grades. This category would be your “4th category”.
So coins that receive detail-grades at CACG need not have been C quality coins in the eyes of CAC. They would have fallen into group 3 above, not group 1. Problem coins aren’t a subset of the C quality coin group, as you’ve mistakenly categorized them.
Again, according to John Albanese, most C quality coins will be straight graded by CACG, but at the next numerical grade down with a plus.
We're probably saying the same thing for the most part. My contention is that CAC/CACG are telling us there's only A,B, and C coins, so by default, straight graded problem coins have to be in the "C" category. It is complicated and mysterious in the sense that CAC/CACG don't publicly have a 4th category. If the above info (i.e JA explaining it) is their policy, they should publish it on their website.
No, by default straight graded problem coins do not have to be in the "C" category. Problem coins span the whole ABC spectrum. For example, there are plenty of coins on all parts of the grading (A, B, and C) spectrum with PVC. An MS65+ with PVC is not a "C" coin. CAC will slab the coin in an MS Details slab. Successfully remove the PVC and it theoretically will cross into a CAC 65 or 65+ slab.
That doesn't make sense if CAC stickers A and B coins. I doubt they're stickering A coins with PVC.
@DisneyFan is right. Coins with issues do not get srickers or straight grades at CACG but that doesnt mean the coin falls on C portion of the spectrum.
@Manifest_Destiny said:
We're probably saying the same thing for the most part.
You keep saying this and then go on to explain why you disagree with what everyone else is saying. That doesn't sound to me like you are saying the same thing that everyone else is.
We're all saying CACG details grades PCGS/NGC straight graded coins. I think they either do that to C category coins or they have other categories that they don't publicly acknowledge which is what Dan was saying with his "D", "E" and "F" comment.
Hopefully that helps you understand better.
I think it's you that doesn't understand regardless of how many times it has been explained by multiple members in this thread. I understand perfectly how CAC is grading.
I guess no one is explaining it in a logical way then, except Dan, which is why I said I think we were saying the same thing for the most part.
@breakdown said:
I think there are a few categories of coins that fail to sticker and some people get these categories mixed up. C coins should end up in straight grade holders. Coins that are deemed over graded by CAC should also. The third category -
problem coins - get detail holders.
In other words, a C coin is not a problem coin. That’s my understanding anyway.
That’s the way John Albanese has explained it for years. I think I’ll go with his analysis, rather than that of others who predict that most C coins will receive detail grades.
CAC and CACG are using the A,B,C system. They don't mention a 4th category, so where do straight graded PCGS/NGC coins end up when they get details graded by CACG. They had to be "C" coins. There's no other explanation unless there's a secret 4th category that CAC/CACG doesn't publicly acknowledge. It can't be both ways.
@breakdown said:
I think there are a few categories of coins that fail to sticker and some people get these categories mixed up. C coins should end up in straight grade holders. Coins that are deemed over graded by CAC should also. The third category -
problem coins - get detail holders.
In other words, a C coin is not a problem coin. That’s my understanding anyway.
That’s the way John Albanese has explained it for years. I think I’ll go with his analysis, rather than that of others who predict that most C coins will receive detail grades.
CAC and CACG are using the A,B,C system. They don't mention a 4th category, so where do straight graded PCGS/NGC coins end up when they get details graded by CACG. They had to be "C" coins. There's no other explanation unless there's a secret 4th category that CAC/CACG doesn't publicly acknowledge. It can't be both ways.
It’s not very complicated or mysterious. Formerly straight graded NGC and PCGS that failed to sticker at CAC were in their opinion, either:
1) straight-grade worthy but of C quality
2) straight-grade worthy but overgraded (as in deserving of a lower numerical grade)
3) problem coins that should have received detail grades. This category would be your “4th category”.
So coins that receive detail-grades at CACG need not have been C quality coins in the eyes of CAC. They would have fallen into group 3 above, not group 1. Problem coins aren’t a subset of the C quality coin group, as you’ve mistakenly categorized them.
Again, according to John Albanese, most C quality coins will be straight graded by CACG, but at the next numerical grade down with a plus.
We're probably saying the same thing for the most part. My contention is that CAC/CACG are telling us there's only A,B, and C coins, so by default, straight graded problem coins have to be in the "C" category. It is complicated and mysterious in the sense that CAC/CACG don't publicly have a 4th category. If the above info (i.e JA explaining it) is their policy, they should publish it on their website.
No, by default straight graded problem coins do not have to be in the "C" category. Problem coins span the whole ABC spectrum. For example, there are plenty of coins on all parts of the grading (A, B, and C) spectrum with PVC. An MS65+ with PVC is not a "C" coin. CAC will slab the coin in an MS Details slab. Successfully remove the PVC and it theoretically will cross into a CAC 65 or 65+ slab.
That doesn't make sense if CAC stickers A and B coins. I doubt they're stickering A coins with PVC.
@DisneyFan is right. Coins with issues do not get srickers or straight grades at CACG but that doesnt mean the coin falls on C portion of the spectrum.
I would think that coins with issues would be disqualified from the A or B categories. If not, can you cite something from CAC that explains it?
@breakdown said:
I think there are a few categories of coins that fail to sticker and some people get these categories mixed up. C coins should end up in straight grade holders. Coins that are deemed over graded by CAC should also. The third category -
problem coins - get detail holders.
In other words, a C coin is not a problem coin. That’s my understanding anyway.
That’s the way John Albanese has explained it for years. I think I’ll go with his analysis, rather than that of others who predict that most C coins will receive detail grades.
CAC and CACG are using the A,B,C system. They don't mention a 4th category, so where do straight graded PCGS/NGC coins end up when they get details graded by CACG. They had to be "C" coins. There's no other explanation unless there's a secret 4th category that CAC/CACG doesn't publicly acknowledge. It can't be both ways.
@breakdown said:
I think there are a few categories of coins that fail to sticker and some people get these categories mixed up. C coins should end up in straight grade holders. Coins that are deemed over graded by CAC should also. The third category -
problem coins - get detail holders.
In other words, a C coin is not a problem coin. That’s my understanding anyway.
That’s the way John Albanese has explained it for years. I think I’ll go with his analysis, rather than that of others who predict that most C coins will receive detail grades.
CAC and CACG are using the A,B,C system. They don't mention a 4th category, so where do straight graded PCGS/NGC coins end up when they get details graded by CACG. They had to be "C" coins. There's no other explanation unless there's a secret 4th category that CAC/CACG doesn't publicly acknowledge. It can't be both ways.
It’s not very complicated or mysterious. Formerly straight graded NGC and PCGS that failed to sticker at CAC were in their opinion, either:
1) straight-grade worthy but of C quality
2) straight-grade worthy but overgraded (as in deserving of a lower numerical grade)
3) problem coins that should have received detail grades. This category would be your “4th category”.
So coins that receive detail-grades at CACG need not have been C quality coins in the eyes of CAC. They would have fallen into group 3 above, not group 1. Problem coins aren’t a subset of the C quality coin group, as you’ve mistakenly categorized them.
Again, according to John Albanese, most C quality coins will be straight graded by CACG, but at the next numerical grade down with a plus.
We're probably saying the same thing for the most part. My contention is that CAC/CACG are telling us there's only A,B, and C coins, so by default, straight graded problem coins have to be in the "C" category. It is complicated and mysterious in the sense that CAC/CACG don't publicly have a 4th category. If the above info (i.e JA explaining it) is their policy, they should publish it on their website.
No, by default straight graded problem coins do not have to be in the "C" category. Problem coins span the whole ABC spectrum. For example, there are plenty of coins on all parts of the grading (A, B, and C) spectrum with PVC. An MS65+ with PVC is not a "C" coin. CAC will slab the coin in an MS Details slab. Successfully remove the PVC and it theoretically will cross into a CAC 65 or 65+ slab.
That doesn't make sense if CAC stickers A and B coins. I doubt they're stickering A coins with PVC.
@DisneyFan is right. Coins with issues do not get srickers or straight grades at CACG but that doesnt mean the coin falls on C portion of the spectrum.
I would think that coins with issues would be disqualified from the A or B categories. If not, can you cite something from CAC that explains it?
It's not a category, it's a spectrum. As JA said (and remember I referred you to listening to some interviews you can probably find on Youtube). Take 100 [problem free, non-details] PCGS MS65 Morgan dollars and rank them from worst to best, the top portion are A coins, bottom portion are C coins, and middle are B. Simple as that.
@breakdown said:
I think there are a few categories of coins that fail to sticker and some people get these categories mixed up. C coins should end up in straight grade holders. Coins that are deemed over graded by CAC should also. The third category -
problem coins - get detail holders.
In other words, a C coin is not a problem coin. That’s my understanding anyway.
That’s the way John Albanese has explained it for years. I think I’ll go with his analysis, rather than that of others who predict that most C coins will receive detail grades.
CAC and CACG are using the A,B,C system. They don't mention a 4th category, so where do straight graded PCGS/NGC coins end up when they get details graded by CACG. They had to be "C" coins. There's no other explanation unless there's a secret 4th category that CAC/CACG doesn't publicly acknowledge. It can't be both ways.
@breakdown said:
I think there are a few categories of coins that fail to sticker and some people get these categories mixed up. C coins should end up in straight grade holders. Coins that are deemed over graded by CAC should also. The third category -
problem coins - get detail holders.
In other words, a C coin is not a problem coin. That’s my understanding anyway.
That’s the way John Albanese has explained it for years. I think I’ll go with his analysis, rather than that of others who predict that most C coins will receive detail grades.
CAC and CACG are using the A,B,C system. They don't mention a 4th category, so where do straight graded PCGS/NGC coins end up when they get details graded by CACG. They had to be "C" coins. There's no other explanation unless there's a secret 4th category that CAC/CACG doesn't publicly acknowledge. It can't be both ways.
It’s not very complicated or mysterious. Formerly straight graded NGC and PCGS that failed to sticker at CAC were in their opinion, either:
1) straight-grade worthy but of C quality
2) straight-grade worthy but overgraded (as in deserving of a lower numerical grade)
3) problem coins that should have received detail grades. This category would be your “4th category”.
So coins that receive detail-grades at CACG need not have been C quality coins in the eyes of CAC. They would have fallen into group 3 above, not group 1. Problem coins aren’t a subset of the C quality coin group, as you’ve mistakenly categorized them.
Again, according to John Albanese, most C quality coins will be straight graded by CACG, but at the next numerical grade down with a plus.
We're probably saying the same thing for the most part. My contention is that CAC/CACG are telling us there's only A,B, and C coins, so by default, straight graded problem coins have to be in the "C" category. It is complicated and mysterious in the sense that CAC/CACG don't publicly have a 4th category. If the above info (i.e JA explaining it) is their policy, they should publish it on their website.
No, by default straight graded problem coins do not have to be in the "C" category. Problem coins span the whole ABC spectrum. For example, there are plenty of coins on all parts of the grading (A, B, and C) spectrum with PVC. An MS65+ with PVC is not a "C" coin. CAC will slab the coin in an MS Details slab. Successfully remove the PVC and it theoretically will cross into a CAC 65 or 65+ slab.
That doesn't make sense if CAC stickers A and B coins. I doubt they're stickering A coins with PVC.
@DisneyFan is right. Coins with issues do not get srickers or straight grades at CACG but that doesnt mean the coin falls on C portion of the spectrum.
I would think that coins with issues would be disqualified from the A or B categories. If not, can you cite something from CAC that explains it?
It's not a category, it's a spectrum. As JA said (and remember I referred you to listening to some interviews you can probably find on Youtube). Take 100 [problem free, non-details] PCGS MS65 Morgan dollars and rank them from worst to best, the top portion are A coins, bottom portion are C coins, and middle are B. Simple as that.
I get that. I was responding to the comment that problem coins exist in all 3 categories and could get a details grade at CACG. I would think coins with issues would fall into the C portion of the spectrum, not the A and B portion.
I can't find the thread that clearly explained that J/A used the A,B, and C categories simply as another way to help some collectors better understand CAC grading. He tries to focus on the positives rather than bashing the other TPG's. The vast majority of numismatists realize that there are far too many over-graded and problem coins in straight graded holders.
When formulating a grade try to focus on the specific characteristics of a coin rather than obsessing on its general alphabetic category.
It's inevitable. It will be done in our life time and a lot sooner than most may anticipate.
As I stated earlier, the way I envision it is AI assigning a grade after analysis of 1 clock to 12 o'clock photos of both obverse and reverse. Then 3 or 5 very senior human graders vote the grade assignment up or down.
This makes the most sense to my thinking.
Your vision has a major flaw.
If AI takes the place of human graders and "very senior human graders" vote on the AI-assigned grade, where will these very senior human graders come from?
How will they achieve that "very senior" status? If AI is doing the grading, how will human graders get the experience necessary to cast the vote of approval? What will junior and mid-level graders do to eventually gain senior status? Will the grading companies keep a shadow staff of humans in a parallel process to train them to become the "very senior" graders who approve the AI grades?
Maybe we can build AI robot coin collectors to collect AI graded coins. 🤔
If he’s under grading them (CACG) , trigger happy on details nothing there for me. Not one waste my money that way. But curious about crackout potential some them get upgraded somewhere else then Flip n Bank. Yay.
. It will be done in our life time and a lot sooner than most may anticipate. @Cougar1978 said:
If he’s under grading them (CACG) , trigger happy on details nothing there for me. Not one waste my money that way. But curious about crackout potential some them get upgraded somewhere else then Flip n Bank. Yay.
You haven’t even been approved for membership/submitting there, have you?
Mark Feld* of Heritage Auctions*Unless otherwise noted, my posts here represent my personal opinions.
@Manifest_Destiny said:
We're probably saying the same thing for the most part.
You keep saying this and then go on to explain why you disagree with what everyone else is saying. That doesn't sound to me like you are saying the same thing that everyone else is.
We're all saying CACG details grades PCGS/NGC straight graded coins. I think they either do that to C category coins or they have other categories that they don't publicly acknowledge which is what Dan was saying with his "D", "E" and "F" comment.
Hopefully that helps you understand better.
Not trying to be argumentative here, just trying to learn. But can you point to where CACG has said there are only A, B and C coins?
"Look up, old boy, and see what you get." -William Bonney.
On the subject of AI, I could see computers used to measure and identify objective elements, such as wear or friction, strike, marks, luster perhaps. Graders would use that information to speed up the grading process and improve accuracy. Not AI, but an assist.
I sent 60 coins to CACG recently. About 1/4 came back details and a handful came back AU58 that likely would’ve gotten MS at PCGS.
Of the details coins, I agreed with most - some I felt would maybe be net graded elsewhere but there were 4-5 that I truly was befuddled at.
I cracked those as well as the sliders and will see how they come back from PCGS.
Turn around time was extremely quick. I had grades for the Economy submissions in about 10 days. The express was around 6.
Overall I have no intentions to send to CACG again. They are not commanding enough of a premium on the market to compensate for the strict grading. Right now it seems a CACG coin is going for somewhere in between PCGS and PCGS + CAC pricing.
@abbyme24 said:
I sent 60 coins to CACG recently. About 1/4 came back details and a handful came back AU58 that likely would’ve gotten MS at PCGS.
Of the details coins, I agreed with most - some I felt would maybe be net graded elsewhere but there were 4-5 that I truly was befuddled at.
I cracked those as well as the sliders and will see how they come back from PCGS.
Turn around time was extremely quick. I had grades for the Economy submissions in about 10 days. The express was around 6.
Overall I have no intentions to send to CACG again. They are not commanding enough of a premium on the market to compensate for the strict grading. Right now it seems a CACG coin is going for somewhere in between PCGS and PCGS + CAC pricing.
Welcome to the forum, you have made some interesting posts and I'm looking forward to seeing more of your content here.
@abbyme24 said:
I sent 60 coins to CACG recently. About 1/4 came back details and a handful came back AU58 that likely would’ve gotten MS at PCGS.
Of the details coins, I agreed with most - some I felt would maybe be net graded elsewhere but there were 4-5 that I truly was befuddled at.
I cracked those as well as the sliders and we see how they come back from PCGS.
Turn around time was extremely quick. I had grades for the Economy submissions in about 10 days. The express was around 6.
Overall I have no intentions to send to CACG again. They are not commanding enough of a premium on the market to compensate for the strict grading. Right now it seems a CACG coin is going for somewhere in between PCGS and PCGS + CAC pricing.
Welcome to the forum, you have made some interesting posts and I'm looking forward to seeing more of your content here.
Thanks! I’ve been lurking for a few months in various coin forums and have seen a lot of your posts. Really impressed with your knowledge and what you are doing with PR!
Just as an aside, I will happily buy CACG in the current market because I do think they are undervalued. I’m not interested in trying to get them into a PCGS holder and then stickered, but when I’m buying for my personal collection I will not hesitate to buy CACG since I can often get a similar quality coin at a discount vs. PCGS + CAC.
I think over time the market will adjust to more closely align with P+C pricing. But if for some reason it doesn’t then I would probably crack them out (assuming they still sticker).
@abbyme24 said:
Just as an aside, I will happily buy CACG in the current market because I do think they are undervalued. I’m not interested in trying to get them into a PCGS holder and then stickered, but when I’m buying for my personal collection I will not hesitate to buy CACG since I can often get a similar quality coin at a discount vs. PCGS + CAC.
I think over time the market will adjust to more closely align with P+C pricing. But if for some reason it doesn’t then I would probably crack them out (assuming they still sticker).
Fully agree with you, I'm an aggressive buyer of nice CACG coins at this time.
@abbyme24 said:
Just as an aside, I will happily buy CACG in the current market because I do think they are undervalued. I’m not interested in trying to get them into a PCGS holder and then stickered, but when I’m buying for my personal collection I will not hesitate to buy CACG since I can often get a similar quality coin at a discount vs. PCGS + CAC.
I think over time the market will adjust to more closely align with P+C pricing. But if for some reason it doesn’t then I would probably crack them out (assuming they still sticker).
Fully agree with you, I'm an aggressive buyer of nice CACG coins at this time.
I’m mystified why coins that are currently graded are submitted to CACG raw or are crossed without specifying a minimum grade. What am I missing? An expensive experiment is my only conclusion.
@oldabeintx said:
I’m mystified why coins that are currently graded are submitted to CACG raw or are crossed without specifying a minimum grade. What am I missing? An expensive experiment is my only conclusion.
There's a lot of experimentation happening, so you would be correct. Many are trying to get a feel for their standards, so they are throwing things at the wall to see what sticks. In addition, I've seen some very high asking prices for CACG coins that were downgraded from their previous holder, often times the seller includes the old certification as justification for the high price. Some sell, and some don't, but stale inventory is never fun to deal with and a fresh holder at CACG provides an opportunity for liquidation.
I get that. I was responding to the comment that problem coins exist in all 3 categories and could get a details grade at CACG. I would think coins with issues would fall into the C portion of the spectrum, not the A and B portion.
Technically, a coin could be an A or B coin except for a "detail." Best example would be a wonderful A grade coin with questionable toning or body bagged because you can't see the PVC.
@Manifest_Destiny said:
We're probably saying the same thing for the most part.
You keep saying this and then go on to explain why you disagree with what everyone else is saying. That doesn't sound to me like you are saying the same thing that everyone else is.
We're all saying CACG details grades PCGS/NGC straight graded coins. I think they either do that to C category coins or they have other categories that they don't publicly acknowledge which is what Dan was saying with his "D", "E" and "F" comment.
Hopefully that helps you understand better.
Not trying to be argumentative here, just trying to learn. But can you point to where CACG has said there are only A, B and C coins?
Understood, and thanks for being cordial. This comes from the CAC/CACG website.
@Manifest_Destiny said:
We're probably saying the same thing for the most part.
You keep saying this and then go on to explain why you disagree with what everyone else is saying. That doesn't sound to me like you are saying the same thing that everyone else is.
We're all saying CACG details grades PCGS/NGC straight graded coins. I think they either do that to C category coins or they have other categories that they don't publicly acknowledge which is what Dan was saying with his "D", "E" and "F" comment.
Hopefully that helps you understand better.
Not trying to be argumentative here, just trying to learn. But can you point to where CACG has said there are only A, B and C coins?
Understood, and thanks for being cordial. This comes from the CAC/CACG website.
And not stated (likely because they thought it would be obvious) is that in the opinion of CAC/CACG there are also 1) straight-grade coins, which are overgraded. Call them D, E, Z, fail or whatever you want. And 2) problem coins, which should receive detail grades.
That’s five categories, in all.
Mark Feld* of Heritage Auctions*Unless otherwise noted, my posts here represent my personal opinions.
Comments
We're probably saying the same thing for the most part. My contention is that CAC/CACG are telling us there's only A,B, and C coins, so by default, straight graded problem coins have to be in the "C" category. It is complicated and mysterious in the sense that CAC/CACG don't publicly have a 4th category. If the above info (i.e JA explaining it) is their policy, they should publish it on their website.
Yea of course, the handling will likely continue to be by humans, and the humans would need to feed the machine. But for photos, PCGS system is almost entirely automated by this point, I presume. Which would explain the cookie cutter light placement and poor quality.
But try to think about this from a non-corporate perspective. Everyone in the hobby realizes that human grading will always be subjective, and highly fallible. We all realize the potential of AI, and how it could change the landscape. Several young entrepreneurs have tried to develop an accurate coin grading app/system already. There will be numerous additional attempts to make AI grading functional. If I were to do it, I would build some type of rotating table with lighting, and the computer would analyze a video or live feed of the coin, being tilted and rotated under light. It would be able to measure the luster and contact marks, and with integrated depth mapping and a precise scale, it could determine the level of wear.
I don't expect PCGS or NGC to be the ones that initially develop this technology, I just expect that they're the ones who will end up with it.
Founder- Peak Rarities
Website
Instagram
Facebook
I don’t disagree with you, Dan. But I hope you’re wrong. There’s something to be said for having fellow humans exercise the nuances of grading a coin. There are also all kinds of ramifications and moral quandaries regarding the ever expanding role of AI. Graders and support personnel are all humans after all, and they need to live and likely have families to raise and support.
I’ll resist until there’s only the option but to use a TPG utilizing AI. I’d rather have the hint of uncertainty and disagreement over a point any day, than know I’m helping put people out of work who are otherwise capable and competent. I’m no communist, I just like humanity.
Having fun while switching things up and focusing on a next level PCGS slabbed 1950+ type set, while still looking for great examples for the 7070.
It's remarkable that a collector just uploaded written ANA grading standards and a couple pics and got a close grade back with a reasoned explanation from AI. It means that the tech probably already exists or will very soon to replace the grading room with your phone and replace the whole concept of TPGS. Imagine being able to identify every individual coin in the world that people put in front of their phones by the tiny individual variations of each coin. Every coin has a fingerprint so to speak. Coins could be instantly imaged/graded with a consistent standard, and it could be done so cheaply that it could be monetized without charging for the service. You could point your cam at a coin, identify, describe in detail, video/image and grade it instantly, see what it's worth in real time, and see who in the world wants to buy or sell it. It could replace TPGS, auctions and shows. Everyone could verify title on every individual coin so stolen coins wouldn't be as valuable. Those are just my initial thoughts after 5 minutes, there's got to be a million other applications for AI in coins.
What if the AI just assigned the initial grade, but human finalizers/QA specialists were still part of the assembly line? I think AI has the potential to level the playing field even more than TPGs and CAC, and would ultimately benefit the hobby in a positive way in the long term. In the short term, as a coin dealer with selfish motives, I'd rather have some inconsistency so I can use arbitrage and exploit the weaknesses of human grading, which is an inherently flawed system.
Founder- Peak Rarities
Website
Instagram
Facebook
This is a compromise I could certainly live with. I just believe after a certain point of AI refinement a company, whose goal is to rightly seek profit, would feel the reviewers are no longer necessary. Kind of a fine line that’s going to have to be reconciled where the free market is going to have to come to terms with humanity. Certainly some interesting times ahead….
Having fun while switching things up and focusing on a next level PCGS slabbed 1950+ type set, while still looking for great examples for the 7070.
They're private companies so we don't have access to their financials so everything here is a supposition. But I think that one of the key factors here is the difference between classics and moderns. We are of course predominantly interested in classics because that's what we like and it's almost completely what we focus on. But we're also the minority of the collecting community. Just like this forum is not an accurate representation of the entire population of collectors.
But there's a huge population of people who don't know or care about the grading or company. I'm not sure how much you've bumped into them in the shows yet but they're there, that's why those TV hucksters have done so well for so long.
I think you're absolutely right about the technology and implementation. I just don't think there is enough money in grading/reholdering/crossing classic coins for such a large capital infusion from the private equity companies that own P & N.
Chopmarked Trade Dollar Registry Set --- US & World Gold Showcase --- World Chopmark Showcase
They never said there were only A,B, and C coins. They said there were those three categories (all straight grade worthy) AND problem/detail grade coins, as well.
They also said they don’t want C coins in their holders, so most C coins will be lowered a grade and given a plus.
Coins in straight grade NGC and PCGS holders that CAC/CACG see as detail grade/problem coins AREN’T C coins.
None of this is mysterious or new and it’s been discussed openly for a long time.
Mark Feld* of Heritage Auctions*Unless otherwise noted, my posts here represent my personal opinions.
Imagine a system where you place the coin in a receptacle, a rotating camera takes thousands of images under multiple magnifications, and AI renders a verdict- all in less that 30 seconds. Again, then have a three to five CACG level graders pass it for encapsulation.
No emotion. No bias. No vested interest in a particular outcome. No financial interest. Just cold, hard, analytical determinations.
At this point, we're either nit picking, or talking past each other to some degree. That's ok, I'll continue a bit longer because we're having a technical conversation and technicalities matter.
First, I'm talking about coins already straight graded PCGS/NGC. If you're talking about raw coins submitted to CACG, then ok. My point is about straight graded PCGS/NGC coins.
CAC absolutely is saying they only have 3 categories, A, B, and C in regard to straight graded PCGS/NGC coins.
There's no mention of a "details" category or that some/many "C" coins should be details graded.
That being said, many "C" coins will receive a details grade from CAC, not necessarily a downgrade. We know this for a fact based on submissions posted here in the forum.
So again, my contention is that CACG will details grade many straight graded PCGS/NGC coins that the CAC stickering service regards as "C" coins in straight graded PCGS/NGC holders. If they have a mysterious "D" category for straight graded PCGS/NGC coins that should be in a details holder, then they need to say that on their web site. Since they don't, it's logical to assume that those coins are a subset of "C" coins.
I think my position is logically consistent with CAC website information and actual CACG grading practices.
1) “CAC absolutely is saying they only have 3 categories, A, B, and C in regard to straight graded PCGS/NGC coins.”
I never said otherwise. And in fact, I’ve repeated it multiple times.
2) “There's no mention of a "details" category or that some/many "C" coins should be details graded.”
I see no reason why a grading company would feel any need to mention that problem coins would receive details, rather than straight grades. That seems quite obvious.
How does one know how many C coins are receiving detail grades? As far as I know, C coins aren’t marked with a “C” or other clue in order to categorize them as such.
3) “That being said, many "C" coins will receive a details grade from CAC, not necessarily a downgrade. We know this for a fact based on submissions posted here in the forum.”
How do we know this for a fact? Because a submitter said so and based on images, it looks like he’s correct?
And again, there’s no “mysterious "D" category for straight graded PCGS/NGC coins that should be in a details holder”. Coins in straight grade holders that CAC/CACG deems problem/detail grade coins receive detail grades. I’m not talking about straight grade C coins.
I give up trying to explain things that have been posted numerous times since before CACG even started taking submissions. Disagree as much as you like.
Mark Feld* of Heritage Auctions*Unless otherwise noted, my posts here represent my personal opinions.
Sorry @Manifest_Destiny, you're one of my favorite posters here but I agree with Mark. Though it may not be explicitly stated on the website, JA has stated on numerous occasions that many coins that fail to sticker are technically not "C" coins. Essentially, there are "D", "E", and "F" coins but JA has been very delicate with his phrasing over the years, in an effort to avoid disparaging the other services. I would say that a minority of CAC rejects are C coins, the rest being problem coins or coins that are severely overgraded.
Founder- Peak Rarities
Website
Instagram
Facebook
Fully agree with everything you are saying re: ai with the exception that using a scale to determine wear won’t be accurate. There is too much variation in weight from the mint for at least classic series.
I have been in some brainstorming sessions with people regarding developing an AI grading service. I am a founder of a tech startup and as I plan on exiting in the next few years, I’m trying to figure out what venture I want to explore next.
I think AI grading is a huge opportunity - like you said, if a service can be developed that provided hyper accurate (or more importantly to the collector, consistent) grading, then market forces will eventually push all existing coins (classic, at least) to be graded by this service. Especially if it can be done quickly and affordably (something like $20 with a one week turnaround).
Building the physical system itself is simple, all you need is a high quality camera, good lights, and a rotating base to simulate what a human does when analyzing luster.
The software part is what would be the trickiest because there would be tons of data needed to train the AI on all of the nuances of grading. Is it hairlines or polish lines? Is it die crack or PMD? Is it wear or a weak strike? Etc etc
In fact in our discussions the biggest barrier to doing this is not necessarily the cost of the physical system or the AI engineers to code it, but the huge amounts of $$ that would need to be spent in order to gather enough examples to train the AI. Sure they could be resold, but it would suck up capital and there is value risk since they would have to be cracked out for the most accurate training data.
I understand the sentiment but you can make the same argument regarding people losing jobs for every single technological advance humans have made. Society will adapt with AI just as it did with the internet and with the Industrial Revolution and everything that happened before.
The thing I don’t really agree with is grading accurately graded “C” coins a grade lower with a +.
If the “C” coin is an accurately graded coin in Fine condition, I don’t get grading it VG+. Why not grade it Fine and grade the “A” and “B” coins F+? Otherwise you are changing the grading standards or is that the ultimate goal?
Joe.
ABC subjective imo. However if he is under grading, players will pick them off submit somewhere else. It’s probably already happening.
No, by default straight graded problem coins do not have to be in the "C" category. Problem coins span the whole ABC spectrum. For example, there are plenty of coins on all parts of the grading (A, B, and C) spectrum with PVC. An MS65+ with PVC is not a "C" coin. CAC will slab the coin in an MS Details slab. Successfully remove the PVC and it theoretically will cross into a CAC 65 or 65+ slab.
And dealers and collectors will still vehemently disagree with the determinations.
For anyone who's dealt with automation, 80% of the job is typically straightforward and easy. It's the last 20% that's the bugger. Getting a car to drive itself down a road and stop at a red light or stop sign or not hit a person is easy. The hard part is programming the car to wait in line at a school to pick up a kid or to find a parking space in a parking garage. That is the problem with grading and to make the problem worse is that all dealers or collectors will assume that their coin is in the small percentage that AI can't accurately grade and use that to insist that AI got the coin wrong and you should pay more because the coin is better than the machine says it is. So are we really gaining anything?
And for those who suggest that a machine grades a coin and then humans finalize it... what's the point? Might as well just have the humans grade the coin too.
I’m getting a headache. Thank goodness it’s happy hour.
That doesn't make sense if CAC stickers A and B coins. I doubt they're stickering A coins with PVC.
That's a valid explanation, but then we're back to my point that if they aren't C coins, then CACG has one (or more) categories that aren't publicly stated.
I feel like we're all essentially saying the same thing. Either there are details C coins, or there are other unstated categories (D,E, etc.) as you're pointing out above.
I feel like the counter argument to my point is: "There are no details C coins" AND "CAC only has A,B,C categories". My point is that both can't be true. CACG either grades C coins with details, or they have more (unstated) categories beyond A,B, and C which they don't publicly state.
I think a coin with obvious PVC green gunk worse category than C lol. What shall it be called - ruined, cull, no go, destroyed, not marketable? What pct can even be salvaged?
In reality / Not all of them can be salvaged. It may have been an A coin b4 the pvc,etc but once it’s got pvc what the heck does it matter what it was b4?
For me:
anything I am keeping for myself goes to PCGS (that would higher dollar stuff, better stuff)
for the business:
Nicer early type, better date seated, bust material, better mid grade & up copper and higher better date dollars PCGS.
Generic gold ,lower to mid grade key dates, Modern to NGC (except like a 95-W that I decided to keep for myself as it was darn nice sent to PCGS.
World coins NGC by far.
Paper Money pretty much 95% PMG (I have used PCGS Banknote on some , but only as a convience becuase they were at a show I was attending .
I like CAC stickering, but have not sent anything to the grading service. From what I have seen I like their grading, but i really am not fond of the holders.
What you keep referring to as “other unstated categories” beyond A, B, and C, are simply coins that CAC/CACG thinks are either 1) are straight grade material, but over-graded (for example, a 63 coin in a 64 holder) or 2) problem/detail grade coins that NGC or PCGS straight graded, but which deserve detail grades. This has been discussed at length, publicly, regardless of whether you’ve seen it.
Mark Feld* of Heritage Auctions*Unless otherwise noted, my posts here represent my personal opinions.
Coins that otherwise would be considered A or B coins that have PVC do not pass at CAC. At CACG, raw coins that would normally be A, B, or C coins are body bagged.
Earlier it was posted:
Q: What percentage of coins CAC reviews sticker? A: As of April 1, 2023, CAC has stickered approximately
43% of the coins received.
So 57% of coins submitted to CAC do not pass. The last thing that coin professionals want to say is most coins are "details." body bagged, or not accurately graded. So JA chose to simply say they are not A or B coins for the good of the hobby. Most often when showing a coin to a coin professional , they will say, "Nice coin." Unless you are selling it.
You keep saying this and then go on to explain why you disagree with what everyone else is saying. That doesn't sound to me like you are saying the same thing that everyone else is.
My Collection of Old Holders
Never a slave to one plastic brand will I ever be.
We're all saying CACG details grades PCGS/NGC straight graded coins. I think they either do that to C category coins or they have other categories that they don't publicly acknowledge which is what Dan was saying with his "D", "E" and "F" comment.
Hopefully that helps you understand better.
I think it's you that doesn't understand regardless of how many times it has been explained by multiple members in this thread. I understand perfectly how CAC is grading.
My Collection of Old Holders
Never a slave to one plastic brand will I ever be.
@DisneyFan is right. Coins with issues do not get srickers or straight grades at CACG but that doesnt mean the coin falls on C portion of the spectrum.
I guess no one is explaining it in a logical way then, except Dan, which is why I said I think we were saying the same thing for the most part.
I would think that coins with issues would be disqualified from the A or B categories. If not, can you cite something from CAC that explains it?
It's not a category, it's a spectrum. As JA said (and remember I referred you to listening to some interviews you can probably find on Youtube). Take 100 [problem free, non-details] PCGS MS65 Morgan dollars and rank them from worst to best, the top portion are A coins, bottom portion are C coins, and middle are B. Simple as that.
I get that. I was responding to the comment that problem coins exist in all 3 categories and could get a details grade at CACG. I would think coins with issues would fall into the C portion of the spectrum, not the A and B portion.
I can't find the thread that clearly explained that J/A used the A,B, and C categories simply as another way to help some collectors better understand CAC grading. He tries to focus on the positives rather than bashing the other TPG's. The vast majority of numismatists realize that there are far too many over-graded and problem coins in straight graded holders.
When formulating a grade try to focus on the specific characteristics of a coin rather than obsessing on its general alphabetic category.
Your vision has a major flaw.
If AI takes the place of human graders and "very senior human graders" vote on the AI-assigned grade, where will these very senior human graders come from?
How will they achieve that "very senior" status? If AI is doing the grading, how will human graders get the experience necessary to cast the vote of approval? What will junior and mid-level graders do to eventually gain senior status? Will the grading companies keep a shadow staff of humans in a parallel process to train them to become the "very senior" graders who approve the AI grades?
Maybe we can build AI robot coin collectors to collect AI graded coins. 🤔
If he’s under grading them (CACG) , trigger happy on details nothing there for me. Not one waste my money that way. But curious about crackout potential some them get upgraded somewhere else then Flip n Bank. Yay.
You haven’t even been approved for membership/submitting there, have you?
Mark Feld* of Heritage Auctions*Unless otherwise noted, my posts here represent my personal opinions.
Not trying to be argumentative here, just trying to learn. But can you point to where CACG has said there are only A, B and C coins?
"Look up, old boy, and see what you get." -William Bonney.
On the subject of AI, I could see computers used to measure and identify objective elements, such as wear or friction, strike, marks, luster perhaps. Graders would use that information to speed up the grading process and improve accuracy. Not AI, but an assist.
I sent 60 coins to CACG recently. About 1/4 came back details and a handful came back AU58 that likely would’ve gotten MS at PCGS.
Of the details coins, I agreed with most - some I felt would maybe be net graded elsewhere but there were 4-5 that I truly was befuddled at.
I cracked those as well as the sliders and will see how they come back from PCGS.
Turn around time was extremely quick. I had grades for the Economy submissions in about 10 days. The express was around 6.
Overall I have no intentions to send to CACG again. They are not commanding enough of a premium on the market to compensate for the strict grading. Right now it seems a CACG coin is going for somewhere in between PCGS and PCGS + CAC pricing.
Welcome to the forum, you have made some interesting posts and I'm looking forward to seeing more of your content here.
Founder- Peak Rarities
Website
Instagram
Facebook
Thanks! I’ve been lurking for a few months in various coin forums and have seen a lot of your posts. Really impressed with your knowledge and what you are doing with PR!
Just as an aside, I will happily buy CACG in the current market because I do think they are undervalued. I’m not interested in trying to get them into a PCGS holder and then stickered, but when I’m buying for my personal collection I will not hesitate to buy CACG since I can often get a similar quality coin at a discount vs. PCGS + CAC.
I think over time the market will adjust to more closely align with P+C pricing. But if for some reason it doesn’t then I would probably crack them out (assuming they still sticker).
Fully agree with you, I'm an aggressive buyer of nice CACG coins at this time.
Founder- Peak Rarities
Website
Instagram
Facebook
Good thing we seem to be in separate markets!! 🤣
I’m mystified why coins that are currently graded are submitted to CACG raw or are crossed without specifying a minimum grade. What am I missing? An expensive experiment is my only conclusion.
There's a lot of experimentation happening, so you would be correct. Many are trying to get a feel for their standards, so they are throwing things at the wall to see what sticks. In addition, I've seen some very high asking prices for CACG coins that were downgraded from their previous holder, often times the seller includes the old certification as justification for the high price. Some sell, and some don't, but stale inventory is never fun to deal with and a fresh holder at CACG provides an opportunity for liquidation.
Founder- Peak Rarities
Website
Instagram
Facebook
Technically, a coin could be an A or B coin except for a "detail." Best example would be a wonderful A grade coin with questionable toning or body bagged because you can't see the PVC.
Understood, and thanks for being cordial. This comes from the CAC/CACG website.
And not stated (likely because they thought it would be obvious) is that in the opinion of CAC/CACG there are also 1) straight-grade coins, which are overgraded. Call them D, E, Z, fail or whatever you want. And 2) problem coins, which should receive detail grades.
That’s five categories, in all.
Mark Feld* of Heritage Auctions*Unless otherwise noted, my posts here represent my personal opinions.