Coin Community Family of Web Sites
Like us on Facebook! Subscribe to our Youtube Channel! Check out our Twitter! Check out our Pinterest!
Username:
Password:
Save Password
Forgot your Password?


Welcome Guest! Need help? Got a question? Inherit some coins?
Our coin forum is completely free! Register Now!

What Has Been Your Worst Grade Experience With A Third Party Grading Company?

First Page
Showing last 15 replies.
 
To participate in the forum you must log in or register.
Author Previous TopicReplies: 21 / Views: 1,995Next Topic Page 2 of 2
Pillar of the Community
United States
1587 Posts
 Posted 08/15/2020  5:51 pm  Show Profile   Bookmark this reply Add jmkendall to your friends list Get a Link to this Reply
My buisness partner was an Anacs and later NGC grader "back in the day". I got a really nice 1879-O Morgan. Well struck, PL, no significant marks on it at all. My partner said "Send it in. It is an easy "66", with a serious shot at a 7 holder".

So, I did. It came back as an AU-58. My partner said "I was afraid of that. With the census that low, they didn't want to affect the value of the one coin already there". The real "tell" is when you get an offer for your coin.

And sure enough.....
Pillar of the Community
United States
4194 Posts
 Posted 08/24/2020  11:10 am  Show Profile   Bookmark this reply Add hfjacinto to your friends list Get a Link to this Reply
This was my worst one ever. I purchased this assuming it was graded correctly. How was this not detailed?

Edited by hfjacinto
08/24/2020 11:11 am
New Member
United States
11 Posts
 Posted 06/23/2021  3:02 pm  Show Profile   Bookmark this reply Add bgreene to your friends list Get a Link to this Reply
I had my coins authenticated by a reputable dealer prior to 2 submissions at PCGS. Both submissions came back without variety designation and at very low grades for coins that came from uncirculated rolls. The coins I submitted were of far higher grade than the examples of the same coins on the PCGS website. I have been informed this is a well known scam by PCGS to increase their bottom line with resubmissions. I will never use PCGS again.

*** Edited by Staff to remove YELLING. All capital letters is the internet version of yelling. Please do not do it in titles or posts. ***
Pillar of the Community
Learn More...
United States
8817 Posts
 Posted 06/23/2021  4:21 pm  Show Profile   Check GrapeCollects's eBay Listings Bookmark this reply Add GrapeCollects to your friends list Get a Link to this Reply
First of all, caps. Second, did you ask for variety and make sure the dealer did? Third, are you sure they were actually BU Rolls? And forth, and don't take this the wrong way, do you know how to grade? PCGS isn't a scam.
My best finds: 1999-WAM:http://goccf.com/t/332161 1988-RDV-6:http://goccf.com/t/335954#2873459 1986-Off-center: http://goccf.com/t/335952
1999 WAM #2:http://goccf.com/t/338710&whichpage=1
1981 Double Struck In Collar: http://goccf.com/t/350199&whichpage=1
ANA id: 3194067
My Type Set: https://www.NGCcoin.com/registry/co...sets/236574/
If you want to buy something or sell something or just talk, shoot me a PM!
Pillar of the Community
United States
7447 Posts
 Posted 06/24/2021  10:51 am  Show Profile   Bookmark this reply Add Earle42 to your friends list Get a Link to this Reply
@bgreene
It is not common to get "high money grades" on coins from uncirculated bank rolls because grades that make money are not that common.

As to PCGS and the other companies being a scam, it depends on how much "gray areas" that exists, in an awful lot of companies and marketing today (not just coins), it takes for an individual to call a company a scam.

PCGS is not, as GrapeCollects says, a provable scamming company like telemarketers can be that is out to take everyone for a ride.

Unfortunately a lot of people feel they have gotten taken by grading companies b/c they do not do the homework to understand the actual part these companies play. Since so many slabs are for sale online, people just assume PCGS, and other grading companies, is/are THE undeniable PROFESSIONALS concerning all aspects of coin collecting. And their marketing departments love this b/c it brings in tons of business.

PCGS is a company, looking to make a profit (as any other company) who GRADES coins. They do not automatically assign varieties, they do not automatically tell you what type of error a coin is. They grade.

Another thing which is said on the website (but who reads directions/instructions anyway - as all good marketing departments love in any business?) is that grading coins is not a science, it is an art.

In other words, there is absolutely no scientific standard applied to grading coins by coin grading companies. You can break a coin one out of a PCGS slab, send it back into PCGS, and never have a guarantee it will get the same grade again.

A former MS64 coin might come back as MS62 (on an uncommonly bad day the graders are having), MS63, MD64, MS65, MS66 (on a good day for you!). No science...no accountability...no verifying the grade by any factual means.


Now the major gray area...
The technology has been there (and was marketed by PCGS in the early 90s) to have a scientific standard, and PCGS even claimed it was more accurate than human grading as an advertising method to get people to their computer grading system. But the scientific approach was dropped with the companies saying since people (back then) did not trust computers, they went back to the less accurate human grading. And here we are...30 years later!

Nowadays everyone carries with them a more powerful computer (and scanner in some cases with facial recognition) than the antique 1990 computers. However, the "art" is kept.

Why? No solid facts are mentioned known. But we do know companies are out to make money. If a verifiable method of grading coins was implemented, then the resubmitting game in hopes of getting a better grade would go out the window. And resubmitting coins is a very profitable area of for these companies. It is likely in their best interest NOT to be based on fact.

There are other instances such as some pretty convincing data pointing to grading companies grading monster boxes of ASEs and probably just skimming 20% off the top to make into MS70s and the rest be put in lesser graded slabs.

There is no actual verifiable difference between an MS69 and MS70 anyway. But people pay more for the MS70, so it makes sense for a business to keep MS70 percentages below a certain rate to encourage better business for the company. How? SO many people are willing to gamble (and pay the company to slab) their ASE to get the "money grade."

Start examining MS70 ASE's for yourself. You will find it not difficult to find an obviously damaged one that should never have been an MS70 if the company-alleged three graders looked at it.

Scam? It depends on how much a person accepts gray areas as being a "legitimate part" of the business world. Its all around us.

Also please look at the PDF linked to in my signature. Here is legit proof the companies are very lacking in variety attribution...and some other claims. These problems have cost thousand$ of dollar$ (all linked int he article to the PCGS website for proof) to the unfortunate who put blind faith in these companies as being THE experts of anything and everything having to do with coins.

When a person understands all of the above, understand to buy the coin and not the slab, and understand the value of slab+label+coin can easily disappear if the plastic does, then slabbed coins coins can, and do, make an enjoyable quest for people to collect.

But just like everything else in life, a person who jumps in feet first and does not understand these companies for what they are can be very disappointed.

Hobbies are about FUN. If a person likes slabbed coins, then all the more power to them. This is the way they enjoy the hobby.
How much squash could a Sasquatch squash if a Sasquatch would squash squash?
Pillar of the Community
Learn More...
United States
3555 Posts
 Posted 06/28/2021  9:08 pm  Show Profile   Bookmark this reply Add BStrauss3 to your friends list Get a Link to this Reply
Actually, we do know why computerized technical grading fails.

Customers don't like it.

There are also technical difficulties. If you go back and read the compugrade patent, it was based on mapping a coin into 50-100 zones, calculating the number of marks & rubs in each zone, and mathematically combining that to create a grade. Each new type of coin or change within a coin would require a new map. Just imagine the complexity of the Washington quarters... 3 or 4 versions prior to 1999, 56 50SG, 56ATB, the new WAxDEL, and then 2022's notable women.

The complexity with a modern AI system is that you need a training set of sufficient completeness for the neural network. It has to contain multiple examples that everybody agrees with, for what 28 grades? And * and + grades.
-----Burton
47 year / Life ANA member (joined 12/1/1973)
Life member: Numismatics International, CONECA
Member: TNA, FtWCC, NETCC, OnLinw Coin Club
Owned by four cats and a wife of 37 years (joined 1983)
Pillar of the Community
United States
7447 Posts
 Posted 06/28/2021  9:38 pm  Show Profile   Bookmark this reply Add Earle42 to your friends list Get a Link to this Reply

Quote:
Actually, we do know why computerized technical grading fails.

Customers don't like it.


Don't or didn't? I may be mistaken, but I thought all these types of grading system were gone back in the 90s. If this is the case, then it makes sense that the companies said the systems were not accepted b/c people didn;t trust computers.

But nowadays, 30 years later, most people have in their pocket a computer much more capable than any computer of 30 years ago and few people don't trust computers even with their financial dealings.


Quote:
Each new type of coin or change within a coin would require a new map. Just imagine the complexity of the Washington quarters... 3 or 4 versions prior to 1999, 56 50SG, 56ATB, the new WAxDEL, and then 2022's notable women.

Also simple with today's machines and not a big deal with computer storage capacities.

In fact it could be a modern iPhone, taking 30,000 data points with IR instantaneously for facial recognition (and being able to recognize with changes such as glasses, mustaches, etc.) could be made to develop a handheld app. The tech may already be in hand, just not developed.


Quote:
The complexity with a modern AI system...

Already in development. Minstate.com has a system they are developing in which they took huge amounts of slabbed Morgans at each grade level and let the computer learn what humans label at each grade point. For awhile it was open for testing to the public and was doing a great job. I do not know what is the current status.

Personally I favor an absolute scientific system that can be verified such as the premise of the original computer systems. With modern technology as compared to the computers and systems from 30 years ago (antique tech literally), and how computers nowadays make scans and evaluations of parts down to levels never needed for coin evaluation, its certainly time for a legit change to science in these systems.
How much squash could a Sasquatch squash if a Sasquatch would squash squash?
Bedrock of the Community
United States
17517 Posts
 Posted 06/29/2021  06:01 am  Show Profile   Bookmark this reply Add Conder101 to your friends list Get a Link to this Reply

Quote:
Don't or didn't? I may be mistaken, but I thought all these types of grading system were gone back in the 90s. If this is the case, then it makes sense that the companies said the systems were not accepted b/c people didn;t trust computers.

Also the complexity problem didn't exist yet when the computer grading was scrapped. At that time the computer grading system had only been set up for Morgan dollars and Saint Gaudens double eagles. But even back them I pointed out the problem they were going to have when they extended it back to the pre-steam press coinage where every die was different. They would need a new "map" for every obv and rev die used, and possibly die states of each die as well to allow for die cracks, chips, cuds, and clash marks so as to not confuse these things with "damage".
Gary Schmidt
Pillar of the Community
Learn More...
United States
3555 Posts
 Posted 06/29/2021  08:25 am  Show Profile   Bookmark this reply Add BStrauss3 to your friends list Get a Link to this Reply
Sure, you can get enough examples to train the popular coins. Even there, I suspect you will lack training materials for Morgans outside of MS and PO/Fr/Ag. But less popular coins?

I'm not saying it can't be done in the small - and you are certainly right that the computer power (and imaging system quality which is more important) have made huge strides in the past 5+ years.

Even purpose build hardware isn't as crazy as it was... I can envision a system of four cameras and a laser scanner. one white light, three R G B light and a silicon belt that moves the coin past the cameras. All using identical 65mm macro lenses on a 20MP+ body.

The ultimate problem is getting people to agree on your training set. Even today if you show three coins to three experts, you will get 5-10 grades.
-----Burton
47 year / Life ANA member (joined 12/1/1973)
Life member: Numismatics International, CONECA
Member: TNA, FtWCC, NETCC, OnLinw Coin Club
Owned by four cats and a wife of 37 years (joined 1983)
Pillar of the Community
United States
7447 Posts
 Posted 06/29/2021  11:13 am  Show Profile   Bookmark this reply Add Earle42 to your friends list Get a Link to this Reply

Quote:
They would need a new "map" for every obv and rev die used, and possibly die states of each die as well to allow for die cracks, chips, cuds, and clash marks so as to not confuse these things with "damage".

That also makes sense. This would have been incredibly hard back then. I don't see much of a problem for tech nowadays though.


Quote:
The ultimate problem is getting people to agree on your training set. Even today if you show three coins to three experts, you will get 5-10 grades.

I think you are correct on this. I think this is why Mintstate.com decided to go with slabbed pieces so people who stick to wanting a human hand in it would be satisfied.

Our entire world that surrounds us is made of engineered and QC-passed parts measuring down into the thousandths (and beyond) of tolerances being scanned and verified by machine. I would like to see coins as scientifically handled. I think with the right PR, our world around us, it seems, is always after the next greatest scientific breakthrough and tech toy, so I would think a non-human opinion basis would be welcomed...except there are already a lot of people out there with a lot invested in the label on their slab who would likely lose a lot of money if the coin itself were scientifically and verifiably graded.

I also am of the opinion (nothing to back it up except thinking like a business) that grading companies would not like the systems anyway. They would likely lose a lot of money with people gambling and playing the re-slabbing game for profit. I admit I wonder if this loss of the "gamble aspect" was not also some of the reason the tech was shut down after a short time. Shareholders (if they had them?) would not have appreciate a drop in business. The system would have made a large, but finite and growing smaller, need for grading company services. There are only so many, for example, 1916-D Mercury dimes to be graded.

I see the system (like you described) not only bringing the coin hobby up to a much needed verifiable standard, but consistency. And since an AI system would be based upon a vast sea of human opinion, it might appease those wishing this as well.

BTW - Mintstate.com also has an AI coin detector. Meaning a system could be made to auto-scan, auto-ID, and then auto grade any coin as well.



Quote:
The ultimate problem is getting people to agree on your training set. Even today if you show three coins to three experts, you will get 5-10 grades.


This is exactly why I would love to see verifiable standards. The above is what I personally see as a "gray area" since thousands of dollars can ride on one simple difference in number printed on a label.

It reminds me of when the Beanie Baby market crashed b/c people faced the reality they were paying hard earned cash for plastic beads in a piece of cloth. The Beanie hobby had a totally artificial market-driven "value" assigned to the Beanie Babies. While our coins are different in that some already are known as rare, competition for registry sets etc. keeps driving up prices on the same coin+label+plastic. Its not just about the coin.

The day when a slabbed coin could be broken out, resubmitted, and get the same exact grade every time is when the COIN itself is factually graded on its OWN merits. No more valuable plastic...just coins.
How much squash could a Sasquatch squash if a Sasquatch would squash squash?
Pillar of the Community
United States
1587 Posts
 Posted 06/29/2021  12:13 pm  Show Profile   Bookmark this reply Add jmkendall to your friends list Get a Link to this Reply
A couple of problems with Ai graded systems.

Market grading. Market grading would be gone. Completely. No more days of days of "Daddy big bucks" collector sending in his/her 1945-P Merc and getting an FSB out of a coin of questionable bands. No more downgrading coins to ridiculas grades in order to keep the census low in order to favor a big customer. I've had PCGS grade three of my coins as an AU58; that were later graded at the gem or above level by another TPG. All three were from single digit census'.

Oth how do you factor in "eye" appeal? Some people love toning. Some people love color, but not necessarilly toning in general. How about a "blast white fresh from the mint presses" finish that is generally worth a point all by itself?

I think, as an Engineer, it would not be as hard as some people think. Sure the learning curve would be a bit harsh and cause problems at first, but it is very doable.
Pillar of the Community
Australia
781 Posts
 Posted 06/29/2021  8:36 pm  Show Profile   Bookmark this reply Add nealeffendi to your friends list Get a Link to this Reply
For training a computer AI system it doesn't matter if many coins get graded differently by different graders and get different grades by the same grader on different days. That is because the database will be built up when a few hundred or a few thousand coins graded by humans are added to the database. The AI will have learnt what each grade (including + grades) will look like on average and the average will be very exact (the system can even be trained to learn what eye appeal and attractive toning is). Then the system can point out those coins that are over or under graded by the humans.
That takes care of the grading of 95% of coins submitted, the remaining 5% are the coins with very few submitted (rarities, many world coins) or problematic coins (hammered coins, coins with unusual features).
Then the problem in future will be that all coins graded pre AI will be suspect for grades, I'm sure the TPGs would love to encourage the owners to resubmit the coins (for a fee ) for an accurate grade. They might even nudge the issue by making pre AI graded coins ineligible for set registries.
Pillar of the Community
United States
7447 Posts
 Posted 07/02/2021  5:21 pm  Show Profile   Bookmark this reply Add Earle42 to your friends list Get a Link to this Reply

Quote:
Oth how do you factor in "eye" appeal? Some people love toning. Some people love color, but not necessarilly toning in general. How about a "blast white fresh from the mint presses" finish that is generally worth a point all by itself?


Eye appeal is already in the eye of the human beholder. Look at the large number of coins with high grades that have little eye appeal to everyone.

As to present slabs then "needing" to be re-slabbed...you bet this is exactly what the companies would promote if they would present the AI system.

In fact one of the points they used to push for computer grading in the 90s was that "to err is human."

I am not being negative as much as just stating opinion. I have a feeling back when the computers were being used with a scientific grid system that the companies realized they were shooting themselves in the foot. Making a scientific standard would kill the profitable resubmission gambling game that brings in huge profits.

A business out to make a profit would recognize this and then default to a system to make more profits throughout the years. Eventually they knew it would catch up with them, but the profits already made have kept the companies going strong for decades now.

Computerphobia was gone decades ago. But there are only so many grades to be assigned if a coin is ACTUALLY graded. Its in the best BUSINESS interest of the companies to keep science out of the picture.
How much squash could a Sasquatch squash if a Sasquatch would squash squash?
Valued Member
Canada
154 Posts
 Posted 08/02/2021  8:51 pm  Show Profile   Bookmark this reply Add DouglasFir to your friends list Get a Link to this Reply
This is a truly great discussion,as a relatively new collector, never having bought a slabbed coin, I'm learning a lot, and really enjoying myself.
Pillar of the Community
United States
7447 Posts
 Posted 08/03/2021  11:21 pm  Show Profile   Bookmark this reply Add Earle42 to your friends list Get a Link to this Reply
Glad you came here first before just jumping onto the slab bandwagon. I hope you take a look at the PDF in my signature for a real eye opener as to the problem these companies have with rookie coin collector level attributions of half dollars.

Everything is linked to the PCGS webpage and uses their own data. There is a lot of incompetence there that has literally cost some collectors thousands who have a blind faith in who they consider to be "THE" experts.

These companies have a great wealth of reference information they post for the hobby. They also have a great track record on identifying fakes (also no magic there though - not hard to do if someone takes the time to learn). As former PCGS president Ron Guth has said, "...anyone can learn to grade... ." So while these companies are a convenience for those who (for any reason including legitimate ones) do not take the time to learn fakes or what better grades actually look like, these companies can serve that purpose for them.

But blind faith in them has cost too many people way too much money. Start examining slabs for yourself to see the problems. Tis is how I started out. A former grader told me to start looking at MS70 graded ASEs b/c its not hard to find them that are obviously not perfect coins. It took me very few to find one with an obvious rim ding! From there on I kept looking and, as this forum says, would buy the coin and not the slab.



How much squash could a Sasquatch squash if a Sasquatch would squash squash?
Page 2 of 2   Previous TopicReplies: 21 / Views: 1,995Next Topic  
First Page
Showing last 15 replies.
 
To participate in the forum you must log in or register.





Disclaimer: While a tremendous amount of effort goes into ensuring the accuracy of the information contained in this site, Coin Community assumes no liability for errors. Copyright 2005 - 2021 Coin Community Family- all rights reserved worldwide. Use of any images or content on this website without prior written permission of Coin Community or the original lender is strictly prohibited.
Contact Us  |  Advertise Here  |  Privacy Policy / Terms of Use

Coin Community Forum © 2005 - 2021 Coin Community Forums
It took 0.56 seconds to rattle this change. Powered By: Snitz Forums 2000 Version 3.4.05