Enough arguing! A Different Look at Grading - Page 4

mikenoodle · 01-05-2009, 12:59 AM

nor are you willing to contribute something to it that might make it so...

**GDJMSP** · 01-05-2009, 01:03 AM

That's because there isn't anything Mike. I disagree with your premise 100%, have said so several times. So how do I contribute something positive to something I know is wrong ?

Sorry, but I don't know how to do that.

mikenoodle · 01-05-2009, 01:13 AM

my Mom always used to say that if you have nothing to contribute, say something nice or say nothing at all.

You could try that...

johnny54321 · 01-05-2009, 02:43 AM

Interesting discussion. Computers not only are getting faster, they also improve with the accuracy and effectivity in handling inputs and outputs. In my mind, this is a big part in what makes them "smarter". I encounter this a lot as an avionics engineer.

I could see computers helping with the grading of the coins if applied right. At the same time, I think it would be a mistake to completely rely on a computer for the grade. Human input is necessary too. Where I see computers helping the most is by detecting problems that may easily be missed by a guy and a 5x loupe. This might be done by analyzing the structure down to the microscopic detail. It could detect the quantity and location of any harmful oxidation levels, and look for the trends commonly found on cleaned coins. There are times when I have examined my coins under a 30x loupe, and have detected problems only after I have photographed it in high res. You guys may nail me for saying this, but I have encountered a fair amount of problem coins residing in NGC or PCGS slabs. I purchased a XF-45 1909-s vdb in an NGC holder at one point that I ended up sending back to the seller because of corrosion. I'd still want an expert to look at the coin personally and make sure things add up; but I think computers can provide a great bit of assistance.

As far as creating a new grading standard, I'm not too sure about that one; but I think things can be improved in the coin market a LOT with the aid of technology if used correctly.

USS656 · 01-05-2009, 01:40 PM

Forget if you believe it can or cannot be done and just think like the technology exists and this is how the technology could be used to improve the current process. People should stop debating one way or the other and just think out of the box on what data they would capture and analyze. FYI - current non contact 3D Measurment systems can be as accurate as .015mm and image acquisition is in the microseconds. If you own a set of digital calipers set them to .015mm and see how small that is. If people approach this with an open mind I think everyone benifts regardless if its possible or not. The less experienced get a lesson on what aspects specifically a person in the hobby at any level shoud be aware of.

**GDJMSP** · 01-05-2009, 11:25 AM

Quote:

Originally Posted by mikenoodle

my Mom always used to say that if you have nothing to contribute, say something nice or say nothing at all.

You could try that...

I don't know how to win Mike - I tried that in the other thread and was ragged on for refusing to continue the discussion

spock1k · 01-05-2009, 11:32 AM

Quote:

Originally Posted by GDJMSP

I don't know how to win Mike - I tried that in the other thread and was ragged on for refusing to continue the discussion

I have a class for that too - how to win and its free now

Lehigh96 · 01-05-2009, 05:20 PM

Quote:

Originally Posted by GDJMSP

I don't know how to win Mike - I tried that in the other thread and was ragged on for refusing to continue the discussion

For the record Doug, I wasn't ragging on you, I just don't understand why coins from different mints and years can't be compared with each other.

And as I recall, that question was never really answered IMO. I am a result oriented person. I don't care if the mint used a lower striking pressure or different die spacing. I care about the end result of the quality of coins they produced. If they use a striking pressure too low in Denver and the coins are not as well struck as they are in Philly, then guess what, they should get lower graded coins for not performing to the same quality level as Philly. Why should we give them a pass by saying, they were all poorly struck and we shall apply a bell curve so that we can have a crappy weak strike MS66.

I am not advocating a whole new grading system, just stricter application of the present one.

**GDJMSP** · 01-05-2009, 09:41 PM

Quote:

Originally Posted by Lehigh96

For the record Doug, I wasn't ragging on you, I just don't understand why coins from different mints and years can't be compared with each other.

And as I recall, that question was never really answered IMO. I am a result oriented person. I don't care if the mint used a lower striking pressure or different die spacing. I care about the end result of the quality of coins they produced. If they use a striking pressure too low in Denver and the coins are not as well struck as they are in Philly, then guess what, they should get lower graded coins for not performing to the same quality level as Philly. Why should we give them a pass by saying, they were all poorly struck and we shall apply a bell curve so that we can have a crappy weak strike MS66.

I am not advocating a whole new grading system, just stricter application of the present one.

The question was answered Paul, numerous times. But you are apparently unwilling to accept the answer. The answer is quite simple - coins from different dates and mints cannot be compared because they are not equal - and you cannot compare two things to each to each other if they are not equal. That is the answer. And that is why we have the system we have.

I have tried to use analogies but that doesn't seem to work. But I'll try one more - can you compare a Cadillac to a Chevy ? A 1957 Chrysler 300 to a 2007 Chrysler 300 ? No, you can't. They are a different as day and night - and you can't compare day to night either.

Now you can take 2 different 2007 Chrysler 300's and compare to each other because they are equal. And you can compare 2 Cadillacs of the same year to each other, or 2 Chevy's to each other - but you can never compare a Cadillac to a Chevy.

Same thing with coins.

And what you are proposing would not be more strict, it would be less strict. For if you were to define what the average strike of a Morgan dollar is - and then grade all Morgans based on that - what you would find out is that all those 65's would become 66's and 67's. And we'd be buried in 68's. For the average strike of a Morgan dollar is really quite poor.

And this bit about computers, people think machines can do a lot more than they really can do. But some of the simplest thing there are - a computer cannot do. For example - to grade coins a computer would have to be capable of complex pattern recognition now wouldn't it. But do you know that even the most advanced computer when using patter recognition cannot even tell a B from an 8. It is known as the B/8 problem.

So how in blue blazes is a computer ever going to be able to grade a coin when it cannot do something that a 4 year old can do in his sleep !

Grading coins requires the ability to think - computers cannot think. They can only compute.

USS656 · 01-05-2009, 10:25 PM

[quote=GDJMSP;494029]And what you are proposing would not be more strict, it would be less strict. For if you were to define what the average strike of a Morgan dollar is - and then grade all Morgans based on that - what you would find out is that all those 65's would become 66's and 67's. And we'd be buried in 68's. For the average strike of a Morgan dollar is really quite poor.[/qoute]

Good discussion - why do you think we would be burried in 68's? Are you assuming that the strike alone would drive the grades up? Wondering whats driving this specific train of thought.

Quote:

Originally Posted by GDJMSP

And this bit about computers, people think machines can do a lot more than they really can do. But some of the simplest thing there are - a computer cannot do. For example - to grade coins a computer would have to be capable of complex pattern recognition now wouldn't it. But do you know that even the most advanced computer when using patter recognition cannot even tell a B from an 8. It is known as the B/8 problem.

Doug - we are not talking about Best Buy electronics here - we are talking about state-of-the-art imaging systems that could record and measure down to the .015mm in microseconds. Systems like these are used to reverse engineer the most complex mechanical designs.

But again, I think you are limiting your thinking to what you think is possible instead of what may be possible. Please - Imagine if anything was possible what would you narrow in on??? That's what this thread should be about and what we need for people that have such a wealth of knowledge about this hobby (like you or anyone else that feels they want to contribute) to provide that type of input. Don't assume anything including how the data has to be used to form a grade.

spock1k · 01-06-2009, 12:23 AM

Quote:

Originally Posted by GDJMSP

The question was answered Paul, numerous times. But you are apparently unwilling to accept the answer. The answer is quite simple - coins from different dates and mints cannot be compared because they are not equal - and you cannot compare two things to each to each other if they are not equal. That is the answer. And that is why we have the system we have.

I have tried to use analogies but that doesn't seem to work. But I'll try one more - can you compare a Cadillac to a Chevy ? A 1957 Chrysler 300 to a 2007 Chrysler 300 ? No, you can't. They are a different as day and night - and you can't compare day to night either.

Now you can take 2 different 2007 Chrysler 300's and compare to each other because they are equal. And you can compare 2 Cadillacs of the same year to each other, or 2 Chevy's to each other - but you can never compare a Cadillac to a Chevy.

Same thing with coins.

And what you are proposing would not be more strict, it would be less strict. For if you were to define what the average strike of a Morgan dollar is - and then grade all Morgans based on that - what you would find out is that all those 65's would become 66's and 67's. And we'd be buried in 68's. For the average strike of a Morgan dollar is really quite poor.

And this bit about computers, people think machines can do a lot more than they really can do. But some of the simplest thing there are - a computer cannot do. For example - to grade coins a computer would have to be capable of complex pattern recognition now wouldn't it. But do you know that even the most advanced computer when using patter recognition cannot even tell a B from an 8. It is known as the B/8 problem.

So how in blue blazes is a computer ever going to be able to grade a coin when it cannot do something that a 4 year old can do in his sleep !

Grading coins requires the ability to think - computers cannot think. They can only compute.

hey i got an idea. now you and me agree that a 2008 P and a 2008 D cent are not the same. but what about the different dies ( die replacement) used in 2008 P in theory those should not be the same either. we recognize this when something silly happens like 1955 doubling 1922 no D etc but what about in other cases. i am studying some coins right now where there are 3 bangles but due to die wear they come together and become a bracelet which fundamentally alters the design. so what are your 2 cents

Lehigh96 · 01-06-2009, 12:54 AM

Quote:

Originally Posted by GDJMSP

The question was answered Paul, numerous times. But you are apparently unwilling to accept the answer. The answer is quite simple - coins from different dates and mints cannot be compared because they are not equal - and you cannot compare two things to each to each other if they are not equal. That is the answer. And that is why we have the system we have.

I have tried to use analogies but that doesn't seem to work. But I'll try one more - can you compare a Cadillac to a Chevy ? A 1957 Chrysler 300 to a 2007 Chrysler 300 ? No, you can't. They are a different as day and night - and you can't compare day to night either.

Now you can take 2 different 2007 Chrysler 300's and compare to each other because they are equal. And you can compare 2 Cadillacs of the same year to each other, or 2 Chevy's to each other - but you can never compare a Cadillac to a Chevy.

Same thing with coins.

And what you are proposing would not be more strict, it would be less strict. For if you were to define what the average strike of a Morgan dollar is - and then grade all Morgans based on that - what you would find out is that all those 65's would become 66's and 67's. And we'd be buried in 68's. For the average strike of a Morgan dollar is really quite poor.

And this bit about computers, people think machines can do a lot more than they really can do. But some of the simplest thing there are - a computer cannot do. For example - to grade coins a computer would have to be capable of complex pattern recognition now wouldn't it. But do you know that even the most advanced computer when using patter recognition cannot even tell a B from an 8. It is known as the B/8 problem.

So how in blue blazes is a computer ever going to be able to grade a coin when it cannot do something that a 4 year old can do in his sleep !

Grading coins requires the ability to think - computers cannot think. They can only compute.

Doug,

You absolutely can compare a Cadillac to a Chevy. The Caddy is better which explains why it costs more money. Regarding coins from different years and mints, IMO, they are equal. They have the same design and should all look the same. They only look different because one mint did a better job producing the coins than the other. You call it apples to oranges (or other apples), I call it accountability. I can look at a Peace Dollar and make in independent judgment on the quality of the strike regardless of the mint or date.

I respect your opinion Doug, but in this case you are making this much more difficult than it really is. What I am saying is simply compare all of the strikes from every date and mint to determine what the weak, average, and good strikes look like for the series. If a particular date and mint produced weak strike coins, the grades of those coins would be correctly limited to the lower MS grades. It really is not that far of a stretch. In fact, if they graded that way in the beginning, you would find people who would vehemently oppose grading strike independently for each year and mint.

Furthermore, just because the so called experts think this is the best way to grade strike does not mean that they are correct. Everyone with an opinion should have the opportunity to have there opinion heard and actually considered before being summarily dismissed.

We completely understand that you disagree with us on this issue, but by no means does that make you (or us) right.

davidh · 01-07-2009, 10:06 PM

Quote:

Originally Posted by GDJMSP

The question was answered Paul, numerous times. But you are apparently unwilling to accept the answer. The answer is quite simple - coins from different dates and mints cannot be compared because they are not equal - and you cannot compare two things to each to each other if they are not equal. That is the answer. And that is why we have the system we have.

And what you are proposing would not be more strict, it would be less strict. For if you were to define what the average strike of a Morgan dollar is - and then grade all Morgans based on that - what you would find out is that all those 65's would become 66's and 67's. And we'd be buried in 68's. For the average strike of a Morgan dollar is really quite poor.

I'm struggling mightily to understand this concept. It sounds like you're saying that it would be impossible to grade a coin without knowing its date and mint. For instance, could you grade the 1922 and 1924 Peaces that Lehigh showed in his post if the dates were obscured? If yes, what grades would give them? If not, why not?

To me it's ludicrous to think that the date is an important and deciding part of grading (assuming the date isn't damaged, or at least not any more that the other parts of the mottoes and devices).

Maybe since I'm looking at this from the viewpoint of a type collector, my thoughts are a lot different from series collectors. The date has no real meaning to me. I simply try to obtain the best representative coin of each type and I couldn't care less what the date or mint is.

mark_h · 01-05-2009, 11:21 PM

I guess what I don't understand with this whole thread is why it needs to change? No matter what system you come up with the buyer will still decide the final grade and the price they are willing to pay. So the seller(or the tpg) can call the coin anything he wants - as a buyer I set my grade and the price I will pay. Why not make it really simple - all coins are genuine or fake then the buyers determine the price? Why even have a grade? We might as well take out varieties as well - make everything equal. If you are trying to get to a buy sight unseen type system - it won't work. Nobody but me gets to determine what I consider a choice coin. Me - I will stick with ANA grading guides. Yes - my interpretation of it, not perfect but it makes me happy. This is the best comparison I can think of - Paul likes toned coins - great for Paul, but a few of them are way over the top for me. Personally I would not have put them in a PCGS or NGC holder - cleaned or AT. But that is only my opinion and it does not make it fact. Even if you proved to me the coins are NT I still would not purchase a few of them. That is an opinion that can not be built into any grading system you come up with. I still would not buy some of them. That is my opinion and my choice.

I mean, we the collectors can't get SGS shutdown, the pocket change MS70's graders - even with a new system they can continue to say "we use ANA standards" these coins are MS70. Then as long as people buy them they are doing nothing wrong. But then when they go to sell them - we the buyers - will demonstrate by the price we pay the mistake they made in buying SGS coins. So what? All we can do is try to educate someone?

I just can't see what the problem is - I mean Doug is allowed his opinion, Paul, Mike(s), Frank(s) and even me. We do not need to agree on a grade eye appeal, cleaning, price or worth of a coin. As potential buyers each of us determine what we will pay. And yes - I may argue with doug or someone else, but we can always agree to disagree. Simple as that. I think sometimes people read too much into a posts tone.

Maybe someday computers can do some technical grading, but they will never replace my eye. Like others I make my living in the industry and I do not think anywhere down the line will computers have a final say in this hobby - human collectors will. You can not convince me otherwise or even convince me to use open source for business critical needs. That should stir up a hornets nest in general discussion if someone started a thread on it(that is if you made it this far). I can not see someone developing a "technology based grading system" that everyone would agree to use. You will still have the SGS's of the world doing there own thing.

All of that to say - "I can live with the way things are now". I still like the discussion.

**The_Cave_Troll** · 01-08-2009, 01:39 AM

Quote:

Originally Posted by mikenoodle

my Mom always used to say that if you have nothing to contribute, say something nice or say nothing at all.

You could try that...

lol, I would say being honest with you was plenty nice. If you want to fight a windmill then that is your Quioxte given right, but if someone is kind enough to point out that it's a windmill you're fighting, they aren't being unkind, just helpful.

Welcome to Coin Talk! Register Now, it's easy and FREE!