How much does a one grade difference make in the registry?

[Oldtrader3]

If lowering the barrier for entry is what the NGC people are doing, then how can a 60% completed Peace dollar set (with a mean grade of 65.1) that is missing 6 key coins, be rated higher than my set which is 100% completed with all key coins (with a mean set grade of 63.9)? The arithmetic says (mean times % completed) otherwise.

 

This is all beyond my ability to rationalize this Registry system. Can you give me a clue?

[Architecht]

Because our system is based on relative rarity, higher grades (and therefore rarer coins) are valued more based on how scarce they are. In many cases, a one point grade-jump adds significant points to a set because there are few of them available at that quality level.

 

Our system adds less of a premium for those coins than the market does in dollars, but they are still valued more than a "single point". What we tried to do was capture the "desirability" of a coin apart from its monetary value. For instance, a grade that seems very obtainable in a 1999 coin might be very scarce and coveted in a 1909 coin. Our system captures this fact.

 

Taking a straight accounting of "grade" ignores how rare those grades are. Taking a straight accounting of the monetary value of a coin, however, shows an exaggerated view of the "desirability" of a coin as the scale goes up.

 

What we have done is apply a system to develop scores for coins that acknowledges grade, rarity, market prices, etc., but delivers a final number that moderates these amounts - coming closest (we feel) to capturing the "gut feel" of what collectors give the highest "whoah cool!" factor to. The truth is that in many series an MS66 is much more exciting than an MS64, which is much more exciting than an MS62. More so than "4 points" of grade difference might indicate.

 

If you think about it, there's what, 22 grades possible in coins (hopefully I counted right). That means even without taking rarity into account, a one grade difference is equivalent to 4.5% of the possible scale - not merely "one point".

 

A 4.5% premium might be a fair number IF, in fact, there was an equal number of coins in existence at each grade. But there aren't. There are far fewer 68's than 67's for example. The market makes that fact abundantly clear.

 

So basically, what you'll see in our system is that in comparing one coin to another, a one point grade difference might mean alot or a little depending on the rarity of those grades. At first glance it might sound mysterious, but really, it's exactly the opposite. Look at any market price guide - this will give you a rough idea of how big a difference there is between coins A and B and which one is better. But know that while the ORDER of which coin is better is fully preserved, the jump in value as grades and rarity go up is not as severe in our system as it is in the market.

 

And be assured that our ranking system is every bit as specific as any price guide or population report. Coins are scored according to each type and variation of coin as well as grade, condition, and (with NGC's star grades) even by things like eye appeal - our system does not merely weight slots or conditions generically.

 

This is what we mean by relative rarity.

 

Arch

[Fairlaneman]

Arch: Thanks for the Explanation. Well I wonder Who Posted to this First.The Person has No Name ??Take Care: Ken

[Weidel Legacy Collection]

Arch -

Great response! I gotta ask, do you clear your throat before pontificating so much in a single response! <ducking and running>

[braddick]

Great response Arch- That answers many of my questions and solidifies my faith in the NGC Registry even more.

[Architecht]

 

(clears throat -ahem-)

 

No.

 

Only for short responses.

 

Arch

[Architecht]

I moved this post from deep in a thread on the certified registry forum to this one, as it seemed worth preserving for those who might happen along later and have the same type of questions.

 

Arch

[EVillageProwler]

It's occurred to me that I have no real idea how the Registry Set ranking formulae works. Either here or at the other place.

 

When I build my collection, I really don't give much thought to the Registries. I'm normally a very competitive person, but just not in this case. I've had conversations with TDN and Keoj where I've felt that they thought I was clued in on all this. I am such a fraud. I just have nice coins, and I collect them mostly in sets, and so they'll perform well as a result.

 

I guess someday the digithead in me will learn all about this, and then I'll see if I can write a program that will allow me to calculate my approximate ranking advantages on the fly.

 

Imagine the benefits of having this program on a handheld, walking around at shows and punching in numbers and dates. Boy, would I be the ultimate in the ``buy the holder, not the coin'' game!

 

 

EVP

 

[lazeroo]

well EVP,

 

I'm adding one high grade lincoln everyday to the 1909 to date set to see if I can make sense of this also.

 

I was quite impressed at the jump in points for the first three coins. then I thought bummer man, this is only registering the % of the highest set. get it ? so then I thought I'd try and figure out %'s of the coins to his set. now i'm really cornfused. .

 

I'll keep entering coins for awhile to see if there's some logic in this but i'm afraid unless someone has the finest set to compare it to I won't be able to really figure out the points.

 

hope this make sense to you.

 

rob.

[dragon]

I can tell you guys this, as far as the weightings go for the Morgan dollar series, the NGC system is light years better than the PCGS weighting system, NO COMPARISON.

 

Dragon