Set Weighting Proposal
tradedollarnut
Posts: 20,162 ✭✭✭✭✭
Copied from another thread:
Yes the sets need weighting, but the current system has a huge fault. As I've stated before, my MS68 1875-S (pop1 - one of only 2 MS68's for the series) adds 0.78 to my set rating and an AU58 1873CC (pop 100? Worth about $1500) would add 6.67. This is not right and cannot be fixed by simply adjusting the weights of the dates in general.
I know PCGS wants to keep it simple, but this really should be fixed. Perhaps a "weight bonus" should be added for the highest grades in a series? For example, in trade dollars a weight bonus of 8 for the two MS68's, of 4 for the 6 MS67's and of 2 for the multiple MS66's. This would at least partially correct the above circumstance and would be as simple as the adders for cameo and deep cameo.
For Bust dollars the weight adder might be as low as MS64, for modern proof coinage it might be all the way up at PF70. It would vary by series. Comments?
Yes the sets need weighting, but the current system has a huge fault. As I've stated before, my MS68 1875-S (pop1 - one of only 2 MS68's for the series) adds 0.78 to my set rating and an AU58 1873CC (pop 100? Worth about $1500) would add 6.67. This is not right and cannot be fixed by simply adjusting the weights of the dates in general.
I know PCGS wants to keep it simple, but this really should be fixed. Perhaps a "weight bonus" should be added for the highest grades in a series? For example, in trade dollars a weight bonus of 8 for the two MS68's, of 4 for the 6 MS67's and of 2 for the multiple MS66's. This would at least partially correct the above circumstance and would be as simple as the adders for cameo and deep cameo.
For Bust dollars the weight adder might be as low as MS64, for modern proof coinage it might be all the way up at PF70. It would vary by series. Comments?
0
Comments
My question is, does it really hurt anything to have these minor problems? PCGS has stated that their system is designed to give a fair weighting for coins in attainable grades. The coins that don't get their due are common date pieces in ultra-high grades. In Washington quarters, the 32-D in XF got more points than any of the MS-68's for the series, $10K versus $200. The only way I could see it being problematic is for sets that were very close, but differed only on a couple of pieces. With your set, don't think that you will have any compeition anytime soon.
Keith
I agree with you that finest known examples should also receive "bonus points". How about having bonus points (added to the base weight before calculation) awarded for the top 2 grade levels known for each date/mintmark? These bonuses could be based on the relative population numbers of only the two highest grade levels for each coin versus the rest of the set. In Lincolns for instance (just because I know them better than any other set) late date pop5 or less finest knowns would have large bonuses for MS67RD examples but no bonus or a bonus of only 1 for the relatively common undergrade of MS66RD; while a less common date such as the 1921 would have similarly high bonus values for MS67RD (6 known?) and MS68RD (2 known?), plus the added weight of the date being less common to begin with. I have been working to develop a model of this but my @#%^%$* job has been eating up all of my time lately.
If one were to use a 1-10 conditional rating scale then the registry rating of a pop1 finest known for a common date (weight of 1 plus 10 condition bonus) would outweigh all but the finest known examples of key dates in the series (weight of 10 but no condition bonus for low grade examples). I think that a well designed system of this nature, designed to be flexible with condition bonus points based on fluctuating populations of the highest two grades for each coin, might be the right fit to move beyond the current "baby steps" system (as WC calls it) without becoming too complex to follow and understand (like some other registry system which I will not name).
RELLA
who boasts of twenty years experience in his craft
while in fact he has had only one year of experience...
twenty times.
I don't disagree at all. A lot of my Type set consists of common date MS pieces in higher than normal grades. They don't get rewarded even though a MS-67 may be the highest MS grade and cost 10 times more than a proof counterpart. Instead that $30 PR-69 beats the $300 MS-67 each time. I have e-mailed D Hall a couple of times on the subject, but the impression I got was that this was the weighting that they were going to go with, at least for now.
I would like to see the system improved, but I think that it will have to be gradual. Personally, I want to leave PCGS alone and let them get my active sets weighted before I start complaining again. Then, all heck can break loose.
For example the 26s carrys a weight of 8 and a red bonus of 5 (13 pts) if you buy the one and only at 65 RED you should get the 3 pt bonus (least they can do for you spending 75k).
It would be a delicate balance to prevent the super coins from dominating the registry, yet enough of a "perk" to acknowledge the collector who stepped up to the plate and acquired that coin (IMHO)
peacockcoins
Does the NGC ranking address these issues appropriately or just differently / also flawed?
double post deleted
BUT - DH specifically stated that PCGS is unwilling to consider detailed weighting. So, if they already are using the bonus method (for cam and dcam), logic says they might be willing to use a weighting bonus for common coins in uncommon grades.
Using a 8 weight bonus for the 1875-S in MS68, all of a sudden it adds 6.44 to the set rating instead of 0.78! This is on par with the 6.67 an AU58 1873CC would add, so it still doesn't fix it totally, but hey!, it's better by a factor of ten and it's simple to use. Instead of having to weight each and every grade for every coin for every set, all they have to do is look at each series and decide if a condition bonus for the top few grades is warranted. For example, in trade dollars there are two MS68's graded and 6 MS 67's graded for the whole series - I'd say those two grades warrant a condition bonus. Probably so does MS66. But for other series the grades will be different and some may not have any. Simple, effective and still leaves the keys with a high degree of desirability - the next best thing to perfection!
TDN: Ever try to do it for Jefferson Nickels dated 1938-2000 in all Mint State grades and with and without full steps relative to each other? Roughly 1,000 entries to deal with. Trust me 3 months is a closer estimate than 3 hours
Sounds like your work with Trade Dollars is very neat!! Wondercoin.
Did NGC do some kind of update to their system? I gained 20 percentage points towards Braddick's Kennedy set, but haven't made any changes to mine, and his hasn't changed either.
Keith
It would be like TDN registering his most favorite and rare Trade Dollar and having a collector who has a near complete set of XF/AU Trade Dollars (missing the couple keys) arguing that he should be #1 because he has more coins and has spent a liefetime accumulating the XF/AU lot. Come to think of it, it not like that, it was exactly that sort of "belly-aching" that, imho, led NGC to acquiesce to this "courtesy" for collectors.
My position has always been that the Registry is about the coins, not the collectors behind them. An MS69 silver Wash Quarter should beat out an 83 coin quarter set in MS60 even if it took a collector 2 minutes to raise his hand at auction and buy the MS69 coin and the other collector spent 10 years building the 83 piece MS60 set of common coins. What would most people rather own if they were really honest - Triple Crown Winner Secretariat, or a stable full of mules?
A post from the staff
So what did they actually do?
peacockcoins
The good thing about the registry is that whether you are going for high grades or going for low grade completion, you can still get that "board effect". The majority of the good of the registry can still be found if you threw out the set rating and just concentrated on viewing the compositions of the sets.
The way it is currently set up, the key dates in low uncirculated grades have an extraordinary effect on a set. Whether this is good or bad is a matter of opinion. Certainly in many series there are key dates in circulated that are not key in unc or key dates in unc that are not key dates in gem. IMO, without some sort of adjustment to the weightings as they now exist, the power of the registry will cause an unusual skewing of the prices of these coins and a lowering somewhat of the prices of high grade common dates. Again, some might agree with this and some might not. The only reason I debate the weightings is because I find the current system inaccurate, and I like to be accurate.
So what is the Registry? Is it strictly a means to discover the best set of a series? Is it a way for average collectors to have fun? Is it a marketing ploy to get more coins slabbed? It seems to me that, however it started, both PCGS and NGC are trying to find balance between all three.
I think they are both listening and are doing pretty good in trying to keep as many people happy as possible. They each balance the issues a little differently. It is great fun to play in both courts.
I'm just taking the devil's advocate approach on NGC. I do believe that key date coins in high grade should get the points they deserve, but I sometimes think that they can get too many points and a "wonder" coin can overpower a solid set of coins. A set of XF Trade dollars is not going to be better than any of TDN's pieces. In his set's case, he doesn't have anything but "wonder" coins.
In the case of my MS Kennedy set at NGC, although I am not #1 (and do not deserve to be), I believe that my set is benefiting too much because I have a few key dates in high grade. I think that NGC's system is good in theory because it tries to take into account that a common coin at a high grade may be rare, but I still think that it has a long way to go as well.
Keith
Keith: Not if anyone is listening to me (or TDN) it won't. But, if you poll the forum members, I have no doubt from my readings here that more than a majority of members would state that a "complete set" beats out a "one coin wonder". But, you know me by now - my posts have never been designed to win a "popularity contest" here and, as Braddick points out may very well actually lose customers offended by my frankness. So be it
But, TDN, in his last post, has hit on perhaps the true essence of what the Registy is really about these days. And, remember, the Registry really started out to get that "board effect" on essentially modern collectors (just take a look at 1998, 1999 Registry books) where PCGS saw its growth coming. But, today, colelctors are enjoying Registry sets of Indian Cents, Buffalo nickels, Trade Dollars, etc. What a bonanza for the entire coin industry!! I hear dealers mumbling at shows that the only thing on fire right now in the coin business are "moderns and Registry coins". Well, they haven't figured out yet that by "Registry coins" they are, imho, suggesting nearly the entire US coin market (yes, I actually agree with Coingame 2000 on something)!!
As far as NGC changing it's policy?? Well let's see what happened, some collectors had some concerns about the "one" coin set ranking number 1. A group of collectors debated the pros and cons. A good spirited debate on both side of the issue. NGC listen to these opinions, and then debated the pros and cons between themselves. Decided an adjustment in the ranking system was needed. Nobody force NGC to change its policy, NGC changed it after looking at the issue and deciding a change was warranted. When did spirited debate on issues become "belly-aching"? Are we all "belly-aching" by debating the topic of this thread?
In God We Trust.... all others pay in Gold and Silver!
I think you are on the money,so to speak.Completeing the set is foremost.Then weighting comes next.
Registry 1909-1958 Proof Lincolns
XF 45 x 5 =225
MS-60 x 5 = 300
The MS-60 wins!
Change the weighting Factor to 10
XF 45 x 10 = 450
MS-60 x 10 = 600
The MS-60 wins!
It has nothing to do with weighting factor, it's winning as many individual (slots) battles to pass the other collector. Scaling the Weighting factor by pop is just like weighting by market value. That is the NGC system, do we really want to have 2 Registies that are the same?
In God We Trust.... all others pay in Gold and Silver!