Roger Burdette's Saint Gaudens Double Eagles Book 4 4

[GoldFinger1969]

3 hours ago, Quintus Arrius said:

If the price of gold were to rise significantly, gold coins including your beloved Saints will be loosed upon the market.  And if even a fraction undergo authentication and encapsulation, I would expect that increased supply would affect demand.

Yeah, rising gold seems to matter here and there.  Of course, if the coin has numimsatic value the person should have turned it in before the rise since it's worth potentially alot more than 1 ounce of gold.

3 hours ago, Quintus Arrius said:

I know nothing about the population of Saints but would it be unreasonable to assume where formerly two existed at a certain grade, they may lose value if seven more were to suddenly become available? And suppose a few graded higher where formerly none had been previously certified at that level?

If you had a condition rarity of 2...and then 7 more hit the market....LOOK OUT BELOW !!! 

The SAINTS book has numerous examples of coins that were once super-rare, but the hoards changed the rarity factor.  Look at my post on the rarity rankings on Page 31....the 1926-D was considered the rarest coin besides the 1933 (which was being confiscated and as we know is basically unobtainable).  But while there are just under 500 coins in all grades, there are ONLY 4 coins MS65/66 and nothing higher.

The 1926-S has just under 1,800 coins total, with 45 MS65/66's and 1 MS67/68.  The 1924-D has just under 1,200 coins but only 18 MS65/66 and over half the total circulated.  These coins did NOT get super-high grade coins back via hoards, but they did get lower MS and Uncirculated coins back from Europe.

OTOH, look at the 3rd most rare grouping and note the 1927-D.  It's now THE RAREST aside from the 1933.  The expectation was that there were tons of them out in Europe -- there weren't.   So while other coins slid down the ranking scale, the 1927-D moved to the top.

If you found some elderly Denver residents who had even 2 or 3 1927-D Saints in Mint State 65 or higher, given the existing base of 20 coins with half MS65 or higher, you'd drop the price by 10-20% from the $2,000,000 level or so.  10 coins would crater the market by 50% I would venture.

BTW, I was at the FUN 2020 auction where the Duckor 1927-D MS65+ was sold for just over $2 MM. 

3 hours ago, Quintus Arrius said:

I would be curious to know whether the reintroduction of the Liberty design on the obverse of the newer bullion coins spurred an interest in the subject matter or only intensified it.  I was going to buy one until I discovered the UHR design necessarily reduced the diameter of the original.  This fast-moving, ever-evolving thread is one of my favorites.

The liberty design on the yearly-issued American Gold Eagle has been there since 1986.  I'm unaware of any change or retintroduction on "newer bullion coins" but then again I only have 1 or 2 I think of the modern AGEs.

The UHR was a separate coin and very nice -- just small. 

Glad you like the thread, it has been a great one to read and contribute to.   

Edited April 3, 2021 by GoldFinger1969

[GoldFinger1969]

Roger, I don't think I asked this before (not sure I could find it in the previous 31 pages anyway ):  how did you arrive at the population census figures for the Saints for each year/mintmark ? 

Did you start from scratch and use the PGCS and NGC figures as of 2016 or 2017 ?  Did you incorporate the estimates from Bowers or Akers' books as a starting point and work forward from there ?  Other materials ?

Finally, did you "top off" any near-final estimate numbers, especially for the high-end MS numbers, maybe after talking with HA folks or other contacts who may have said "we think there's another 4 coins in MS66 for (fill-in-the-blank rare Saint) that aren't in the market yet" ?

Edited April 4, 2021 by GoldFinger1969

[Henri Charriere]

[I cannot quote book and verse, but somewhere in the Holy Scriptures (maybe Numbers) is a quote from a pre-eminent expert on the subject of populations and censi, VKurtB, who cautioned in substance:  "one is not to place one's trust in the accuracy of such things."]  😉

[GoldFinger1969]

33 minutes ago, Quintus Arrius said:

[I cannot quote book and verse, but somewhere in the Holy Scriptures (maybe Numbers) is a quote from a pre-eminent expert on the subject of populations and censi, VKurtB, who cautioned in substance:  "one is not to place one's trust in the accuracy of such things."]  😉

Oh, they're guestimates at best, QA.....with all the re-submissions, you don't know how inflated the numbers are though I suspect the market price of a coin would be much higher/lower if the population census was way off.

Someone once threw out 20-30% for double-counting in the numbers but they may have just made it up.

Unfortunately, unless the 2 leading TPGs coordinated pictures of specific coins and the population census we'll never have super-accurate numbers like say 90-95% accuracy.

[RAJ on COINS]

Slightly off topic, but curious as to Roger's take.  Is there any reason to believe that any Saint's have been melted in the last 20 years, and is there any scenario by by which it would make (economic) sense for any to be melted in the future?

The only case I can imagine would be if the metal price spiked because of some gold based medical or technological breakthrough, increasing demand for the raw metal above numismatic premiums.  Some gold based cancer drug, or space/energy technology related breakthrough for example.  Otherwise, I have a hard time believing that anyone would melt any, and yet I think the hobby might benefit if some 100,000 lower grade 1924's or 1908's hit the melting pot.  Going to coin shows and seeing so many "bullion" date coins makes Saints look anything but rare.  While they are at it, we do with fever American Eagles as well.   Any thoughts?

   

[RAJ on COINS]

Just now, Ross J said:

While they are at it, we do with fever American Eagles as well.

   

What I meant to "type" was...

"While they are at it, we could do with fewer American Eagles as well."

[Henri Charriere]

1 hour ago, GoldFinger1969 said:

Oh, they're guestimates at best, QA.....with all the re-submissions, you don't know how inflated the numbers are though I suspect the market price of a coin would be much higher/lower if the population census was way off.

Someone once threw out 20-30% for double-counting in the numbers but they may have just made it up.

Unfortunately, unless the 2 leading TPGs coordinated pictures of specific coins and the population census we'll never have super-accurate numbers like say 90-95% accuracy.

I appreciate the fact you understand my sense of humor and accept your well-reasoned response completely and unequivocally.  

[Henri Charriere]

4 minutes ago, Ross J said:

What I meant to "type" was...

"While they are at it, we could do with fewer American Eagles as well."

The likelier scenario is mining operations that have lain fallow.or dormant, would spring back to life when the economic equation favors them.

It is true there is a great deal of metals on the market but I believe like anyone else that there are interests beyond ours that effectively regulate the markets in metals, diamonds and oil -- and a group of experts speculating on it all.

 

[RWB]

2 hours ago, Ross J said:

Is there any reason to believe that any Saint's have been melted in the last 20 years, and is there any scenario by by which it would make (economic) sense for any to be melted in the future?

Most of the coins are simply metal. When commercial need exceeds available supply, it can be a good business decision to melt any reserve supplies to keep the business operating. An individual who had purchased a quantity of routine DE during the 1960s-70s, could easily convert these "cheap" coins into cash and not report the capital gain. (Ignoring the foolishness of losing a much greater profit just by buying a mutual fund at that time.)

I did some consulting for a company that had built up a large stockpile of rare earth elements in anticipation of future technological uses. Gold is not different, except its not a useful.

[GoldFinger1969]

6 hours ago, Ross J said:

Slightly off topic, but curious as to Roger's take.  Is there any reason to believe that any Saint's have been melted in the last 20 years, and is there any scenario by by which it would make (economic) sense for any to be melted in the future? The only case I can imagine would be if the metal price spiked because of some gold based medical or technological breakthrough, increasing demand for the raw metal above numismatic premiums.  Some gold based cancer drug, or space/energy technology related breakthrough for example.  Otherwise, I have a hard time believing that anyone would melt any, and yet I think the hobby might benefit if some 100,000 lower grade 1924's or 1908's hit the melting pot.  Going to coin shows and seeing so many "bullion" date coins makes Saints look anything but rare.  While they are at it, we do with fever American Eagles as well.   Any thoughts?   

Good question....I think I posted this elsewhere.....how do we know that lots of bullion Liberty and Saint Double Eagles aren't owned as bullion (maybe at a slight discount since they're not as convenient as gold bars) by ETFs or other similar funds ?

Some of thee closed-end funds and ETFs have to hold physical gold...I don't think it matters if it is a gold nugget, gold bar, or gold coin.  Ditto Private Banks that hold it directly for wealthy clients.

Just remembered another big potential buyer....CENTRAL BANKS !!  

Edited April 5, 2021 by GoldFinger1969

[GoldFinger1969]

4 hours ago, RWB said:

I did some consulting for a company that had built up a large stockpile of rare earth elements in anticipation of future technological uses. Gold is not different, except its not a useful.

Rare earth price moves were a big thing a few years ago....then they settled down....making a comeback with the growing conflict with China focusing interest on national security, supply chains, microchip scarcity, etc.

I suspect you might have more consulting jobs in the near future. 

[RWB]

9 hours ago, Quintus Arrius said:

somewhere in the Holy Scriptures (maybe Numbers) is a quote

This is in the "Book of Fractions." it originally came after the "Book of Numbers," but was removed during the First Council of Nicaea [pronounced: Ne ky 'a] (325 CE). when none of the Bishops could separate "numerator" from "deuteronomy" and they agreed to call the whole thing off.

[Henri Charriere]

On 4/4/2021 at 8:58 PM, RWB said:

This is in the "Book of Fractions." it originally came after the "Book of Numbers," but was removed during the First Council of Nicaea [pronounced: Ne ky 'a] (325 CE). when none of the Bishops could separate "numerator" from "deuteronomy" and they agreed to call the whole thing off.

Always make my day!  😉

[GoldFinger1969]

On 4/3/2021 at 11:07 PM, GoldFinger1969 said:

Roger, I don't think I asked this before (not sure I could find it in the previous 31 pages anyway ):  how did you arrive at the population census figures for the Saints for each year/mintmark ? 

Did you start from scratch and use the PGCS and NGC figures as of 2016 or 2017 ?  Did you incorporate the estimates from Bowers or Akers' books as a starting point and work forward from there ?  Other materials ?

Finally, did you "top off" any near-final estimate numbers, especially for the high-end MS numbers, maybe after talking with HA folks or other contacts who may have said "we think there's another 4 coins in MS66 for (fill-in-the-blank rare Saint) that aren't in the market yet" ?

Roger, in case you missed this.  No rush on a response, just curious.

[RWB]

The population estimates are the product of a simulation model built on multiple variables including: pieces destroyed from multiple sources, statistical validation and adjustment of TPG populations, behavioral patterns (such as tracing extant specimens; appearance ratios) and adjustments recommended by several academic people who regularly deal with population estimates (and validation of estimate) from incomplete data sets. Available hard data was/is always the starting point, and the model then expanded and adjusted to accommodate the gaps. A clear example occurs in the 1930s dates where we have exact quantities distributed and melted.

After the estimates were completed for all dates/mints, David Akers' work was reviewed to see if there was at least a little correlation from his business-derived estimates and my data-based modeling approach. I then attempted to explain notable differences and select the most plausible final estimates. Fortunately, I had access to much larger and more reliable data sets than Akers, but unfortunately lacked clear information about how he really arrived at his approximations. I did not consult Bowers at all - being largely derivative and more indirect than presumed for Akers.

TPG data began as early as possible - mostly around 1990. This helped identify the effects of resubmissions, crossovers and grade instability. My grade "bins" in the book are designed to minimize instability in what is inherently an opinion-centric system.

[GoldFinger1969]

Fantastic, what detail you put into it.  Thank You !  

[GoldFinger1969]

5 hours ago, RWB said:

The population estimates are the product of a simulation model built on multiple variables including: pieces destroyed from multiple sources, statistical validation and adjustment of TPG populations, behavioral patterns (such as tracing extant specimens; appearance ratios) and adjustments recommended by several academic people who regularly deal with population estimates (and validation of estimate) from incomplete data sets. Available hard data was/is always the starting point, and the model then expanded and adjusted to accommodate the gaps. A clear example occurs in the 1930s dates where we have exact quantities distributed and melted.

Fascinating....off the top of your head, aside from super-low population numbers like for the 1927-D (where we can ID individual coins) or the very low numbers for the 1930's Saints.....would you agree that combining PCGS plus NGC population numbers for most Saints leads to an overcounting of about 20-30% ?  Higher ?  Lower ?

If anybody here has had any talks with PCGS or NGC or dealers who have thoughts on the extent of double-couting, chime in.  I would think it is numerically logical that double-counting for 1930-S Saints will probably be much much less percentage-wise than that for 1923-D's or 1924's. 

But maybe not so ?

5 hours ago, RWB said:

After the estimates were completed for all dates/mints, David Akers' work was reviewed to see if there was at least a little correlation from his business-derived estimates and my data-based modeling approach. I then attempted to explain notable differences and select the most plausible final estimates. Fortunately, I had access to much larger and more reliable data sets than Akers, but unfortunately lacked clear information about how he really arrived at his approximations. 

Very interesting.  You basically didn't just improve on earlier estimates, you used modeling to build upon earlier estimates.

5 hours ago, RWB said:

TPG data began as early as possible - mostly around 1990. This helped identify the effects of resubmissions, crossovers and grade instability. My grade "bins" in the book are designed to minimize instability in what is inherently an opinion-centric system.

Both NGC -- and especially PCGS -- see crossovers.  Initially, back in 1986-90, ALL of the coins/Saints they got were 1st-tier submittances, raw coins.  I don't know when cracking-out began (mid-1990's ?) but it gradually picked up in the 2000's.  But by now, the number of raw coins to previously graded coins should be near all-time lows.  Most super-rare, expensive coins should have been turned in for grading (I would think).  Maybe not for generic Saints because grading is not going to mean much monetarily.

Thanks for the feedback, Roger. 

Edited April 8, 2021 by GoldFinger1969

[RWB]

11 hours ago, GoldFinger1969 said:

You basically didn't just improve on earlier estimates, you used modeling to build upon earlier estimates.

Not quite. Models were built and tested against data and did not involve use of any earlier guesses or estimates. Comparisons with previous estimates - published and private - occurred only after the model results were complete.

Also, TPG populations were lower reliability data than some other sources; they also had to be adjusted for resubmission and crossover events whose ratios changed over time. In effect, TPG population data decreases in reliability and validity as one approaches the present. It is like starting with thick, rich bean soup, then adding water hour after hour. Eventually one has runny bean broth - the beans remain, but are harder to find in the excess liquid.

Edited April 8, 2021 by RWB

[GoldFinger1969]

1 hour ago, RWB said:

Not quite. Models were built and tested against data and did not involve use of any earlier guesses or estimates. Comparisons with previous estimates - published and private - occurred only after the model results were complete.

Understood, thanks ! 

1 hour ago, RWB said:

Also, TPG populations were lower reliability data than some other sources; they also had to be adjusted for resubmission and crossover events whose ratios changed over time. In effect, TPG population data decreases in reliability and validity as one approaches the present. It is like starting with thick, rich bean soup, then adding water hour after hour. Eventually one has runny bean broth - the beans remain, but are harder to find in the excess liquid.

I'm sure you spoke to higher-ups at both PCGS and NGC (let alone HA).  Did you find their data and anectdotal information on crackout numbers (especially PCGS, since they were the preferred destination for a long time) useful or did you take it with the proverbial grain of salt ?

[RWB]

Empirical data and anecdotes were treated separately as is consistent with their information quality.

[GoldFinger1969]

6 minutes ago, RWB said:

Empirical data and anecdotes were treated separately as is consistent with their information quality.

Gotcha....finally... adding up PCGS and NGC population numbers.....20-30% overcounted ?  More ?  Less ?  I realize it varies for each year and mintmark, just wondering if that is a good ballpark in general.

I saw that 20-30% figure but can't recall if it was just for Saints or for all coin populations with all the re-submittances leading to double-counting.

Edited April 8, 2021 by GoldFinger1969

[RWB]

4 minutes ago, GoldFinger1969 said:

Gotcha....finally, adding up PCGS and NGC population numbers.....20-30% overcounted ?  More ?  Less ?  I realize it varies for each year and mintmark, just wondering if that is a good ballpark in general.

I saw that 20-30% figure but can't recall if it was just for Saints or for all coin populations with all the re-submittances leading to double-counting.

I don't recall there being any consistency....one guess is as good as another on this point. The reports for the last few years are bloated with cumulative authentication events and nearly useless for my purposes.

Edited April 8, 2021 by RWB

[VKurtB]

1 hour ago, RWB said:

I don't recall there being any consistency....one guess is as good as another on this point. The reports for the last few years are bloated with cumulative authentication events and nearly useless for my purposes.

This is what I have been saying consistently for well over 30 years now, and time has not improved the situation. Pop. reports are WILDLY inaccurate, and to the extent they have value at all, it might be as a testament to the crack out culture. Having spent several months in Alabama now, and finding it a numismatic wasteland compared with Pennsylvania, I am left again stressing how much places like central Pennsylvania disprove the “if it’s worth slabbing, it already has been slabbed” meme. High quality coins there in PA literally ABOUND outside of plastic tombs. Yes, there’s some mishandled gar-BAGE too, but it is the mother lode of valuable “raw” coins as well.

Every major coin show in Pennsylvania invariably produces significant amounts of “new to the market” material, some of it walking in on “free to the public” day. The amount of recently good but recently improperly cleaned material is enough to break your heart.

Edited April 8, 2021 by VKurtB

[RWB]

To be clear - an "authentication event" occurs when a TPG receives a coin, medal or other object. This is the raw count of items submitted. Following submission, several actions occur which separate damaged, defective, counterfeit and other ungradeable or limited-grading, from graded coins.

[GoldFinger1969]

2 hours ago, VKurtB said:

This is what I have been saying consistently for well over 30 years now, and time has not improved the situation. Pop. reports are WILDLY inaccurate, and to the extent they have value at all, it might be as a testament to the crack out culture.

Kurt, if the numbers were way out of line by adding PCGS + NGC -- and then making an estimate for double-counting (maybe my 20-30% guestimate).....wouldn't the price of the coins be dramatically different if the raw PCGS+NGC number (with or without the haircut for double-counting) were WAY off ? 

I agree the numbers may be "off" -- but they still have value.  Question is, how much ?  And if the numbers are off for the 1924 Saint in various grades, does it matter as much if the total numbers (and grades) are off for the 1927-D or 1930-S (coins with dramatically lower population numbers across the board) ?

Edited April 8, 2021 by GoldFinger1969

[RWB]

For rare pieces, one coin represents a significant portion of the known population. For common pieces, 1,000 examples might be a trivial quantity. Thus, piece-count errors in the TPG populations are of very limited utility.

[GoldFinger1969]

The Amazingly Stable 1921 Saint:  Earlier, I talked about how Saints had shifted over the years in rarity rankings because of hoards.  The 1921 Saint is unique in that it's ranking hasn't changed much in 80 years.  There's a decent amount available in circulated and low-60's grades, but only 4 coins at MS64 and above with NOTHING at MS67 or higher. 

Roger's book has these comments from Akers:

"...The standing of the 1921 in the overall hierarchy of Saint-Gaudens double eagle rarities has changed less over the last seven decades than any other regular issue in the series. During that time, some issues have dropped precipitously from their place at the top (1924-S and 1926-S for example) and others have risen substantially (1920-S, 1930-S and especially 1927-D) but the 1921 has always been recognized as being among the top four rarities of the series, both 70 years ago and today, at least with respect to value. The only thing that has changed is the other three coins with it at the top. The 1921 is now considered to be the second most valuable regular issue Saint-Gaudens double eagle, surpassed only by the 1927-D whose extreme rarity was not recognized fully until the 1950s, at least in comparison to other issues in the series. Judged solely on its population rarity, meaning the total number of specimens known in all grades, the 1921 is certainly rare, but not exceptionally so, comparable overall to the 1920-S, but actually less rare than the 1930-S and 1932. However, as a condition rarity it is the unrivaled "Queen" of the Saint-Gaudens series because the condition at which it becomes extremely rare and valuable is lower than for any other issue. Of course, every Saint is a condition rarity at a certain level. For example, any issue is (or would be if one existed) a great rarity in MS68 or 69. For some issues MS67 is the rarity point, for others it is MS65 or MS66. But no issue, not even the 1927-D, is as difficult to locate in MS64 or higher grades as the 1921. Only four or possibly five specimens are known in the MS65 and MS66 grades combined with nothing finer. Even in the MS63 and 64 grades, the 1921 is a major rarity with no more than 12-15 examples known of those two grades combined."

[VKurtB]

19 hours ago, GoldFinger1969 said:

Kurt, if the numbers were way out of line by adding PCGS + NGC -- and then making an estimate for double-counting (maybe my 20-30% guestimate).....wouldn't the price of the coins be dramatically different if the raw PCGS+NGC number (with or without the haircut for double-counting) were WAY off ? 

I agree the numbers may be "off" -- but they still have value.  Question is, how much ?  And if the numbers are off for the 1924 Saint in various grades, does it matter as much if the total numbers (and grades) are off for the 1927-D or 1930-S (coins with dramatically lower population numbers across the board) ?

What Roger said just below your question. On this issue, Roger and I are in complete agreement. 

[RWB]

Also note the comments from Mint Cabinet Curator Comparette to Goddard about acquiring a 1921 DE.

[GoldFinger1969]

20 minutes ago, RWB said:

Also note the comments from Mint Cabinet Curator Comparette to Goddard about acquiring a 1921 DE.

Yes, fascinating stuff you dug up on the 1921.   Usually, most articles (and the few books) focus on TOTAL number of coins available, not the ones in Mint State or Gem or Superb Gem quality.

Can't believe Goddard hijacked a coin and didn't get one for the CT Museum. 

Sometimes I wonder if a child or grandchild has inherited a few coins they don't know the worth of and they might come out in the future, like the 1921's.  Though you figure that they likely would have been sold during previous manias in 1980 or 1989 by heirs or descendants or an estate.