Realcent.org

I am trying to put together a collection of 100,000 wheat pennies. Neat goal huh? I have 15,000 that are unsorted. I recently bought 65,000 wheats from the 40's and 50's. I want to put this to a natural order similar to an unsorted group. The seller neglected to tell me all steel pennies were removed. So how many steel pennies should I add to make it have a normal percentage of steel pennies as if they had not been picked out? Also how many 30's 20's and teens?

I was told for 65,000 in the 40's and 50's that to get normal numbers I would need to add about 10,000 30's 4000 20's and 2000 teens. My guess is about 3000 steels.

gubni wrote:I am trying to put together a collection of 100,000 wheat pennies. Neat goal huh? I have 15,000 that are unsorted. I recently bought 65,000 wheats from the 40's and 50's. I want to put this to a natural order similar to an unsorted group. The seller neglected to tell me all steel pennies were removed. So how many steel pennies should I add to make it have a normal percentage of steel pennies as if they had not been picked out? Also how many 30's 20's and teens?

I was told for 65,000 in the 40's and 50's that to get normal numbers I would need to add about 10,000 30's 4000 20's and 2000 teens. My guess is about 3000 steels.

The term unsorted is a misnomer as we all sort if you see something odd you are going to pull it. Steel cents stick out like a sore thumb so they are all ways pulled out. They also stick to a magnet so most coin counters pull them out. We have a hundreds of steel cents from coin counters but you never find any 1954 cents. see --http://realcent.org/viewtopic.php?f=7&t=5364&hilit=+1954. So I would think that you would have to at zero to be normal on the steel cents.

P1090909 by henrysmedford, on Flickr

henrysmedford wrote:The term unsorted is a misnomer as we all sort if you see something odd you are going to pull it. Steel cents stick out like a sore thumb so they are all ways pulled out. They also stick to a magnet so most coin counters pull them out. We have a hundreds of steel cents from coin counters but you never find any 1954 cents. see --

Well then I did not convey my question properly. I bought several bags of wheat in the past and there was always steel pennies in them. I was wondering if someone could help me know the percentage of steel.

if your unsorted wheaties contained steel cents then they were put there intentionally by the seller. Steel cents were only made in one year, 1943 so if you want to have an even mix of 40's & 50's they should make up 5% of the total. (1 year out of 20 = 5%)

My first thought to get a good mix of wheats would be to combine all the mintage numbers in a spreadsheet. Get the total wheats minted, then do the quick math on percentages of the total for each year and mint mark...take all that out of 100,000 and you would have a proportional mix of exactly how they were minted.

Assuming on a just few things:
That all Wheats remained in circulation the whole time from 1909-1958
There were adequate forces in place to circulate all the coins with mint marks to each side of the country (P, D, S)
That all Indian head pennies were "out" of circulation by the first 1909 Lincoln
..........etc. etc.

To make it more realistic of a mix, what you could do is account for the loss among years as time went on to account for how many coins would actually still be in circulation at any given time. Like 5% of the wheat population each year is lost to ball park bleachers, grocery store parking lots, storm drains, hoarded, coin counter magnets etc. Speaking of time, I would also think that you would have to pick a date that these "Un-searched Wheats (Proper mixture is what I'm calling it)" were gathered. Due to the the loss factor decided in the previous step, I would think the wheat mix would be quite different if assembled in 1959-1960 vs being assembled in 2013.

What I'm trying to figure out in my head is that if you were to collect all 100,000 Wheats from roll searching in one year (For Example, not a new target for HCBTT), what is a realistic mix you would come up with. Obviously you would be getting much more 40s and 50s, but what portion of that is because they are the most recent Wheats and what portion is because they had much higher mintages than the others?

One would think with all of this available data, that someone would have crunched these numbers already I might throw some of this in to a spreadsheet and see how far I can get, but surely we have some sort of statistician here that could take this on....

Ultimately there will be no Ideal mix or ultimate blend of un-searched wheats, but I'm thinking with some of the techniques I have outlined above coupled with a few more ideas that are still in my head, I'm pretty sure you could get close

Okay, I couldn't resist getting started.
http://en.wikipedia.org/wiki/Lincoln_cent_mintage_figures

These are the numbers that I came up with just based off of the mintage figures of each year expressed as a total out of a theoretical 100,000 Wheats. I didn't account for loss like I mentioned above , but that can be done later if you really need it.


Based upon ideal percentages you should also have:
(2) 1909 VBD S
(5) 1914 Ds
(5,557) 1944 Ps, the most common Wheat/Mint Mark

The same math showed
(4,236) 1943 Steel, includes all 3 mint marks

Good luck, hope this gets you started, and BTW cool project

Awesome! I am guessing 10% loss per year while in circulation also.

gubni wrote:Awesome! I am guessing 10% loss per year while in circulation also.

Hi there,

I think you are close, but I remember looking at one of our fellow member's site (sorry I do not remember which one), and the attrition rate was 6% for normal years and 12% for key-dates.

Cheers,

mtalbot_ca wrote:Hi there,

I think you are close, but I remember looking at one of our fellow member's site (sorry I do not remember which one), and the attrition rate was 6% for normal years and 12% for key-dates.

Cheers,

Because I wasn't really satisfied with the numbers I got while not accounting for loss, I decided to see if I could include it fairly easily into the math. I started with 10% across the board, but then saw this post and changed it to a universal 6% across the board. I did not vary based on key dates etc.

For those following the math, here is what I changed from my previous response:
I eliminated Proofs and Matte Proofs from the equation. Most of these has mintages below 20,000. I also removed all of the individual Date/Mintmark combos with less than 100,000 minted total, for both ease of calculations and to make some of the figures not round down to "0". (I'd say since 11/50 years had mintages over 1 Billion, those are small potatoes). In statistics you are allowed to remove data if you don't want to use it and feel it's irrelevant (My math, My rules)

Using my discretion, I grouped everything by year (All 1909s, 1922's etc. I just combined all mint marks for a given year and that was the total minted). I modeled that this Theoretical 100,000 Wheatbacks was assembled in the year 1959, all at once. Using the 6% loss figure above, I accounted that for each year any given coin was in circulation, that loss took effect. For example, the 1958s were in circulation for 1 year, whereas the 1932 coins were circulating for 27 years.

For the equations, I took the 94% and multiplied it to the total mintage for that year, raised to a power of the years it was in circulation (1955*0.94*0.94*0.94*0.94) or (1955 total mintage*(0.94^4)) I did this for each Wheat year and got a total of 12,139,108,588 wheats theoretically still in circulation as of 1959. Even though this is an ideal case scenario, it should still be a little more accurate than the original numbers I posted above.

Here are the new Numbers split by "Decade"

Shown above is that the ratio of 40s/50s is above 9 to 1 with the other years. This does make since though if you think about it for 2 reasons, the earlier ones were in circulation longer therefore you will have more "modern" wheats, and the other is the total mintages of the newer wheats compared to the earlier years. There were just plain more of th 40s/50s made, pretty simple. So IMHO those 2 factors are working directly against you and all of us, for having a good "solid mix" of wheats. That term may need some revisting as this math shows a 9+:1 ratio. I was not expecting that

This still doesn't factor in any remaining Indian heads or any different loss for '43 steels or VBDs, just starting from no pennies (Not true) and going from all 1909s in year one to some of each year, in year 50 (1959). Please point out any flaws in my logic as I am no expert by far. There has to be someone else who can figure this out better than I and shed some light on the situation.

I found the site: (Market Harmony) http://marketharmony.net/uploads/pennie ... lation.pdf

Cheers,

First I must say I admire your math abilities and thank you so much. This idea as if they were picked in 1959 is excellent. Can you give me a number for steel pennies? I would think they were picked out more so than other pennies so in 1959 wouldn't be as many.

First off, gotta give props to Market Harmony for the breakdown of theoretical circulation numbers. I have a case of SpreadSheet envy
I probably shouldn't have removed the small numbers as that will (slightly) skew the results, especially since I grouped each by year anyway. I wonder how you came up with the figure of circulation decline like the 6%/12% etc. and the way you have it set up looks pretty slick for adjusting any of the variables.

gubni wrote:First I must say I admire your math abilities and thank you so much. This idea as if they were picked in 1959 is excellent. Can you give me a number for steel pennies? I would think they were picked out more so than other pennies so in 1959 wouldn't be as many.

To your question Gubni:
My math at 6% loss per year showed (3,348) 1943s across all mint marks in 1959.
Based on MH's loss rate of 8%, I got about (2,397). There were over a billion minted, so changing the loss rate over 16 years really makes a difference in the numbers.