Circulation Statistics: Please Help

[Guadalcanal]

Circulation Statistics

Hello everybody! I enjoy programming a lot, and I thought to myself "Hey, why not try to make an application that shows the chances of finding old coins at parks?" I just started today, and I want to make that a reality. I think it would be neat to have a program that looks for things like:

• What was there before the park?
• How Old is the Park?
• How many parks are there in the city?
• How many people live in the city?
• How many tourists visit the city each year?
• What is the soil like at the park?
• What is the climate like in the city?
• What is/was at the park?(Picnic areas, ice cream shops, baseball diamonds, ticket booths, etc.)
• What is the population density of the city?

It would then return things like

• Average age of coins underground
• Likelihood of finding artifacts
• Depth of items of different ages

Status of project: I'm currently adding in mint statistics. It's going to take a while, since I have to add in each year for every coin, but I've finished all pennies and Jefferson nickels so far. I also did a quick check on the forums to see which kinds of clad people found, and I'm glad to say that I found correlation between findings and mintage!!!

Take a look for yourselves:
Mintage:
605,175,770,000 Pennies minted 1959-present 91.4%
56,749,710,000 Nickels minted 1959-present 8.6%
Discoveries:
2519 Pennies found 90.6%
262 Nickels found 9.4%

As you can see, the percentages are very close to each other. I will be checking the numbers again each time I add a new denomination of money, but so far it's looking good. I hope to finish entering in all the mintage dates in about two weeks. I've got 2000 lines of code just inputting all these dang numbers xD

I'll keep you all posted as I learn more.

Thank you all very much, I hope this program will be useful to everybody!

 

Amazon Forum Fav 👍

DR.ÖTEK Metal Detector, Waterproof Gold Detector, Pinpoint & Terrain Modes, Professional Metal Detector - Grab it through Amazon!

[SusanMN]

Just curious how you will estimate and take account the old coins that once were in the park but have been taken out by other hunters over the last 50 years .

 

[Guadalcanal]

Unfortunately, that is going to be difficult. It's going to be impossible to find how many people have detected for metal over the years, so I can't get a solid number. I will either not factor it in at all, but mention to the user that this assumes that nobody has detected there, or I'll give them an option to choose how hard it's been hit over the years. This program will just estimate how many coins were dropped/how deep they will be buried. I wish I could tell how many had been dug up, but sadly that will not be plausible.

 

[ConceptualizedNetherlandr]

He just wants to know where to go hunt

 

[Dave Rishar]

This is an interesting idea, but I'm not sure that it's possible to estimate what's at a site with even to even a partially accurate degree...and that's before we factor in how the site's been worked in the past, which can be difficult to do. I'm not sure how one would model coin drops, as they're random by nature. I've found too many things where they shouldn't have been, and not enough things where they should have been. It will be extremely difficult to model something that approximates a number of different theories which don't necessarily agree with each other, cannot be easily proven, and tend to be vague at best.

In fact, I'd say that half the fun of the hobby for me is looking at a find and saying to myself, "Now what in the hell is this doing here?" The flipside to that ("THIS SITE HAS BEEN USED FOR OVER A CENTURY AND NOT ONE EFFING PRE-65 COIN? ARE YOU EFFING SERIOUS?") is not as much fun, but it can be just as mysterious. (For the record, the site I'm referring to finally turned up a WWII-era wheat, at about three inches down, with no trash or anything around it. It was right next to the sidewalk, just a slam-dunk "dig me" signal that any machine would have hit, in a high traffic area. If the park had been pounded so hard as to be nearly free of targets, how did such an easy target sneak through the net? Another mystery.)

 

[Guadalcanal]

Perhaps this is a fool's errand.

@ConceptualizedNeanderthal Haha if that's what I wanted I would have asked. I've already got a couple good places scouted out for when I return to the US

@DaveRishar
What you're saying makes sense. Most of it is just blind luck, but I still like to look at the data and make estimates about whether or not a spot is good for detecting. Congrats on that WWII Wheat by the way, that's pretty awesome!

Yes, this stuff may be way too random to predict, but at least I'm getting some good programming practice and I can learn about metal detecting, so it's still a win even if the program's predictions rarely come true. If anyone has any data on coin dates I'd still be very grateful, but if you don't have the time, no worries. I can get to it when I'm back stateside.

 

[Guadalcanal]

Update:

I've entered in some statistics so far. As of now I will be looking at:

• Number of parks in the city. I will check for number of parks in the city at the time of the target park's creation and the current number of parks in the city.
• Current year and the year the park was created.
• Population of the city when the park was founded and the current population. Later I plan to add an option for entering the population of years in between for enhanced accuracy.

Since it's impossible to predict the quantity of certain coins or visitors, I will only be estimating the percentages of different coins on the ground. After a couple hours of searching this forum for people's clad collections, I came up with an estimate that the coins underground are:
61% Pennies
6% Nickels
19% Dimes
and 12% Quarters
I will have to adjust these figures slightly when I take into account half dollars and dollar coins, which I believe were more frequently used in the past.

I've also entered in the starting and ending dates of each coin's time being minted. I will also add yearly mintage for every US coin to help increase accuracy.

How will I calculate which years were the best for coin dropping?
I will assign each year a select amount of "visitor points". These points will be based on the tourists, commuters, and residents of the year. They will also be influenced by the number of parks in the city. (For example, if there are 50 parks, then I assume that this park will get 1/50, or 2% of the park visitors for the year.) In the future, I will also take into account the park's features. I plan to rank these either "high", "medium" or "low" for amount of coins dropped. For example:
Hot dog stand, vendors, or ticket booth: High boost to coin probability.
Soccer field, baseball diamond, playground, or volleyball court: Medium boost to coin probability
Public restroom, dog park, or basketball court(hard to lose coins on concrete): low boost to coin probability
Therefore, if a park has/had a hot dog stand or other feature, I would add a boost to visitor points for that year. So every year that has a feature is affected by it. I will then calculate each year's visitor points based on the current year and rank them from highest to lowest.

After I find out which years have the most people, I need to find out what coins those people may have dropped. Once again, since I can't get real numbers, I will use percentages. I found that the average time a coin spends in circulation is 30 years.

Assuming that the half-life of a certain year's worth of coins is 30 years, I can make assumptions on what coins those people would have been carrying. I will also take into account the times when different models of coins were use. (For example, when the half dollar switched from a Franklin to a Kennedy, the amount of Franklins began to decrease more quickly because they were pulled out of circulation by the government or they were collected.) I also need to factor in the value of coins. Someone is much less likely to lose a dollar coin than a penny, right now I'm trying to find out how much less likely that would be.

What will the results look like?
Right now I don't have a U.I., this will all be run in the command prompt. The results will order each type of American coin from most to least likely. (I have silver dimes/quarters and clad dimes/quarters in separate groups to make things easier) It will also list ten or so of the most likely dates. I also will make a static example, where the population and the number of parks doesn't change. I will compare the individual results to that park, and list some coins that are very likely to be found in your park. This feature can help you if you're looking for a certain type of coin. The program could say "This park is unusually good for 'Walking Half Dollars'" or something to let you know that the chance for that coin is particularly high.

I'll keep you all posted!
(Also, would you say that the low percentage of nickels found by metal detectors is because of the low amount of nickels dropped, or the fact that people don't dig the signal that nickels give off?)

 

[Charlie P. (NY)]

If you're available, please do me a favor and look through all the change that you have (regular pocket change) and tell me what dates you come up with. For example:
1970: 10 coins
1971: 11 coins
1972 : 20 coins
1980: 50 coins
etc. etc. etc.

I want to use this information so that I can make estimates on the age of coins in the ground. For example, say that 1% of the coins in circulation today are from 1974, and the park you're searching in was created in 1940. If you extrapolate backwards, you could estimate that in 1940, 1% of the coins in circulation were from 1900. Then if you compile each year with its changing population and dates, you can guess at the average age of the coins under the ground! Obviously it's far from foolproof, and there could be flukes (Someone drops a roll of new pennies, or someone loses an already old coin from their collection), but I hope that it will give a small amount of pre-knowledge of what you're getting into before tackling a park.

That census ratio will not work because of the differences in mint volume of certain years. For instance, the Philadelphia mint produced over 10 billion Lincoln cents in 1982 but only 763 million in 1963. But for wheat cents most years were well under 500 million; closer to 100 million or less. Indian Head cents only exceeded 100 million in one year. They were in the 50 million range annually +/-.

Also, do you carry enough coinage in your pocket now to buy two loves of bread and three gallons of gas? In 1940 that was common. Now you'd need $15 in change to do that.

 

[Guadalcanal]

That's a great point. I only thought of including mintage statistics after I started, but I forgot that it would apply to what you mentioned as well.

I forgot to mention that I plan to add the value of the dollar as opposed to the value of gold in my calculations. Higher value of the dollar = increased usefulness of coins = more coins used = more coins dropped. Hopefully this extra modifier will help show the difference in the amount of coins dropped every year more accurately. I also plan to add a decay rate for the number coins in the ground, because eventually coins can get too deep to find, become covered by cement, or any number of things. If you looked at a park in Boston, maybe there would be a high probable drop rate in the 1700's. The reason you would find a small amount of those today is because they have become more and more difficult to find over the years.

Thank you for your input, every suggestion helps make this program more accurate!

 

[Dave Rishar]

(Also, would you say that the low percentage of nickels found by metal detectors is because of the low amount of nickels dropped, or the fact that people don't dig the signal that nickels give off?)

The latter. Nickels live in the trash range and some (most?) folks can't be bothered to dig that sort of thing up, so they discriminate them out or otherwise ignore them. Your statistics seem to confirm this.

Truth be told though, I don't dig very many nickels and I most assuredly do dig the stuff down there. Perhaps there's more to this than I'd initially thought. Or maybe not? I just glanced at a few of my spreadsheets and saw sites that exceeded this nickel ratio by a wide margin, but also some that didn't even meet it. Odd.

This will be very difficult to model accurately.

 

[Charlie P. (NY)]

I make sure to include nickles in my range (I prefer to see and hear a target) and dig anything remotely solid sounding.

These are from one-hour visits to a local park showing every target I went after: good and bad. Plenty of trash and some jewelry . . . and a good showing of nickels. I run about even quarters vs. nickels.

 

[Guadalcanal]

Truth be told though, I don't dig very many nickels and I most assuredly do dig the stuff down there. Perhaps there's more to this than I'd initially thought. Or maybe not? I just glanced at a few of my spreadsheets and saw sites that exceeded this nickel ratio by a wide margin, but also some that didn't even meet it. Odd.

Yes, I think that it is more complex than I originally thought as well. The way I see it, there are three factors that determine the drop rate of coins:
1. Value: If people drop a dollar coin, they will be much more likely to pick it up/notice that it's missing than if they drop a penny. Since the dollar has changed over the years, I plan to use the value of the dollar in comparison to the price of gold to find the likeliness of finding certain coins.
2. Usefulness: Some coins are more useful than others. For example, all vending machines accept quarters, so kids are likely to carry those around.
3. Change ratios: If someone gives you 99 cents change, they will give you 4 pennies, 3 quarters, 2 dimes, and 1 nickel. I think this would also contribute to drop rates.

As far as I can tell, mintage statistics are more useful for finding the average dates of denominations, not the amount of certain denominations. Any chance I could look at those spreadsheets you're talking about? I'd love to find data that I can use to make this more precise.

These are from one-hour visits to a local park showing every target I went after: good and bad. Plenty of trash and some jewelry . . . and a good showing of nickels. I run about even quarters vs. nickels.

Thanks for showing that, I love seeing all of people's finds after a hunt, including the trash It's interesting that you got roughly the same number of nickels as quarters... it could possibly be a fluke, since your sample set isn't terribly large, or it could be that almost everyone skips the signal that nickels give off. I'll need to do some research on this. I fear that the result will be that many factors are involved. This will make things more complicated, yes, but not impossible.

Any other factors that you guys can think of? I really don't want to miss anything, or it will be for naught.

Some interesting facts I've found while doing this:

Mintage vs. MD finds:

Mintage:
605,175,770,000 Pennies minted 1959-present 81.5%
53,387,040,000 Nickels minted 1959-present 7.2%
83,793,650,000 Dimes minted 1959-present 11.3%
Discoveries:
2519 Pennies found 70.4%
262 Nickels found 7.3%
795 Dimes found 22.2%


605,175,770,000 Pennies minted 1959-present
29,152,670,000 Wheat Pennies minted 1909-1958
1,848,510,000 Indian Head Pennies minted 1859-1908
42,060,000 Flying Eagle Pennies minted 1856-1858
157,480,000 Large Pennies minted 1793-1856

56,749,710,000 Jefferson nickels minted 1938-2014

6,368,480,000 Silver dimes minted 1946-1964
80,995,960,000 Clad dimes minted 1965-2014

 

[Msbeepbeep]

And then there are the oddities, like people purposefully throwing pennies on the ground because they can't be bothered with them, then there are the coin spills, lost rolls of coins, caches, storms, etc. there are so many oddities that pertain to this hobby and what we find, and I don't think everybody wants to be on the radar screen with their "exact" finds.
Percentages are probably as close as it gets.
I applaud you inventiveness and willingness to take on a task as hard as this to help out your fellow MD'ers.

 

[Guadalcanal]

Haha I'd pay good money for a program that could estimate locations of caches, I think everyone would Unfortunately that's unrealistic right now, so things like that will remain mysteries. I would actually prefer that the big treasures stay mysteries, that way they'll stay around a little longer so more people can get shots and finding them

From watching these forums for a while it seems that people who do the research find the coins. I just hope that this will make it easier for beginners to know what to research. People will have to look into the history of their city to find the data that this program needs. Chances are this program will be equally helpful when someone discovers something while researching their city as when they actually finish entering the input.

I've almost finished the mintage stats for nickels, then I'll be finishing up dimes.

-Guad

 

[Charlie P. (NY)]

Here's a free site that allows you to find mintage amounts for all US coins. Some good history as well.

Coin Mintages

 

[Guadalcanal]

I've made some progress!

It's slow going, but I'm steadily plowing ahead. I still haven't finished entering in all the mint amounts, but I already have a rough equation worked out for the rate at which coins fall out of circulation. There are several factors that go into this rate that I've come up with so far:
*Completed* 1. Value of the coin. Pennies fall out of circulation the fastest, then nickels, then dimes, etc.
*Completed* 2. Composition of the coin. Silver dimes fall out of circulation faster than clad. Same with wartime nickels, old quarters, etc.
*Incomplete* 3. Catastrophic events, such as the Gold Recall, the recalling of large pennies when the Flying Eagles came out, and so on.
Here is a chart showing the number of Jefferson Nickels in circulations. (Numbers of nickels should be multiplied by 10,000)



And here is the readout file that the program currently gives, if anyone is interested. In here is included the percentage of each date of wheat pennies and Jefferson nickels remaining every year:
View attachment readout.txt

Thanks for reading!

 

[Guadalcanal]

Great news! I have a working prototype. First thing that should be noted:
**This is only for pennies**
I still need to enter in some more numbers before I start calculating other coins too. This will show you what you're likely to get for pennies only.
**I have not calculated the drop in large pennies when they were melted down for flying eagles**
I'm getting to that... really...

Here's an example of what I get for a park that was established in 1944:

Percentage of <lincolnPenny> in park: 69.45976866789852%
Percentage of <wheatPenny> in park: 30.4020417526426%
Percentage of <indianPenny> in park: 0.13790780867660196%
Percentage of <flyingEaglePenny> in park: 2.813505615084882E-4%
Percentage of <largePenny> in park: 4.202207577151718E-7%

And here's what happens for a park that started in 1926:

Percentage of <lincolnPenny> in park: 54.85640225634923%
Percentage of <wheatPenny> in park: 43.76569054943659%
Percentage of <indianPenny> in park: 1.3756063643439254%
Percentage of <flyingEaglePenny> in park: 0.002286793005833308%
Percentage of <largePenny> in park: 1.4036864420833183E-5%

Download:Host-A Hosting » guadalcanal » PennyScanner

There is some information you will be asked for!
1. Year the park was established
2. Current year (Entering something higher than 2014 will mess it up. Feel free to mess around and enter lower years if you'd like though.
3. Population of the city when the park was established. I understand that this may be hard to find, and you may have to estimate, but this is essential to figuring out the percentages. If you really don't want to do the research (bad) just enter the same numbers for #3 and #4
4. Current population of the city.
5. Number of parks in the city when your park was established. I use a population per park sort of thing for determining how many people go to the park each year. This information is quite hard to find, but I found it for my park and I bet you can find it for yours if you look hard enough.
6. Current number of parks in the city.

There are still some variables I need to factor in before this number is accurate, including, but not limited to:
Special features of your park. Playgrounds, ticket booths, taco trucks, etc.
Dollar:Gold ratios, Higher ratio, lower value coins dropped.

Tell me what you think! What results do you get for parks near you?

 

[Davers]

Interesting ..
Keep up the good work.
and Thanks
Davers

 

[Dave Rishar]

Yes, I think that it is more complex than I originally thought as well. The way I see it, there are three factors that determine the drop rate of coins:
1. Value: If people drop a dollar coin, they will be much more likely to pick it up/notice that it's missing than if they drop a penny. Since the dollar has changed over the years, I plan to use the value of the dollar in comparison to the price of gold to find the likeliness of finding certain coins.
2. Usefulness: Some coins are more useful than others. For example, all vending machines accept quarters, so kids are likely to carry those around.
3. Change ratios: If someone gives you 99 cents change, they will give you 4 pennies, 3 quarters, 2 dimes, and 1 nickel. I think this would also contribute to drop rates.

1. It's more complicated than that. The US gold dollar was smaller than a dime and was largely hated because it was so easy to lose on account of its size. (Imagine dropping a dime in tall grass. You may look for it, but good luck finding it! The same goes for sand, mud, etc.) One could theorize that not only was it small and easy to lose, but it was also dense enough to work its way into the ground quickly. But yes, someone would certainly be more inclined to look for a coin that's worth more...assuming that they knew that they'd lost it in the first place, which in some cases they probably did not.

2. Agreed, although if we go back to the days before vending machines, all coins were useful. With the advent of vending machines, this places and adverse pressure on the value of coins. Did any vending machine ever accept a $0.50 coin? I know that some vending machines now accept the dollar coin, but this came well after the advent of the dollar coin.

3. Agreed conditionally. The $x.99 model of psychological pricing hasn't been around forever, and changes in taxation have changed pricing models as well. Based on the change that I currently encounter, I'd expect pennies, nickels, dimes and quarters, but what I encounter now was not what was necessarily encountered at any other time. 110 years ago, dollar coins circulated in this area; nowadays, you only see them as change from certain vending machines, or maybe as a cashier. Things change.

On a related note, I keep somewhat detailed spreadsheets maintained on sites that I've worked, Guadalcanal. I can provide you with some of them if you think that they'd help, but the only thing that they've proved to me so far is that there's only a tiny relationship between what should be in the ground and what's actually there. PM me if you'd like them, but I'm not sure that they'll help.