Geocaching Growth

[+fly46]

fizzy, you may very well be a good person, but all I heard from your posts was blah blah blah blah blah..

 

Asking someone else this....

How do you find your user number?

[+Sputnik 57]

I don't care if it makes me a geek. I love fizzy's numbers. I thought the Parade Magazine exposure would wreak havoc on the site, but there are obviously a limited supply of geeks who are actually willing to go outdoors!

[+Moose Mob]

fizzy, I like your charts. I find them interesting. The curiousity question I have is... when will we run out of 6 character GC numbers?

[+fizzymagic]

GCZZZZ will be cache 512400. If the quadratic growth keeps up, looks like we'll run out around June 2007.

 

One more letter included in the alphabet (base 32 instead of base 31) would have helped a lot. But it's not exactly a pressing issue.

[+Sputnik 57]

GCZZZZ will be cache 512400. If the quadratic growth keeps up, looks like we'll run out around June 2007.

 

One more letter included in the alphabet (base 32 instead of base 31) would have helped a lot. But it's not exactly a pressing issue.

I'm not sure we'll make it to June 2007. I read somewhere (Markwell's update to the FAQ?) that certain GC numbers are edited out to avoid names that might otherwise offend the sensitive among us.

[+Moose Mob]

That would be a lot of four letter words to force the trend to be that far off.

 

And thanks for the info fizzy. That question has come to mind several times recently.

[Tahosa and Sons]

Nice Bump - And I've seen a lot of growth since I started playing this game, and hope to see more. It's ironic that this thread surfaced the same day I received an

invitation from a manager of a resort in CO. to talk with them about doing some caches near there place so they can use it in their agenda. Now what a decent way to expose people to CO and caching. I hope their guest are ready for some mental and physical challenges.

[+mtn-man]

Asking someone else this.... 

How do you find your user number?

fly46, in the forums all you have to do is mouse over the user's name and you can see it in the bottom left corner of the IE browser. You are 130099. I am 3512. On cache pages it is all guid numbers now.

 

Interesting graphs fizzy. Thanks for the updates.

 

(Edit, I meant my other right )

Edited November 25, 2004 by mtn-man

[+Nurse Dave]

Of course there are not unlimited places for a cache. As areas become saturated eventually the placements will slow. I mean, there are only so many light poles ya know.

Edited November 25, 2004 by Nurse Dave

[+amytincan]

wow. There are some smart cookies out there. Fizzy thanks for sharing.

[+Team Virginia Browns]

Very interesting data, enjoyed your analysis.

 

We have circles arrows and a paragraph to explain what each one was...

 

but where are those rakes shovels and implements of destruction and the VW bus?

[+fly46]

Asking someone else this.... 

How do you find your user number?

fly46, in the forums all you have to do is mouse over the user's name and you can see it in the bottom left corner of the IE browser. You are 130099. I am 3512. On cache pages it is all guid numbers now.

 

Interesting graphs fizzy. Thanks for the updates.

 

(Edit, I meant my other right )

Thanks MtnMan.

[+Hynr]

GCZZZZ will be cache 512400.  If the quadratic growth keeps up, looks like we'll run out around June 2007.

While it is pretty iimportant to use maximally 6 characters (that's the most that many older GPSr units support), it is not particularly important to have a C in the second spot; Even the G in the first spot is relatively unimportant. Thus if we continue codes with GD0001 after GCZZZZ, then we can go a few more months without someone having to worry about it.

[+The Puzzler]

I can't resist.

 

1st: Thanks Fizzy for the fun plots. I really like them a lot especially when you include your analysis. It makes a lot of sense.

 

2nd: But, I am puzzled. Maybe it's because I'm a biologist and not a physicist, but why do you expect growth of geocaching to be exponential by default? Exponential growth, by my understanding, comes from the continuous reproduction of offspring over multiple generations. Growth is only exponential if the increase in the number of geocachers with each time step (generation, year, whatever) is dependent on the number of geocachers in previous time step, without any loss. Real systems, even biologically repoductive ones, seldom demonstrate exponential growth except in very controled settings over finite periods of time (generally only when a population is very new, resources are plentiful, and predators have not found them yet - maybe a bit like geocaching?)

 

3rd: So, what do we expect in the future? Do we have evidence of resource depletion (no more hiding places)? If geocaching is primarily spread by word of mouth (not Parade Magazine Articles) so that be can assume exponential growth for a little while anyway, what is the loss term of geocachers - i.e., the difference between exponential and the actual plot? What is the growth vs. loss rate of us?

Do we see any sign of geocacher saturation - the growth rate decreasing with time due to all the interested people already being involved in the sport?

 

4th: I still think the best piece of information is when the waypoint ID numbers will run out.

 

Thanks again for the fun thoughts.

[+Moose Mob]

2nd: But, I am puzzled.  Maybe it's because I'm a biologist and not a physicist, but why do you expect growth of geocaching to be exponential by default?  Exponential growth, by my understanding, comes from the continuous reproduction of offspring over multiple generations.  Growth is only exponential if the increase in the number of geocachers with each time step (generation, year, whatever) is dependent on the number of geocachers in previous time step, without any loss.  Real systems, even biologically repoductive ones, seldom demonstrate exponential growth except in very controled settings over finite periods of time (generally only when a population is very new, resources are plentiful, and predators have not found them yet - maybe a bit like geocaching?)

Here's my thoughts on why folks would assume it would be exponential... cacher A introduces cacher B, The following month, Cachers A intro's cacher C, Cacher B intros cacher D... the following month Cachers A,B,C, & D intro Cachers E, F, G,H, So, 1 becomes 2, 2 becomes 4, 4 becomes 8, etc. That woul be exponential growth.

 

What we seem to have is linear, wher Cacher A into's B, B intros C, C intros D. (Or.... A intros B, C, & D, and B, C & D don't intro anyone.)

 

What would be interesting is to see the rate of active cachers. A look from different angle of course. The criteria for counting active cachers could be "found or hide cache within 30 days", 90 day, or 6 months. We know that some lay low for several months then get active again. I doubt there are enough to really throw off the curve, but it one would have to see that data.

 

3rd: So, what do we expect in the future?  Do we have evidence of resource depletion (no more hiding places)?  If geocaching is primarily spread by word of mouth (not Parade Magazine Articles) so that be can assume exponential growth for a little while anyway, what is the loss term of geocachers - i.e., the difference between exponential and the actual plot?  What is the growth vs. loss rate of us?

Do we see any sign of geocacher saturation - the growth rate decreasing with time due to all the interested people already being involved in the sport?

 

Cache saturation.. to pull a number out of my *&^%, I would say 80% of the great views are taken care of, unless someones plant a cache 528 feet away from an existing hilltop cache. Urban caches are growong fastest. That level 4 hike up a hill takes too long to get to.

 

Am I close?

Edited November 26, 2004 by Moose Mob

[+Moose Mob]

I can't resist.

 

1st: Thanks Fizzy for the fun plots. I really like them a lot especially when you include your analysis. It makes a lot of sense.

 

2nd: But, I am puzzled. Maybe it's because I'm a biologist and not a physicist, but why do you expect growth of geocaching to be exponential by default? Exponential growth, by my understanding, comes from the continuous reproduction of offspring over multiple generations. Growth is only exponential if the increase in the number of geocachers with each time step (generation, year, whatever) is dependent on the number of geocachers in previous time step, without any loss. Real systems, even biologically repoductive ones, seldom demonstrate exponential growth except in very controled settings over finite periods of time (generally only when a population is very new, resources are plentiful, and predators have not found them yet - maybe a bit like geocaching?)

 

3rd: So, what do we expect in the future? Do we have evidence of resource depletion (no more hiding places)? If geocaching is primarily spread by word of mouth (not Parade Magazine Articles) so that be can assume exponential growth for a little while anyway, what is the loss term of geocachers - i.e., the difference between exponential and the actual plot? What is the growth vs. loss rate of us?

Do we see any sign of geocacher saturation - the growth rate decreasing with time due to all the interested people already being involved in the sport?

 

4th: I still think the best piece of information is when the waypoint ID numbers will run out.

 

Thanks again for the fun thoughts.

Fizzy, any input on this one? (personally, I don't think what you have to say on this stuff is blah, blah, blah).

[CoyoteRed]

On the issue of cache saturation, don't assume cache growth will be affected by the number of parks or areas to place a cache. It will be determined by the total area of places to place a cache and will be affected by the ease at which you can place a cache.

 

A vistor a while back commented on the total lack of cache seperation in my area. While we might have a cache or two per park, where they were from they would have many caches in the same area. We might think that we are tapped out with the number of caches placed, folks from other parts of the country will think we've got plenty of room.

 

In other words, local mindset will determine the potential for growth at any one point.

[+Moose Mob]

Agreed CR. Although there are a finite number of light poles, it'll become more difficult to find a light pole without a cache. </cynasism>

[CoyoteRed]

Agreed CR. Although there are a finite number of light poles, it'll become more difficult to find a light pole without a cache. </cynasism>

 

Around here, the lamp post cache population has massive room for growth. I don't recall there being the first one!

 

I wonder if my fellow local cachers will kick my butt if I hide a few.

 

...probably.

[+fizzymagic]

The original analysis was posted in response to some people who claimed that Geocaching was undergoing exponential growth. Please don't think that I thought it would be exponential!!!

 

On the other hand, I don't completely understand the almost perfect quadratic growth we have seen in both caches placed and in number of cachers. I believe that the cubic growth in the number of logs is expected (more or less) given the quadratic growth in cachers and caches.

 

Systems that exhibit quadratic growth are usually 2-dimensional, like a bacterial culture on the surface of a culture medium. Growth is limited to the perimeter of the culture, giving an instantaneous growth rate that is linear with time, and a cumulative growth that is quadratic. There is nothing about geocaching, in my opinion, that imposes similar constraints.

 

There may be some argument that ties the growth of new geocache placements to existing caches; since caches are arrayed on a 2-D surface, maybe there is something there. But I admit that it is a puzzling growth pattern for which I have no good explanation.

[+The Puzzler]

On the other hand, I don't completely understand the almost perfect quadratic growth we have seen in both caches placed and in number of cachers.

I did some playing around with my spreadsheet.

If we make some simplifying assumptions, I may be able to explain the power-function growth of caches.

 

Lets assume that the knowedge of caching is still limited enough that over the time frame of these data, awareness of caching and thus the addition of active cachers is roughly exponential - most people don't know about it and publicity is proportional to the number of cachers.

 

We must also assume that some proportion of cachers regularly become inactive (quit hiding caches).

 

Then, if we assume that the number of caches hidden is proportional to the number of active cachers at any point in time, we get a function that departs from exponential and approaches a power function as the loss rate of active cachers increases and as the number of caches hidden per active cacher increases.

 

I did not include the loss rate of caches which would surely push the fit even further from a simple exponential.

 

So, the power function may just be the result of loosing active cachers while continuing to accumulate caches hidden by the remaining active cachers.

 

If the number of cachers quit increasing because the loss was equal to the gain, we would achive a trend in cache abundance that was simpily linear.

[+fizzymagic]

It's been a while since I updated this data. So here we go:

 

Number of caches:

 

Number of registered cachers:

 

Number of logs:

Edited August 8, 2005 by fizzymagic

[+fizzymagic]

Further analysis:

 

Square root of number of caches:

The line is to guide the eye. Seems like there is a recent upsurge in cache placements, but it is remarkably consistent over time. I still have no idea why this would be linear.

 

By the way, this predicts that the cache GCZZZZ will be placed between March and June of 2007. (GCZZZZ = cache number 512400; the sqrt of that is about 716)

 

Square root of number of registered cachers:

As above, the line is to guide the eye. Also quite consistent.

 

The ratio of caches placed to total registered users seems more or less constant over time. Interesting.

 

Cube root of number of logs:

The number of logs is growing faster than the cube of time, but still slower than the fourth power of time, as might be naively expected from the growth in caches and users.

Edited August 8, 2005 by fizzymagic

[+The Leprechauns]

Here is a great example of an old thread that SHOULD be bumped every once in awhile. Thanks for the update, Fizzymagic! It is always so interesting to review this thread when you add new charts.

[+budd-rdc]

Here is a great example of an old thread that SHOULD be bumped every once in awhile.  Thanks for the update, Fizzymagic!  It is always so interesting to review this thread when you add new charts.

I echo that sentiment.

 

Thanks fizzy, for giving me something that doesn't require a headache medicine. This thread answered several things I've wondered about.

[+Team Cotati]

The rapid growth that I so much like to see, continues unabated. Hopefully the pace will pick up a bit real soon. Wonderful.

Edited August 8, 2005 by Team Cotati

[+Team Maccabee]

Fascinating thread!

 

So I have a research question. professor(s).

 

What is the ratio of premium members to active, non-premium registered users (say, someone who logs at least once per month or so) over time? And are there geographical differences?

 

A practical application: is my premium-member-only cache really shutting out a lot of cachers in my area (Seattle)?

 

Thanks for the great analysis!

 

Team Maccabee

[+Douglas_Clan]

Will this explain what happened to the missing sock from the dryer ?

[+wimseyguy]

Will this explain what happened to the missing sock from the dryer ?

I have noticed that I loose far fewer socks to the dryer monster now that I am married. Therefore I conclude that the use of dryer sheets helps prevent the sock loss. (C'mon what single guy uses those things? )

 

Back OT, thanks for the great charts and insight fizzy. Keep 'em coming.

[+Vinny & Sue Team]

Fizzymagic, thank you for your effort at producing these stats and charts. This is interesting stuff to me, and I have wondered in the past about exactly the types of questions which you so kindly answered. Very illuminating! Thanks!

[+fizzymagic]

OK, it's time for the end-of-year update!

 

Caches:

 

Square root plot:

 

Seems like cache placements have started to grow faster than they were.

[+fizzymagic]

Registered users:

 

And a square root plot:

 

The number of users also seems to be growing more quickly.

[+fizzymagic]

Logs:

 

Cube root of logs:

[+Nerves]

Interesting.

 

Now, how about a study on the behavioral and personality traits of geocachers?

[+Elde]

Here's my thoughts on why folks would assume it would be exponential... cacher A introduces cacher B, The following month, Cachers A intro's cacher C, Cacher B intros cacher D... the following month Cachers A,B,C, & D intro Cachers E, F, G,H,  So, 1 becomes 2, 2 becomes 4, 4 becomes 8, etc.  That woul be exponential growth.

That's the misperception that Amway and their ilk encourage folks to believe, because it makes their downline potential look impressive. Or are we hardwired into believing that this exponential growth is the 'norm' - I've seen at least three ad campaigns over the years that trade on that misperception.

 

What we seem to have is linear, wher Cacher A into's B, B intros C, C intros D.  (Or.... A intros B, C, & D, and B, C & D don't intro anyone.)

I suspect the first is correct - among my circle of friends only one has treated caching as anything other than "another weird interest of D____'s". (The cost of entry may be a barrier as well, to many people even a $100 Garmin Yellow for a hobby represents a significant psychological barrier.)

[+Team Cotati]

Thanks fizzy! Arent' those numbers great!!? I wish that my investment portfolio was doing half as well, trust me.

[+Moose Mob]

Here's my thoughts on why folks would assume it would be exponential... cacher A introduces cacher B, The following month, Cachers A intro's cacher C, Cacher B intros cacher D... the following month Cachers A,B,C, & D intro Cachers E, F, G,H,  So, 1 becomes 2, 2 becomes 4, 4 becomes 8, etc.  That woul be exponential growth.

That's the misperception that Amway and their ilk encourage folks to believe, because it makes their downline potential look impressive. Or are we hardwired into believing that this exponential growth is the 'norm' - I've seen at least three ad campaigns over the years that trade on that misperception.

 

What we seem to have is linear, wher Cacher A into's B, B intros C, C intros D.  (Or.... A intros B, C, & D, and B, C & D don't intro anyone.)

I suspect the first is correct - among my circle of friends only one has treated caching as anything other than "another weird interest of D____'s". (The cost of entry may be a barrier as well, to many people even a $100 Garmin Yellow for a hobby represents a significant psychological barrier.)

Actually, that isn't a misperception, it really happens that way.

 

The misperception with Amway is that you alienate your friends when you try to sell them on it, you spend countless hours trying to have the "hard sell" personality that you really don't have.

 

Edit to bring back on topic:

The exponential growth is a common occurance in anything new and popular, until it reaches a point of saturation and can no longer sustain the growth.

Edited January 31, 2006 by Moose Mob

[+reveritt]

I can't pass this up. "Exponential Growth" is always funny to hear. ...

Exponential growth does have a technical meaning, and is often confused with geometric growth. If you measure something at regular intervals--monthly, say--and each measurement is X times the previous measurement, where X is greater than 1, then you have exponential growth. If X is less than 1, but greater than 0, you have exponential decay. The quantity you are measuring, and all its derivatives with respect to time can be expressed as funnctions of e.

 

This is different from geometric growth where the measured quantity is some polynomial function of time.

[+reveritt]

...

Exponential growth curves would actually be better trends because attrition rates are less effectual on the bottom line. We are stable and upward currently. But we are close to an inflection in the data is my interpretation.

I agree, and I think a better model for the growth curve might be a sigmoid function

 

[+fizzymagic]

...I think a better model for the growth curve might be a sigmoid function

Great idea, except the sigmoid function doesn't fit the data. Nor does the more general function you actually meant, the logistic function (the sigmoid function has a width of 1).

 

And exponential growth is the limit of geometric growth as the time interval between changes goes to zero.

Edited January 31, 2006 by fizzymagic

[+Elde]

Here's my thoughts on why folks would assume it would be exponential... cacher A introduces cacher B, The following month, Cachers A intro's cacher C, Cacher B intros cacher D... the following month Cachers A,B,C, & D intro Cachers E, F, G,H,  So, 1 becomes 2, 2 becomes 4, 4 becomes 8, etc.  That woul be exponential growth.

That's the misperception that Amway and their ilk encourage folks to believe, because it makes their downline potential look impressive. Or are we hardwired into believing that this exponential growth is the 'norm' - I've seen at least three ad campaigns over the years that trade on that misperception.

Actually, that isn't a misperception, it really happens that way.

If it really does happen that way - then why is it not doing so in this case? As you say, people assume it would be - why would they have that assumption? That's what I seek to examine.

 

Watching grass roots marketing campaigns, etc, over the decades - it seems to depend on luck as much as anything else. (And substantial grassroots buzz can arise without assistance from any marketing. Austin Powers flopped in the movie theaters - but went gangbusters in the DVD/video market for example. This lead to two sequels and more and more movie companies pushing the DVD version out as early as possible with as much buzz as their budget can stand.)

 

Therefore, I believe the assumption arises because folks remember the times it worked, and the many more times it didn't they don't - because it never appeared on their radar in the first place. MLM's and pyramid schemes play on this.

 

The exponential growth is a common occurance in anything new and popular, until it reaches a point of saturation and can no longer sustain the growth.

Except - as Fizzy's graphs plainly show, that's not happening in the case of geocaching despite your assertion that 'it really does happen that way'. I'd say it does happen in instances - but it's not universal and not guaranteed. (It's hard to say if 'popular' is the chicken or the egg.)

 

What will be interesting to see will be the 4thQ 2006 and 2007 version of this chart. GPS unit are rapidly becoming 'the' auto acessory and GPS enabled phones are also starting to enter the market place. I suspect Geocaching will grow radically over the next two years - but because their will be multiple 'seeds' rather than growth from 'infection' by individual geocachers.

[+Moose Mob]

 

I think I'm better off leaving this one alone for awhile.

[+fizzymagic]

I think I'm better off leaving this one alone for awhile.

Well, maybe. My main point in starting this thread way back when was to try to understand why the growth is not exponential. Because the naive expectation is that it should be, and that was certainly my prediction.

 

There may be some expectation that as geocaching reaches saturation that there will be a slowdown in the growth. That may happen someday; but the data don't indicate that yet. In fact, recent trends are above the historic growth curve, not below it.

 

I find it amazing that the growth in the number of cachers and caches is approximately quadratic. I don't know of any simple model that would predict that behavior.

 

That's what this thread ought to be about.

[+fishingfools]

cool, neat to see math in action, not just laying around lazy in a math book

[+WalruZ]

I would be interested in seeing a graph of active caches over time. Unfortunately, not having history, you would need various details from all cache pages in order to do this.

[+fizzymagic]

There's been some recent discussion about reaching the end of 4-character waypoints, so I thought I would update the data here.

 

First, the total number of caches submitted.

 

And the sqrt plot:

The line is drawn just to guide the eye; but you can see that the rate of cache submissions has increased in the last year.

 

The reason for that can be seen in the plot of the number of users:

 

Or, rather in the sqrt plot of users:

 

As you can see, starting last summer there was a distinct upsurge in the trend of new users.

[+fizzymagic]

Some additional analysis:

 

First of all, here is where we stand in relation to the end of 4-digit waypoints. The last of those will be GCZZZZ. It looks to me like we might not even make it to the end of this year!

 

And here is a semi-log plot illustrating that the growth is not exponential.

 

Exponential growth gives a straight line on a semi-log plot. Yeah, I understand that it will be more complicated if the exponential growth starts after a period of non-exponential growth, but I still claim it's not exponential.

 

But the semi-log plot can give us a limit on how early we might reach the magic number. Assuming exponential growth, it looks to me like we could hit is as early as October. Perhaps we should start a pool?

 

Interesting new data, I hope.

Edited May 6, 2006 by fizzymagic

[gerboa]

PLease how can I save this stuff ????

[+Team Cotati]

PLease how can I save this stuff ????

 

I think that gc.com is doing that for you.

[+bigeddy]

Interesting new data, I hope.

Yes! Thank you.

 

Am I reading the charts right?:

 

At the start of 2003 there were roughly 50k caches submitted by 100k users or a 1:4 ratio.

At the start of 2006 there were roughly 330k caches submitted by 550k users or a 1:1.7 ratio.

 

Despite the huge increase in users at some point the number of caches may overtake them. That includes archived caches, of course, but it is still amazing growth. I'm starting to see more people now who have over a hundred placed caches most of which are still active.