Jump to content

Cache centroid


Stokesy
Followers 1

Recommended Posts

I've been using GSAK and the excellent stats macro for my profile for a while, and occasionally click on the link for my cache centroid (i.e. the centre point of all my finds).

 

At the moment, its only about a quarter mile from my home coords.

 

Now the sad part....

 

I'm going to deliberately do caches in a certain direction, to try and get the cache centroid as close to my home coords as possible.

 

Does anyone else actually monitor their centroid? And how far from home is it?

 

Presumably if you have done some far-away caches, Oz for example, it will throw your centroid out.

Link to comment

.. and I thought I was sad.....

 

Have you seen my collection of Banana labels?? It's amazing how many different ones there are...

 

...

 

If it's in OZ.. the opposite of the planet, would it not have an equal effect all round?

Good point, and yes, it does sound a bit sad....

Link to comment

Does anyone else actually monitor their centroid? And how far from home is it?

 

Yep, we must be sad too! Although ours is currently a few miles from home, it tends to hover near the location of the first cache we placed. At one point our centroid got to within 100 feet of it.

 

Regarding banana labels.....if anyone's got any swaps, we are desperate for a "Lush Buganda" label :):)

Link to comment

.. and I thought I was sad.....

 

Have you seen my collection of Banana labels?? It's amazing how many different ones there are...

 

...

 

If it's in OZ.. the opposite of the planet, would it not have an equal effect all round?

Good point, and yes, it does sound a bit sad....

 

No - it's the sort of thing I'd do..

 

One of my fondest memories of my late father, was asking him when I was 12 or so... "Does the Mileometer in the car go backwards when you reverse"? then being taken to a local large car park to reverse for half a mile to see it it did.. I don't think it's sad... I think it's scientific.. having a theory and having to test it out to if it works... I'd be stuffed in Cornwall I think as I can't move far North or South as the sea gets in the way!

 

Go for it and lets us know if you get the centroid dead centre on your house... it may be pretty pointless - but it's great fun trying.

 

Have FUN!

 

:):)

Link to comment

Does anyone else actually monitor their centroid? And how far from home is it?

 

Yep, we must be sad too! Although ours is currently a few miles from home, it tends to hover near the location of the first cache we placed. At one point our centroid got to within 100 feet of it.

 

Regarding banana labels.....if anyone's got any swaps, we are desperate for a "Lush Buganda" label :):)

 

Would that be the yellow or the rarer Blue or the almost mystical green version?

Link to comment

According to GSAK, my centroid is just south of the Isle of Man (in the Irish Sea). This is a somewhat strange as I have only found 52 caches north of these coordinates and 1001 south of them. I would love someone to explain how this can possibly be correct as some of the 1001 are a lot south of the coordinates and none of the 52 are very much north of them.

 

On the main subject, I think it's an excellent idea to cache by trying to manipulate your cache centroid! Like cricket, caching is heaven for stats lovers!

Link to comment

I'd not taken much notice of my centroid before this posting, but I've just had a look and plotted it on a world map. It is 747 miles NW of my house, about mid-way(ish) between Devon and Iceland at N57, W17.

 

This is confusing though, as I have never cached north of N56 before, and found almost 500 caches (a little bit under 20% of my total finds) in Los Angeles one week at N33, W118. I have also cached in Australia and have a handful at S33, E151. About 2200 caches are within a few degrees of N50, W2 near my home.

 

So, with all these "well south of N57" caches that I have found, how can my Centroid be at N57? I'm sure a mathematician may come forward and explain it in terms that the Wiki link failed to explain to me. Is it to do with the sphere that we live on? Gut instinct tells me that it would be somewhere near N44, W23.

 

Now who's sad? :):)

Link to comment

According to GSAK, my centroid is just south of the Isle of Man (in the Irish Sea). This is a somewhat strange as I have only found 52 caches north of these coordinates and 1001 south of them. I would love someone to explain how this can possibly be correct as some of the 1001 are a lot south of the coordinates and none of the 52 are very much north of them.

 

On the main subject, I think it's an excellent idea to cache by trying to manipulate your cache centroid! Like cricket, caching is heaven for stats lovers!

 

I'd not taken much notice of my centroid before this posting, but I've just had a look and plotted it on a world map. It is 747 miles NW of my house, about mid-way(ish) between Devon and Iceland at N57, W17.

 

This is confusing though, as I have never cached north of N56 before, and found almost 500 caches (a little bit under 20% of my total finds) in Los Angeles one week at N33, W118. I have also cached in Australia and have a handful at S33, E151. About 2200 caches are within a few degrees of N50, W2 near my home.

 

So, with all these "well south of N57" caches that I have found, how can my Centroid be at N57? I'm sure a mathematician may come forward and explain it in terms that the Wiki link failed to explain to me. Is it to do with the sphere that we live on? Gut instinct tells me that it would be somewhere near N44, W23.

 

Now who's sad? :):)

Any chance you guys have some Locationless Caches in your database?

Link to comment
Any chance you guys have some Locationless Caches in your database?

 

The macro output includes this text: "Note: excludes locationless caches", so I expect they are not included in the maths. Having said that, I have found some virtuals, but not very many at all compared to the rest of my finds.

Link to comment

So, with all these "well south of N57" caches that I have found, how can my Centroid be at N57? I'm sure a mathematician may come forward and explain it in terms that the Wiki link failed to explain to me. Is it to do with the sphere that we live on? Gut instinct tells me that it would be somewhere near N44, W23.

 

You can check the calculation by exporting the relevant caches to Excel (with the coordinate format set to dd.ddddd) and taking the average of the lat and long columns.

Link to comment

So, with all these "well south of N57" caches that I have found, how can my Centroid be at N57? I'm sure a mathematician may come forward and explain it in terms that the Wiki link failed to explain to me. Is it to do with the sphere that we live on? Gut instinct tells me that it would be somewhere near N44, W23.

 

You can check the calculation by exporting the relevant caches to Excel (with the coordinate format set to dd.ddddd) and taking the average of the lat and long columns.

 

Thanks for that Nick. I just did that for all my finds and it comes out at N47.5, W21.5 (hey, I was just 245 miles out with my gut instinct!). If I ignore the virtuals and locationless caches it comes out at N48, W22 (not much difference there). So... is the centroid calculation on the GSAK stats macro correct?

Link to comment

So, with all these "well south of N57" caches that I have found, how can my Centroid be at N57? I'm sure a mathematician may come forward and explain it in terms that the Wiki link failed to explain to me. Is it to do with the sphere that we live on? Gut instinct tells me that it would be somewhere near N44, W23.

 

You can check the calculation by exporting the relevant caches to Excel (with the coordinate format set to dd.ddddd) and taking the average of the lat and long columns.

Taking a simple average in Excel doesn't work - let me try and explain why. As the earlier poster correctly pointed out it's all to do with living on a sphere (or an oblate spheroid). Great circles and centroids on the surface of the Earth are often somewhat non-intuitive in their behavior.

 

Here's a simple example that might help you visualize what's going on - imagine two points, one in New York and the other at the same Latitude as New York but around the other side of the world at 180° different Longitude in China. Where is the mid point? It's NOT anywhere on that latitude - it's actually over the top at the North Pole. You can see that flying over the North Pole would be the shortest route to get from one to the other and that also tells you where the centroid is. Here's a map of that route.

 

In general when you find a new cache the centroid we use here will be 'dragged' along a great circle route between its previous position and the new cache. If those positions vary significantly in longitude then the great circle route will deviate significantly northwards or southwards from where you might expect. It's the same great circle routes that airlines fly to minimize route lengths.

 

In general centroids in the Northern hemisphere will be further North than they would be on a flat map (and where your intuition might think) and those in the Southern hemisphere will be further South. The further apart points are in longitude the more this effect comes into play.

 

The FSG macro works by converting every latitude/longitude point to 3D x,y,z cartesian coordinates using true WGS84 ellipsoidal math. All those 3D points are then averaged which gives a result under the surface of the earth. Finally that result is projected back up to the surface of the ellipsoid again using an iterative routine to get the result that the macro uses. (There are other ways to define different types of 'centers' but this was the simplest one to use here to give us a rough answer. The result will always be somewhat incorrect as we don't know the altitude of each point so there's no point in getting too anal about it)

 

With that caveat AFAIK it works correctly! :)

Link to comment

In general centroids in the Northern hemisphere will be further North than they would be on a flat map (and where your intuition might think) and those in the Southern hemisphere will be further South. The further apart points are in longitude the more this effect comes into play.

 

The FSG macro works by converting every latitude/longitude point to 3D x,y,z cartesian coordinates using true WGS84 ellipsoidal math. All those 3D points are then averaged which gives a result under the surface of the earth. Finally that result is projected back up to the surface of the ellipsoid again using an iterative routine to get the result that the macro uses. (There are other ways to define different types of 'centers' but this was the simplest one to use here to give us a rough answer. The result will always be somewhat incorrect as we don't know the altitude of each point so there's no point in getting too anal about it)

 

With that caveat AFAIK it works correctly! :)

 

Makes sense to me now, thanks for popping by to explain it :)

Link to comment

 

The FSG macro works by converting every latitude/longitude point to 3D x,y,z cartesian coordinates using true WGS84 ellipsoidal math. All those 3D points are then averaged which gives a result under the surface of the earth. Finally that result is projected back up to the surface of the ellipsoid again using an iterative routine to get the result that the macro uses.

 

:) And I thought my computer was slow taking 19 seconds to run the GSAK script. Now that I know it's doing things that the International Cartographers Society would be proud of, I'll give my home computer a bit more respect! :) As others have said, thanks for the explanation.

Link to comment

My Centroid turns out to be 150 miles west of South Uist in the Outer Hebrides, about 80 miles east of Rockall, just above the Anton Dohrn Seamount (that well-known Scottish peak!). About 350 miles from home.

 

My total cache-to-cache distance is 68349 Miles (excluding locationless).

Edited by Happy Humphrey
Link to comment

 

The FSG macro works by converting every latitude/longitude point to 3D x,y,z cartesian coordinates using true WGS84 ellipsoidal math. All those 3D points are then averaged which gives a result under the surface of the earth. Finally that result is projected back up to the surface of the ellipsoid again using an iterative routine to get the result that the macro uses.

 

:) And I thought my computer was slow taking 19 seconds to run the GSAK script. Now that I know it's doing things that the International Cartographers Society would be proud of, I'll give my home computer a bit more respect! :D As others have said, thanks for the explanation.

Agreed. Also respect to the folk who wrote the software to do all this. You know who you are. I never used the stats program before (odd, as I am a statistician by qualification) and it must've taken a bit of doing. Nice one. My centroid is at the same coords as a pub in Middlesbrough. Off we go, then. :)

Link to comment

Watch out for the armchair like caches and also any travelling caches... removing 5 from our 250 caches moved our centre about 150miles SE!!

Yes, I'd removed my only "armchair" cache from my finds some time ago so as not to upset my records. I actually signed a travelling cache recently too, but declined to log it on GC.COM for the same reason.

Link to comment

Watch out for the armchair like caches and also any travelling caches... removing 5 from our 250 caches moved our centre about 150miles SE!!

Yes, I'd removed my only "armchair" cache from my finds some time ago so as not to upset my records. I actually signed a travelling cache recently too, but declined to log it on GC.COM for the same reason.

I also found a travelling cache recently but altered the coordinates (and country) in GSAK to correct everything.

Link to comment

Watch out for the armchair like caches and also any travelling caches... removing 5 from our 250 caches moved our centre about 150miles SE!!

Yes, I'd removed my only "armchair" cache from my finds some time ago so as not to upset my records. I actually signed a travelling cache recently too, but declined to log it on GC.COM for the same reason.

The FSG macro provides a means to deal with these "armchair" caches. Just include the string 'FindStatsExclude' anywhere in the GSAK User Notes for that cache (Right click and select Add/Change/Delete note) and it will be treated as a Locationless for the purposes of the statistics and excluded everywhere Locationless caches are excluded. For example, like many folks I've found the 'Four Windows' cache in Germany but exclude it from the maps and centroid calculation using this method.

Link to comment

Don't forget...

 

The user can chose to exclude locationless caches from the 'Finds by		 
Distance', 'Map' blocks and 'Total Distance' figures.

 

This also applies to the centroid calculation.

 

Caches which you wish to be excluded like travelling and armchair caches can be excluded too by adding "FindStatsExclude" anywhere in the usernote B)

Edited by rutson
Link to comment

 

No - it's the sort of thing I'd do..

 

One of my fondest memories of my late father, was asking him when I was 12 or so... "Does the Mileometer in the car go backwards when you reverse"? then being taken to a local large car park to reverse for half a mile to see it it did.. I don't think it's sad... I think it's scientific.. having a theory and having to test it out to if it works...

 

:laughing::anicute:

And did it reduce???... something I've always wondered about.... did here rumours of people with drills winding the clock back :laughing::laughing: MaxKim

Link to comment

 

No - it's the sort of thing I'd do..

 

One of my fondest memories of my late father, was asking him when I was 12 or so... "Does the Mileometer in the car go backwards when you reverse"? then being taken to a local large car park to reverse for half a mile to see it it did.. I don't think it's sad... I think it's scientific.. having a theory and having to test it out to if it works...

 

:D:D

And did it reduce???... something I've always wondered about.... did here rumours of people with drills winding the clock back :unsure::ph34r: MaxKim

 

OK we're talking Triumph Herald Vitesses here - ! Buy yes it did...

Link to comment

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

Loading...
Followers 1
×
×
  • Create New...