Jump to content
Sign in to follow this  
Followers 5
Harry Dolphin

Cache Carnival

Recommended Posts

13 hours ago, The A-Team said:

 

That would be nice, but currently there's no way to search for caches by the FP percentage. They would need to give members some way of identifying the relevant caches (without having to use third-party tools).

 

Weird though, because the % is reported when you click the FP area on the cache page. 

Share this post


Link to post
2 minutes ago, lee737 said:

 

Weird though, because the % is reported when you click the FP area on the cache page. 

There's macro for that for GSAK too. Sorting on that value makes it easy to go for a cache with 8 favs and 10 logs over a cache with 50 FP and 1000 logs.

 

  • Upvote 1

Share this post


Link to post
2 hours ago, lee737 said:

 

Weird though, because the % is reported when you click the FP area on the cache page. 

 

Not weird at all.  The percentage feature is somewhat hidden, and only available one cache at a time, because of the drain on server resources to calculate favorite percentages for a large number of caches "on the fly" in search results.  Here is a prior post I made on this subject:

 

On 4/27/2014 at 4:58 PM, Keystone said:

You can find the requested tool for ranking caches by percentage of favorite points here on Project-GC.com. This is a CPU-intensive calculation over a large number of caches, so it's a perfect task for an offiste application that's a Geocaching Live! API partner. When this has been requested as an on-site feature, Groundspeak has always pointed to the database hit as the reason for not developing the request.

  • Helpful 2

Share this post


Link to post

If you're looking for a mathematically relevant high rating based on favourite points, look at the Wilson score.  PGC has a page for searching by this rating.  Percentage alone isn't as relevat, since a cache with 1 log and 1 FP is 100%, which is less informative or helpful than a cache with 100 logs and 80 FPs, or 1000 logs and 400 FPs, even.  The Wilson score is the one that'll generally indicate a more reasonable rating based on ratio of FPs and logs.

  • Helpful 2

Share this post


Link to post
7 hours ago, thebruce0 said:

If you're looking for a mathematically relevant high rating based on favourite points, look at the Wilson score.  PGC has a page for searching by this rating.  Percentage alone isn't as relevat, since a cache with 1 log and 1 FP is 100%, which is less informative or helpful than a cache with 100 logs and 80 FPs, or 1000 logs and 400 FPs, even.  The Wilson score is the one that'll generally indicate a more reasonable rating based on ratio of FPs and logs.

 

Yes, for my area at least I really can't argue against the Wilson score. All its highly rated ones are caches I'd recommend to others and are mostly ones I've favourited myself. It's interesting to compare the scores of the top thirty in my region (the NSW Central Coast) with that of central Sydney where most of this state's high FP count caches are and hence are drawing the most attention in this current promotion:

 

image.png.5bc46946bfbdf26b5dd7c02b63612ecf.png 

 

Only two of those caches in the Central Coast's top 30 Wilson scores have had any finds during the promotion, and those were by people attending the local CC event last Saturday.

Share this post


Link to post
9 hours ago, Keystone said:

 

Not weird at all.  The percentage feature is somewhat hidden, and only available one cache at a time, because of the drain on server resources to calculate favorite percentages for a large number of caches "on the fly" in search results.  Here is a prior post I made on this subject:

 

 

It is obvious that % favourites is going to be a far superior measure of cache quality, surely the database could calculate the % for each cache after each find and store it in an easily searched field?

Share this post


Link to post
9 hours ago, thebruce0 said:

If you're looking for a mathematically relevant high rating based on favourite points, look at the Wilson score.  PGC has a page for searching by this rating.  Percentage alone isn't as relevat, since a cache with 1 log and 1 FP is 100%, which is less informative or helpful than a cache with 100 logs and 80 FPs, or 1000 logs and 400 FPs, even.  The Wilson score is the one that'll generally indicate a more reasonable rating based on ratio of FPs and logs.

Yes - this is obvious, when talking about using %FP as a metric, it is assumed you need some sort of minimum to avoid the 1/1 situation. I'd say though once you get to 4/5 or 5/6 the cache is probably worth checking....

Share this post


Link to post
7 hours ago, lee737 said:

Yes - this is obvious, when talking about using %FP as a metric, it is assumed you need some sort of minimum to avoid the 1/1 situation. I'd say though once you get to 4/5 or 5/6 the cache is probably worth checking....

There is math behind the Wilson score that does all that. You should look into it if you're interested in how it handles weighting :)

  • Helpful 1

Share this post


Link to post
Posted (edited)
11 minutes ago, thebruce0 said:

There is math behind the Wilson score that does all that. You should look into it if you're interested in how it handles weighting :)

Unfortunately, The Project GC page does not specify what confidence interval they are using, and although this should not affect the sort order it does make the scores not especially meaningful.

Edited by fizzymagic

Share this post


Link to post

More meaningful than straightforward raw counts of FPs or percentages though; if you care about that meaning at least. (I dno't really use the Wilson score option anyway; but it does inherently have more meaning than the raw numbers).  Unless of course one very much disagrees with how they weight the algorithm.

Share this post


Link to post
9 hours ago, thebruce0 said:

More meaningful than straightforward raw counts of FPs or percentages though; if you care about that meaning at least. (I dno't really use the Wilson score option anyway; but it does inherently have more meaning than the raw numbers).  Unless of course one very much disagrees with how they weight the algorithm.

It would be nice to see how my caches compare.  For that I would need the confidence interval they used.  I agree that the Wilson score is a better indicator of the actual rating that the average, though!

Share this post


Link to post
On 3/29/2019 at 3:11 PM, barefootjeff said:

Project-gc shows there are 8 caches in my region that have 70% or higher FP percentage, although it only includes those with ten or more FPs - extending that down, I can add another 9 to the list which are either quite new or very high D/T ones that have only had a handful of finders. These 17 I'd consider the cream of the crop from my region, but I'd wager that none of them get any finds during the promotion period.

 

So just to wrap up, how did those 17 caches fare across the three weeks of the promotion? These are the final find counts on them:

 

GC Code      Number of FPs     Pecentage FPs    Number of finds during this promotion

GC7H4RW           6                               100%                                         0

GC7YP51            4                                100%                                         0

GC5TMBE          10                                91%                                          0

GC68VZP            7                                 88%                                          0

GC5HW5M         19                                83%                                         0

GC6XHHJ            4                                 80%                                         0

GC6T5PZ             4                                 80%                                         0

GC5E7A3             7                                 78%                                         0

GC62WZJ           19                                76%                                         0

GC6WPQ5            3                                 75%                                         0

GC5JZQM           17                                74%                                         0

GC4ATHC            11                                73%                                         0

GC59NNC            24                                73%                                         0

GC6RKKH            13                                72%                                         0

GC6XVK3              5                                 71%                                         0

GC6JMDK           20                                 71%                                         0

GC7GT14              7                                  70%                                         0

 

So yep, none of them got a single find so I'd have won my wager. Apart from the local CC event which I thoroughly enjoyed, this promotion pretty much killed caching here as everyone chasing points sought out the higher FP count caches in the more populous regions.

Share this post


Link to post

Another factor to keep in mind is how many active cachers have already found the high FP caches. It's not conclusive to say (imply) a promotion was a failure just because a list of high-fp caches in some location weren't found. Far too many unanswered questions.  It's going to be a global statistical analysis HQ will look into, and general social media reception and feedback.

Share this post


Link to post
6 hours ago, thebruce0 said:

Another factor to keep in mind is how many active cachers have already found the high FP caches. It's not conclusive to say (imply) a promotion was a failure just because a list of high-fp caches in some location weren't found. Far too many unanswered questions.  It's going to be a global statistical analysis HQ will look into, and general social media reception and feedback.

 

Almost half the caches in my list of 17 have had less than ten finds so I don't buy your argument that everyone's already found them. The highest find count on any of them is 33 (GC59NNC); by contrast, most of the ones around Sydney harbour that attracted all the action in this promotion had find counts in the thousands but weren't especially creative or diverse. Most were either virtuals (old and new) with a basic logging task (post a photo of yourself and GPSr at GZ) or a basic traditional with a harbourside view. The promotion's theme was "Celebrate geocaching creativity and diversity around the world!" but all it did was attract people away from the truly creative and diverse caches in the regions to the quick-and-easy caches in big city tourist hotspots.

Share this post


Link to post

Yeah, see, I feel for you, but that didn't refute the point - yes there are areas that won't be super-positively impacted. So? NO strategy will be 100% beneficial around the world in every region for every cache that any person feels should be benefitted. I don't know why the promo didn't seem to work so well in your region. In stark contrast, in my area most caches with over 50 points have seen increased finds and FPs for the period. Other new caches have been placed that are more creative than usual. People have been more actively seeking creative caches in general, than just going out to find random caches for a day.

This isn't a problem with hq's promotion or strategy, dare I say it's just unfortunate for your community that it's not taking hold and having the effect that it is in other regions.  I can't agree with the implication that this promotion is failure because you're not seeing a benefit in your region. It's unfortunate, it sucks, but you do what you can do and try to avoid pessimism especially when there's good happening elsewhere. Maybe HQ will take the results in your region into consideration for next time, who knows.

  • Upvote 1

Share this post


Link to post
On 4/16/2019 at 8:15 AM, barefootjeff said:

 

So just to wrap up, how did those 17 caches fare across the three weeks of the promotion? These are the final find counts on them:

 

......

 

So yep, none of them got a single find so I'd have won my wager. Apart from the local CC event which I thoroughly enjoyed, this promotion pretty much killed caching here as everyone chasing points sought out the higher FP count caches in the more populous regions.

 

To be fair, you can't really say that this promotion had a meaningful negative effect on these caches, they are a group of scarcely found caches at any time - hard puzzles, hard terrain, long walks, complex multis. A lot of them only had 1 or 2 finds in the year before the promotion..... No doubt they are good caches, and also no doubt the promotion didn't give them any assistance, but even if each of these had 50 points on them, I doubt many more would have been found, for the same reasons they weren't being found last year.... We like these sort of caches (maybe not so much the hard puzzles) but don't have the time to do as many as I'd like to..... especially when we needed 500 points! :)

  • Upvote 1

Share this post


Link to post
On ‎4‎/‎16‎/‎2019 at 10:15 AM, barefootjeff said:

 

So just to wrap up, how did those 17 caches fare across the three weeks of the promotion? These are the final find counts on them:

 

GC Code      Number of FPs     Pecentage FPs    Number of finds during this promotion

GC7H4RW           6                               100%                                         0

GC7YP51            4                                100%                                         0

GC5TMBE          10                                91%                                          0

GC68VZP            7                                 88%                                          0

GC5HW5M         19                                83%                                         0

GC6XHHJ            4                                 80%                                         0

GC6T5PZ             4                                 80%                                         0

GC5E7A3             7                                 78%                                         0

GC62WZJ           19                                76%                                         0

GC6WPQ5            3                                 75%                                         0

GC5JZQM           17                                74%                                         0

GC4ATHC            11                                73%                                         0

GC59NNC            24                                73%                                         0

GC6RKKH            13                                72%                                         0

GC6XVK3              5                                 71%                                         0

GC6JMDK           20                                 71%                                         0

GC7GT14              7                                  70%                                         0

 

So yep, none of them got a single find so I'd have won my wager. Apart from the local CC event which I thoroughly enjoyed, this promotion pretty much killed caching here as everyone chasing points sought out the higher FP count caches in the more populous regions.

 

OK I haven't read all this topic, so apologies if I'm way off the mark here, but...  All those caches might be high in FP% but they are pretty low in overall FP.  This promotion encouraged finding caches with high FP not high FP%.  I'm betting there are caches in the same geographic area with much higher outright FP...?  So really, any result whether these 17 caches got lots of finds or not could only be at best indirectly linked to the promotion.  The fact they didn't get any finds can't really be at all linked to the promotion, after all the promotion was at least encouraging finds on any caches.

 

I think suggesting that the promotion killed caching there might be a bit of a stretch.  Over the 3 weeks (was it 3 weeks?) the promotion was running, you could get your 500 points from 10 highly favourited caches (>50FP each) and then you're done with needing points.  I doubt anyone decided they needed to forgo finding some good local caches because they didn't have time for them as they needed those 3 weeks to hunt for 10 caches in the city.  I know I didn't particularly modify my caching during the promotion (ok, on the last day or 2 of my overseas holiday I did specifically seek out a handful of the high FP caches where I was, but then they were largely great caches so I'm glad I did) and I ended up with over 3600 points.

  • Upvote 2

Share this post


Link to post
45 minutes ago, lee737 said:

 

To be fair, you can't really say that this promotion had a meaningful negative effect on these caches, they are a group of scarcely found caches at any time - hard puzzles, hard terrain, long walks, complex multis. A lot of them only had 1 or 2 finds in the year before the promotion..... No doubt they are good caches, and also no doubt the promotion didn't give them any assistance, but even if each of these had 50 points on them, I doubt many more would have been found, for the same reasons they weren't being found last year.... We like these sort of caches (maybe not so much the hard puzzles) but don't have the time to do as many as I'd like to..... especially when we needed 500 points! :)

 

I agree, only six of those seventeen caches have had any finds this year, but a few of those were found in the weeks leading up to the promotion and the fact that none of them got any finds during it still suggests some potential finders might have been diverted away. I guess my gripe is mainly with it being promoted as showcasing the top creative and diverse caches when in reality it just seemed to be all about getting quick and easy finds on the most popular ones. Maybe it's just me, but the most diverse and creative caches I've done have been the ones that the CO has put a lot of effort into and which in turn require some effort back from the finder, either in solving a satisfying puzzle or the time and exertion of getting to GZ. They're the sort I give FPs to; perhaps it's telling that I only awarded one FP across the sixteen caches I found during the promotion, and that was one I would have done then anyway, whereas my average FP awarding rate is right on the limit at ten percent of my finds. I just didn't see much connection between most of the caches I found over those three weeks and what I consider "diversity and creativity". Had the promotion been called "Fun with Favourites", "Popularity Stakes" or some such thing it would have sat a lot better with me.

Share this post


Link to post
Just now, barefootjeff said:

I guess my gripe is mainly with it being promoted as showcasing the top creative and diverse caches when in reality it just seemed to be all about getting quick and easy finds on the most popular ones. Maybe it's just me, but the most diverse and creative caches I've done have been the ones that the CO has put a lot of effort into and which in turn require some effort back from the finder, either in solving a satisfying puzzle or the time and exertion of getting to GZ. They're the sort I give FPs to; perhaps it's telling that I only awarded one FP across the sixteen caches I found during the promotion, and that was one I would have done then anyway, whereas my average FP awarding rate is right on the limit at ten percent of my finds. I just didn't see much connection between most of the caches I found over those three weeks and what I consider "diversity and creativity". Had the promotion been called "Fun with Favourites", "Popularity Stakes" or some such thing it would have sat a lot better with me.

 

GS can only use the data they have in order to produce a scoring regime that attempts to push what they are promoting.  It seems they decided FPs was the best they had in this case.

 

I'm thinking your real gripe might be at least partly with the way people award FPs to caches in the first place.  If so, then I'd largely agree with you on that.  Mingo is a case in point, along with many other oldest-in-the-area caches.  It's a pretty crappy cache really and even breaks guidelines, but it's got lots of favourites because it's the oldest active and that makes people enjoy it.  And of course the more visits, the potentially more FPs (and then more FPs encourage more visits, etc).  But at the end fo the day, people award FPs to whatever they've decided was their favourite caching experience...

  • Upvote 3

Share this post


Link to post
1 minute ago, funkymunkyzone said:

I'm betting there are caches in the same geographic area with much higher outright FP...?

 

No, there are are only two caches in the region with 50+ FPs so one really had to leave the area to have much chance of getting all the souvenirs. With one exception, a traditional close to a motorway rest stop, activity across my 32 hides in the region has been pretty much dead over the past three weeks.

Share this post


Link to post
Posted (edited)
2 minutes ago, barefootjeff said:

 

No, there are are only two caches in the region with 50+ FPs so one really had to leave the area to have much chance of getting all the souvenirs. With one exception, a traditional close to a motorway rest stop, activity across my 32 hides in the region has been pretty much dead over the past three weeks.

 

Well there's a big difference between 3 FP (one from your list) and 50+ FP.  I bet there's a LOT of caches with more than 3 FP.... right?

 

And I stand by my assertion that it would not take all of the 3 weeks of the promotion to make one visit to the big smoke to pick up 10 high FP finds.  One day perhaps...

Edited by funkymunkyzone

Share this post


Link to post
Posted (edited)
21 minutes ago, funkymunkyzone said:

I'm thinking your real gripe might be at least partly with the way people award FPs to caches in the first place.  If so, then I'd largely agree with you on that.  Mingo is a case in point, along with many other oldest-in-the-area caches.  It's a pretty crappy cache really and even breaks guidelines, but it's got lots of favourites because it's the oldest active and that makes people enjoy it.  And of course the more visits, the potentially more FPs (and then more FPs encourage more visits, etc).  But at the end fo the day, people award FPs to whatever they've decided was their favourite caching experience...

 

No, most of the caches in this state (and probably the whole country) with 50+ FPs are in tourist hotspots and got them as a tiny percentage from thousands of finds.

 

18 minutes ago, funkymunkyzone said:

Well there's a big difference between 3 FP (one from your list) and 50+ FP.  I bet there's a LOT of caches with more than 3 FP.... right?


Okay, here's the stats from project-gc for the Central Coast region:

 

5-9 FPs       87 caches

10-24 FPs  74 caches

25-49 FPs  10 caches

50+ FPs       2 caches

 

The one at the very top of the list, with 55 FPs, got eight finds during the promotion, and a few of the others got finds from people heading to the local CC event, but most of the rest got zilch. As you said, the big incentive was for anyone chasing the souvenirs to head down to Sydney harbour and grab a bunch of easy high FP-count caches there.

Edited by barefootjeff

Share this post


Link to post
29 minutes ago, barefootjeff said:

 

Okay, here's the stats from project-gc for the Central Coast region:

 

5-9 FPs       87 caches

10-24 FPs  74 caches

25-49 FPs  10 caches

50+ FPs       2 caches

 

The one at the very top of the list, with 55 FPs, got eight finds during the promotion, and a few of the others got finds from people heading to the local CC event, but most of the rest got zilch. As you said, the big incentive was for anyone chasing the souvenirs to head down to Sydney harbour and grab a bunch of easy high FP-count caches there.

 

TBH I think if you wanted to draw any kind of statistical conclusions about the success of the promotion, you'd have to pick a region with a lot more caches and general caching activity, so that you can actually see a statistical difference between caching activity within and outside the promotion period.

 

From the limited data presented to me, a more highly favourited cache in your area got 8 finds, while caches with much fewer got less or none.  That may be normal, or it might be in part an effect of the promotion.  But you'd have to look at a much larger area or one with lots more caches and ongoing caching activity.  You said yourself, most of the 17 caches you highlights had no finds at all all year, so you can't conclude anything from the fact they continued to have no finds during the 3 week promotion...

 

  • Upvote 2

Share this post


Link to post

Awarding points to placers was a game changer for us. Having long ago found most of the highly favorited in our area that we can do, we thought we might pick up a couple of the souvenirs. We were stunned  to receive enough favorites on our hides to get the final souvenir, including 8 on a cache we released during our cache celebration event. 

  • Upvote 3

Share this post


Link to post

I've just noticed that I have three Cache Carnival souvenirs. I haven't taken any notice of the Cache Carnival thingy and haven't found any caches in over a month. How did this happen?

Share this post


Link to post
1 hour ago, colleda said:

I've just noticed that I have three Cache Carnival souvenirs. I haven't taken any notice of the Cache Carnival thingy and haven't found any caches in over a month. How did this happen?

 

People awarding FPs to your caches during the promotion period would have given you points.

Share this post


Link to post
21 hours ago, barefootjeff said:

 

People awarding FPs to your caches during the promotion period would have given you points.

ok. Cool. Thanks

Share this post


Link to post

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...
Sign in to follow this  
Followers 5

×
×
  • Create New...