Better Favorite Search Criterion

[+Bluespade00]

I recently saw a link on the website for "most favorited" caches or something to that effect.  This statistic as a search criterion is a little wonky.  It skews results to those with high find counts.  A better statistic for a search criterion would be highest favorite to find ratio.  This will include many of the same caches from the "most favorited" but also  caches that have lower favorite points only because they are more difficult or remote but high percentage of finders who awarded favorite points.

[+Hügh]

I'd love to see this integrated into Geocaching.com, but since the functionality is offered by Project-GC, it may be a while.

 

https://project-gc.com/Statistics/TopFav (Ordered by raw FP count)

 

https://project-gc.com/Statistics/TopFavPct (Ordered by FP percentage: essentially FP / PM finds with some adjustments)

 

https://project-gc.com/Statistics/TopFavWilson (Ordered by the lower bound of the Wilson score confidence interval for a Bernoulli parameter, which apparently is "the best" way. I don't know. Math.)

Edited August 19, 2022 by Hügh

[+niraD]

27 minutes ago, Bluespade00 said:

I recently saw a link on the website for "most favorited" caches or something to that effect.  This statistic as a search criterion is a little wonky.  It skews results to those with high find counts.  A better statistic for a search criterion would be highest favorite to find ratio.  This will include many of the same caches from the "most favorited" but also  caches that have lower favorite points only because they are more difficult or remote but high percentage of finders who awarded favorite points.

This has been suggested before, for example:

 

[+TeamRabbitRun]

Just a reminder, it shouldn't be a ratio of FAVES to FINDS (or the other way around; whatever), it has to be a ratio of FAVES to PREMIUM MEMBER FINDS, since only PM's can award FAVES, as displayed on the Cache Page: