There is a much more fun mefi game than trying to find the users who joined on the same day as you
, namely computing your Cryptic FPP Advantage.
posted by jeffburdges to MetaFilter-Related at 2:20 PM (138 comments total)
5 users marked this as a favorite
First, sort your past FPPs into cryptic vs. non-crptic1 columns, placing the number of favorites into the appropriate column. Next, average each column, assuming you've enough data, and post the ratio avg(cryptic) / avg(non-cryptic) here.2
That's how many times more favorites your cryptic posts get than your non-cryptic posts. Let call it your Cryptic FPP Advantage γ.3 I'm batting γ = avg(cryptic) / avg(non-cryptic) = 2.74 if I compute back to 2006. It drops back around 2.4 if I ignore more early years with few favorites or cryptic FPPs.
I'll conjecture that most people who FPP sparsely, and only occasionally crypticly, have a γ > 2, but I'm reluctant to extend the conjecture to very frequent FPPers. If you don't have enough FPPs, you could try the same game with snarky vs. non-snarky comments, but that'll take too long for many people.
1 Yes, it's obviously subjective whether or not a fpp is cryptic, but here are some guidelines : (1) consider your intentions when posting, (2) non-cryptic [more inside] material doesn't count, we're only worried about the fpp itself, (3) single letter links help make a post cryptic, but might not suffice, especially not in a (previously), and (4) double check yourself if you got called out for crypticness on post you don't consider cryptic, but don't just declare it cryptic though. Your callout ratio data isn't so easy to obtain. 2 We only want the ratios not the averages because the bare averages would distract us into **** measuring and the ratios should be more consistent across different users. And we don't want to complicate the statistics more than necessary. 3 Yes, I know γ gives a crappy gamma. :(