Quoted By:
So I've seen a lot of people posting over the poll and whether or not its results are representative. So I decided to run some numbers.
First. The argument:
>Polling region by region is biased against mons from larger regions (read: Charizard) because there are less votes to go around. Leading to mons from smaller regions (read: Greninja, being from the smallest generation) being over represented in the vote.
This assumes.
>Due to each voter being able to vote once for each region, each region *should* have the same number of votes in total.
Based on that, let's begin.
We start with the total number of votes, Vt. Then we distribute those votes among the mons in that region's section of the dex, D, to get an average.
The actual votes received, Vr, are of course different to this average. This difference is represented by the Popularity Factor, P.
Putting that together we get.
Vr = PVt/D
Or
PVt = DVr
Since we don't have Vt (either regional or total from a brief google search) we will instead combine it with the P-factor for a P-score. Which is the metric we're using here.
P-Scores
Charizard - 14,189,168
Umbreon - 6,706,200
Fug - 8,226,765
Lucario - 10,941,713
Chandelure - 7,947,108
Greninja - 10,120,248
Mimikyu - 8,718,776
Dragapult - 4,695,813
This still isn't an accurate model, because 1, we cannot assume that each region gets the same number of votes. If anyone has the regional votes however, you can find the P-Factor and see how that pokemon weighs against its region.
One could argue that each region did get the same number of votes, and any empty votes are essentially "there is nothing here i like" cast as a vote. In which case, P-score is the most accurate descriptor of popularity. So there, charizard can have a circumstance underwhich they "win" for what that's worth to you.
inb4
>butthurt charitard bullshits until his fat lizard wins
Sorry friend
luv maths
luv ma waifus
luv stantler
ate shitposters
Simple as
