***If you want to see stats for a metagame that isn't listed, shoot me a PM, and I'll throw it up. It literally takes me two minutes.***
Hi there, folks!
It's September, and I'm back with more stats. Same as last month, they're culled from battle logs, so only pokemon that appear in battles are actually counted. This is less than ideal, but we have strong reason to believe that, by re-normalizing the stats so that the sum of all usage percentages is 600% (which I call
MaestroXXVI's Method), the resulting stats should be roughly equivalent to what would have been generated using the old method, which took into account ALL pokemon on a team, whether it appeared in battle or not (the reason we don't think it matters is because we looked at what would happen if we took into account only teams where all six pokemon appeared. The results were almost identical. For more, read
here).
The scripts that I used to generate this data can be found
here. If you see something that looks wrong, say something. I'm more than willing to double-check my code.
So now, without further ado, I present you with the Smogon PO usage stats for September 2011:
By request of waterwarrior...
By request of Sniffle Puzz
By request of HuntSaboteur,
Full Team Stats
For comparison, I've done a separate run where I only count full teams (teams where all six pokemon appear):
1337 Stats!
MaestroXXVI, among others, asked what would happen if I only counted teams where the trainer was in the top echelon of the ratings. So I chose an arbitrary cutoff (1337), and applied a rule wherein only trainers with a rating at-or-above the cutoff at the time I dled the ratings (Oct. 2) got their teams counted. Below are the results. Note that there were only 123 users with a ranking at-or-above 1337 when I dled the ratings.
I can, in principle, do this for the other tiers as well, but I doubt I'm going to. It's too much of a hassle to retrieve the rankings. I can also move the cutoff up or down (it's arbitrary, after all), but I probably won't unless it's requested.
The inverse of this would be "troll stats."
Lead Usage Statistics!
I'm actually surprised more of you didn't harass me about this. Keep in mind when comparing to above stats--the sum of all columns should add up to 100%, not 600%.
Link to previous month's stats
August 2011