I’M ADDING ONE MORE THING, I did this in a game to Leafia once (while hydraing with Zug) and found her to be heavily postcount polarised over like 23 games, 141 ppd as wolf vs 66 ppd as town with a standard deviantion of 52 ppd. Now, obviously there is overlap between her wolfrange and townrange there, any game could be an outlier. A postcount number alone is unreliable evidence! It’s not usually going to get you somebody’s alignment!
But what this information is useful for is indicating underlying causes of its fluctuations. Leafia posts more as wolf because she has more to say as wolf, because she’s more motivated as wolf, because she gets into extended arguments and doesn’t let things go as wolf. And the postcount thing is a flag to look for more specific behaviours than just “posting a lot” or “posting a little”.
It’s similar to other “objective tells” you can spot: Leafia uses the word “gamethrowing” as wolf much more than town, not because there’s some forcefield making her, but because she more often tries to shut down people’s arguments by saying “you are literally gamethrowing by doing [X]” (this was discovered through manual forum search in a BotF I was playing and not through any kind of forbidden tool, I would not be allowed to mention it if it were). The exact word “gamethrowing” is volatile and unreliable, but trying to paint other people as “objectively in the wrong” when they are not? That comes from TMI, and you can use that to make good reads in games!
People bash this kinda thing in mafia all the time because they assume people are trying to use numbers alone to make reads, which is not super reliable and is easy to jam the signal of once you’re aware of it. But numbers can be the basis of more thoughtful arguments! They can give you insight into people’s minds in the same way looking at their posts can! Let me have my numbers. I deserve my numbers