For instance, if we look at Lithodora's PK Statistics, it looks like AP is super deadly to thief. But I don't believe for a second that is generally true. Chances are it was because that particular thief faced a particularly difficult to avoid and beat AP. Namely, Sodsob.

I'm clearly cherry picking, and with PBF's as the only data source, we all have to. But as an Imm, you should be able to get access to a better data set.

Then you need to break apart stuff like what is assassin good at fighting overall, not just on PBF chars, but all chars. How does that change at major miletstone... Levels, edges, losing XP penalty, etc. It is entirely possible the data doesn't support my feeling that assassin has moved from good to a bit OP (under the right circumstances, which includes a couple of edges).

Also, I didn't break it down for all the goodies you mentioned (most of which are pretty old at this point), but Air offense shifter is a particularly bad matchup for assassin for obvious reasons. Doubly so when you consider jack blaguar, and Oshui's ratio vs assassins was still 92% which is going to be well within the margin of error for the sample size vs his overall ratio of 88%. He also was only 50% vs necros, which is also clearly a matchup that doesn't favor shifter in most cases. Conjurer, Ranger, Polespec (perma-ing) warrior, are all unfavorable matchups, but all you have to look at is who most wins are against to realize the game is mostly warriors, and has been for quite a while.

I'd love to see some real analysis on the total data set, but that's impossible for us, so it's really up to you.