I don't really like ratings being blended across various kingdom selection methods. If I had gone onto my old councilroom page and found absolute homogeneity on my "win % given available"s, I would probably figure it doesn't make a difference, but given that there are some differences, I think it does. And I think it might even be extra true for people long me in the middle who have mastered some strategies but not other. With the Obi Wan Bonogi thing it's like, well I'm sure he know how to play a BM mirror just fine, he probably just doesn't want to, his rating would be very near that, so whatever. I don't really want a play experience where people who are rated similar to me actually have a subset of skills of the game that are their favorite, and have that box checked.
I'd like to see separate ratings for different kingdom selection methods. Maybe they would all come out super similar, and I'm just so terribly wrong and overly particular, but you can tell me that as many times as you want then. That's just how I feel.
It seems like a good minimum or compromise would be for a 1/N chance for each of N participant's preferred kingdom selection method to be used, so that Prosperity lovers or Prosperity haters, whichever are the minority, don't get as much of a reduced rating for playing their favorite when they can and just sucking it up when they have to play someone else's preferred setting to get a game going.