Some observations:
I've skimmed over a majority of this discussion. I apologize if something I say has already been mentioned. Consider it my agreeing with you.
Perhaps some of us have lost sight of what we're actually talking about. AFAIK nobody asked Qvist to do these rankings, he thought it would be a good idea and did the work of collecting data from everybody, compiling it, computing all the relevant statistics he could think of, AND putting together a write-up of EVERY card given those statistics. The write-ups are thoughtful and insightful not only for people new to the game (I recommend the lists in my Dominion video tutorial series for people who want a starting point for which cards are more likely to shape the course of a game) but also for experienced people to say "RAWR I THINK SCOUT SHOULD BE HIGHER!" and express the different ways the view the game. The different ways we can interpret the data, agree/disagree with it, and still all end up agreeing that it's a great thing goes to show the awesome thing that Qvist has put together and put so much work into.
I'm worried that he might be put into a position where no matter what he decides to do with the +cost cards, some people will think it's wrong. I'm not saying that's happening because there's a lot of text here that I haven't read, but none of us are entitled to any of this. It's Qvist's decision what to do and pretty much no matter what he does, I'll probably support it and continue to try and find good conclusions to make from it, as I have done in the past.
If you want my opinion: card costs, what these categories attempt to distinguish by, are a "cost to entry." What does it take to get these cards? If you can do more to get them then sometimes you get a benefit on top of that (this goes beyond overpaying, like Fool's Gold doesn't work well unless you have a lot of them. That means that $2 and 1 Buy is not the "cost to entry" for FG, it's higher. Chapel is different. We already take that into account in the rankings). I don't understand how overpaying is any different. Of course cards like Masterpiece and Peddler are going to be awkward, but Qvist is already able to paint the broad strokes just by compiling stats from tons of people with different opinions, and he does a good job of going a step further buy inferring situational cards by high deviation and other stuff.
I think the list works, and we should let it continue to work and show nothing but support.
I like fish.