I don't know of any conclusive tournament evidence. That's my point, though.
Sure, there are theoretical advantages that we can GUESS would make characters broken. Sometimes it may *seem* obvious. But until someone abuses it in tournament, we can't be sure.
Can't we? We know what situations are common, we know what moves break those situations, we know the fall-back options.
Keep in mind that street fighter is a far more controlled game due to having considerably less movement and therefore it's easier to examine from a MU prospective.
I'm trying to get people to quit making decisions based merely on their own personal judgments. Doing so biases the decision based on who is making the observations.
If people's biases are coming into play, then get a better paper. We know that MU numbers are less then reliable in smash, and I'm personally made an effort to improve on that (nobody seems to wanna help though).
But that is not the case in SF, MU numbers are a GREAT DEAL more reliable in SF.
But the problem is, your "in practice" numbers have much deeper issues relating to statistical validity, and you can never have perfect statistical conclusions anyway, it still only proves correlation and we can't know that player population isn't a significant influence.
anti-ban doesn't want to ban MK because he doesn't *seem* broken. But we do have data on MK, unlike all these other things we've banned right off the bat. And the data not only shows that MK is on top, it shows that he's on top by a ridiculous margin.
Now why is that? Some have argued that it's merely because MK is popular. Well that makes a difference, but not as much as you'd think. I already gave my popularity speech a long time ago, but it boils down to it doesn't matter how popular or unpopular a character is if they always lose. Same if they always win. (that's just a summary, popularity DOES have an influece, it's just that the influence is small)
You've made every statistician in the world cry.
If a character always loses, yea you're right. But no character always loses, there's always that small percentage, especially when you account for varying skill, the fact is that as you increase the number of people maining a character the raw number of players of amazing skill increases while the ratio of good to bad players simply remains the same.
Again, by this reasoning, you should take up smoking to avoid cardiovascular disease.
That's why unless you account for population size, any correlation drawn is presumed invalid, period.
The word "bannable" is kind of weird to me, first of all because it isn't a real word, and second because any character or anything is bannable. If we collectively decided we don't like Lucas' hairstyle (I mean it is pretty 1974) we could ban him. If we decided we didn't like people grabbing ledges at all we could ban that, and say that anyone who grabs a ledge loses a set. Anything is "bannable," and any criteria is going to be arbitrary even if it's an official criteria.
So please for the love of Sakurai stop saying things like "he doesn't fit ban criteria" or "he isn't gay enough" or "he isn't broken enough" or "he isn't overcentralizing" because these are all just expressions of a personal ban criteria and mean absolutely nothing in an argument like this. Actually, none of this really means anything, but we do have a lot of data showing that a lot of people play MK and he wins a lot. The only thing we should be discussing in this thread is whether or not it's too much, and whether or not micromanaging MK's strengths to keep him playable is worth M2k's bankroll.
Which is exactly why a large number of people don't want him banned, because it's acknowledged that there is no rigid universal criteria, and banning without such is ridiculous because that means it comes down to philosophy.