Why wouldn't they add PIchu? Plenty of reasons.
Keep this in mind. It'd take several hundred hours to add a new character. The time and effort must be justified. Both Roy and Mewtwo have had significant demand. Not only were their characters more popular, but their movesets, while weak in melee, just needed a few tweaks for improvement. As expected, they got a huge amount of hype in their trailers.
Well, Mewtwo took 700 hours, so we can safely say that any character that needs to be made from the ground up like that will take 700. In fact, for someone like Isaac, who would need a moveset made from scratch to do him justice, as well as specials the game has never had before, it may be several hundred more hours.
Roy took 300 hours, even though he was a clone, but he has a lot of new animations, too. Different tilts, aerials, smashes, and a totally revamped side-B.
Pichu wouldn't need new animations on anything. He wouldn't even really need to be thoroughly QA tested to make sure he's well balanced. The sense I'm getting from most Pichu fans in this thread is that they think it'd be really cool if he was a viable pick, but they'd rather have him come back as a derpy gimp than not have him come back at all.
So at that point, how tough is Pichu? You have to re-create the self damaging mechanic, which admittedly might be tough because there are only a few characters who have that on them (G&W's hammer hitting 1, for example, or Ike charging up his B). So if that's a specific property like crawling or wall-jumping that can't be given to other characters, then Pichu's probably impossible to bring back, and there's no reason for the PMBR to drive themselves crazy trying to code around that.
But if it's not, I really think Pichu is the
easiest choice for the PMBR, of all these characters. He doesn't need any different moves, just tweaks to his physics and the additional of the self-damaging mechanic. If the PMBR wants to play with those variables until he's in a place where he's a decent pick, that'd be nice! But it's not really necessary in this case.
So I don't think the comparison should be "let's pit Pichu against Isaac/Dixie/Lyn/Ridleyetc, and see who's more popular." I think it's "for a much smaller time investment than a full-bodied character, we can get an extra one! On top of that, he's a character fans are already familiar with, and while controversial, he has his following."
I could see Isaac taking anywhere from 700 - 1000 hours, since Mewtwo's moveset was already mapped out, so at least the PMBR wasn't inventing stuff from scratch. (The added development time of making a move, then realizing it's not working on the character and needing to make a new move, would spike up the time on Isaac so much).
I could see Dixie, Lyn, and others like them taking 300 - 500 hours, depending on how different the PMBR wants to make them feel from their bases.
Pichu, without the need for new animations or the level of QA testing that the others have, might be ~100. The PMBR would still want a new model for him, because that's their style, so that's a pretty significant time investment. Outside of that, they're playing with his properties.
So is he as popular as Isaac? Nope! But just based on what we know about how long Mewtwo and Roy took, Pichu would take a fraction of the time that Isaac or Ridley would take. This makes him a nice compliment to them, since you could definitely see a "4.0" update adding Ridley, who took a tremendous amount of work and effort, and then Pichu, who the team was able to get done during Ridley's development with just a little bit of work.
The only time it becomes an issue is if the PMBR fills up every character slot, but that's a pretty daunting task. Asking them to fill the slots with characters who all require 300-700+ hours of work is intense.
Sorry, I wrote too much again. : >
tl;dr Pichu has a smaller following than most chars, but he's also just about the easiest character to make and he's certainly bringing something unique to the table.
