Sukapon
Popularity: He's probably even forgotten in Japan. But he's a retro, so that doesn't matter that much.
Relevancy: Lead character in Nintendo's first fighting game. But he was Japan only.
Design: A bit odd. He has moves, of course. The animations might need some work, and he'd have some wonky hitboxes and so forth. He's doable, but he'd be more work than many characters getting the balance right and keeping him from being confusing to watch on the screen (I watched some clips from the NES game and it was a little confusing). Even Rayman, who is similar, would not be so bad since his body mostly stays in the same configuration and his body parts don't fly all over the place.
Roster and Competition: We might expect one to two retros on the basis of Melee(Ice Climbers + Mr. Game & Watch) and Brawl(Pit + R.O.B.). And Little Mac may or may not count as one of them (yes, he had that Wii game, but that was after a 15 year hiatus). Takamaru, Mech Rider and some other characters are on the list. I don't know if we should
expect two retros. After all, at some point Sakurai will decide all the important retro characters have been done. He might think there's only one, or none, worth putting in. Or at least none from the pre-SNES or pre-N64 eras. This is probably the last game I'd expect any NES newcomer.
Sakurai comments: He said not to expect Sukapon. Now, he might change his mind about it. But unlike the Villager, this comment wasn't based on Sukapon not being a fighter. So while Sakurai might change his mind, we don't have any evidence that he's changed his mind about Sukapon. We might rate other non-violent characters higher because of the Villager (i.e. almost as high as if they were fighters), but that certainly doesn't apply to Sukapon. So I think Sakurai's comment should still carry a fair amount of weight.
Sukapon chances: 0.9%
Most retros don't rate that high, in general. I think there are a fair number of choices but also that the remaining pre-SNES era characters are not the greatest bunch. And he's pretty obscure. And he'd not be the easiest to implement. And Sakurai's anti-Sukapon comments still matter. All adds up to Sukapon being a pretty surprising choice. He's still possible though.
Sukapon want: 40%
I really don't care that much. But I'd prefer a lot of other characters.
Little Mac re-rate
I don't really think much has changed from when I rated him
before. There's no new information specific to him, the only thing we've gotten is general.
His major plus points are the stage, his Western popularity, and his recent well-received and decently selling Wii game. He might also count as a retro, sort of. If it weren't for the Wii game, he'd certainly be the most likely retro. Also, it's not necessary that the boxing ring have the Punch-Out logo on it, and it would also be extremely easy to change for pre- and post-Little Mac reveal screen shots, since it's simply a design that goes on a flat surface. The stage otherwise looks pretty much just like the Punch-Out ring.
What has changed is that I'm slightly more optimistic about the number of characters. But I'm also slightly more pessimistic about Little Mac. In the end, it's about a wash.
Little Mac chances: 82%
Little Mac want: 59%
I'm coming around a little bit.
Andy prediction: 9%
Brawl was the best time to put a Wars character in. But he's still got a shot.
Nominations:
Sheik x5
I like seeing that Sheik is racing up the nominations list. Hopefully we can get all the veterans done quickly (even though, yes, Pichu and Young Link are going to get obliterated).
Statistics
Also, I'm going to have some statistics in a little bit. Maybe Groose will want to put them in the OP. They'll be our benchmarks of what we want to beat in accuracy.
I can tell you right now that the previous iteration of the game will do pretty horribly, because they massively overestimated the number of characters (in other words, they massively overrated everyone but the most obvious choices like Diddy). So I will try out some transformations to apply to the data like normalizing the probabilities to the actual roster size. That way I can apply the same transformations to our predictions to produce normalized predictions.
If we rate the likelihood of different roster sizes or at least get an average expected roster size, I can even use that information to produce a normalized score that takes that into account. That would make our individual character ratings consistent with our expectations with the roster size.