A lot of people may not remember that Amy in SCIII (PS2) was worst than Raph. Then 6 months later, in SCIII: AE Amy became much better by being a complete different character, but Raph got nothing.
Well, that's a bit of a false analogy: Amy got much more work than Raph between the two versions for a pretty obvious reason: she (as with all of the "bonus characters" from the console version of SCIII) had about 60% of the total moves that the average member of the regular roster (such as Raphael) had at that point. There was vastly more work needing to be done to make her into a true competitive style. In fact, while all of the main roster characters that were included in SCIII:CE got a small amount of touch-up work to their moveset (in an inverse of what had usually happened in games in franchise up to that point, where the game would be released to arcades and go under several rounds of refinement in that context before a more comprehensive and polished version was released for consoles) the lion's share of the work between SCIII:CE and SCIII:AE was spent in fixing the former's notoriously heavy number of balance issues and outright glitches that destroyed even a hint of potential for competitive play. Some of that balance work went (as always) into tweaks for existing moves, but a lot of it went into just fixing blatantly broken aspects of the universal mechanics for the game.
All things considered, AE did a pretty phenomenal job (albeit too late to save the reputation of SCIII in general) of making a balanced final product out of the initial offering, and there's nothing much surprising that Raph (or any previous SCIII:CE main roster character) got fewer changes than Amy (or any of the three bonus characters promoted to the main roster) got: because again, Amy, Li Long, and Hwang were not actual competitive characters before that point, but rather just kind of placeholder "bonus" proof-of-concept content that got left in the console version because the team for that release had a "everything and the kitchen sink, even if it's half-baked and doesn't allow us enough time to properly balance this sucker" approach to development of the original version of the game.
Which was, of course, a colossal mistake, considering SCII had arguably been close to the gold standard for competitive fighters just a few years earlier. If SCIII:CE had 1) omitted 14 of its bonus characters and made Amy, Hwang and Li Long fully fledged characters from the start, and 2) dropped a mode or two from its single player content in favour of kicking more man hours into balance and basic programing, it would probably be considered one of the golden entries of the franchise, rather than one of the black sheep--and the entire course of the franchise (which went downhill for some time following that entry) might have been very different indeed.
