well, you could determine who's the best with matchups. technically, if you had all the matchups (like what ritt was originally intended to do), you could use some mathematics to determine various aspects of what the tier list should be. and i speculate it would be a lot more accurate than people just taking into account general character strengths and weaknesses. that would take a lot of time though, and i usually find it better to simply look at and analyze various important matchups and see which ones would be useful/detrimental and to whom.
another thing to look at is the character's usage in tournaments. for instance, take kirby, who can reasonably take sheik and ganon. who's going to be more useful, in a counter-character standpoint? kirby, though his matches against everybody else that matters are pretty bad, can serve as a secondary character to work in conjunction with, oh, marth, who has a fair amount of trouble with sheik and ganon. this type of back-scratching usage allows lower-tiers to be used at much more of an extent than we see (since most people just use one char, and the ones that double up usually do so with fox and sheik, or some other high tier). the question brings in influences in two directions - if people do starting going to tourneys with azen-like lineups, matchups against lower tiers become increasingly important. but to become useful, lower-tiered characters have to break even or win against some rather hard-to-beat upper tier (captain falcon has a lot of these matches, it seems >_>). i don't know how this would work out ;)
DDR: want to pass Paranoia Survivor ~ Religions are just cults with more members. -- The resident SSB:M Captain Falcon user. -- "Let's eat fast so we can eat again!" ~ Ash