Ranking built by repeated wins: take the ranker's winning tier, then re-run
the full ranker restricted to the remaining candidates to fill the next
place, and so on.
Differs from ranking() whenever a method's full ranking disagrees with
how it ranks subsets: round-based methods order losers by elimination time,
and score methods can reshuffle once a winner's votes are gone.
Ranking built by repeated wins: take the ranker's winning tier, then re-run the full ranker restricted to the remaining candidates to fill the next place, and so on.
Differs from
ranking()whenever a method's full ranking disagrees with how it ranks subsets: round-based methods order losers by elimination time, and score methods can reshuffle once a winner's votes are gone.