Approval voting is Condorcet-compatible voting under a constraint: A critique of Approval, and both Range and Star, voting
Summary: Two systems of electing one candidate from many are compared and proved identical: approval voting, and voting with a Condorcet-compatible method and ranked-choice ballots, but ballots on which voters must assign each candidate one of but two distinct ranks.
Approval, range, and star voting compare unfavorably to Condorcet-compatible methods for electing a single candidate. Examples show approval voting has a marked tendency in common elections to elect fringe instead of representative candidates. Range and star voting undermine the principle of one-man, one vote, and disproportionally empower voters who are either passionate, or willing to misrepresent how passionate they are to get an election outcome that they want. Arguments commonly advanced in support of approval, range, and star voting are examined and shown to be without substance. In particular, the ideas of maximizing social utility, as they have been used to argue in favor of an election system weighing the intensity of a voter’s preference for one candidate over another, instead of noting the mere fact a voter has a preference, are shown to be inapplicable. While range and star voting are easy to manipulate, an attempt under a Condorcet-compatible method to elect by tactical voting a candidate who is not the Condorcet winner is shown to both be extremely unlikely to succeed, and to punish those who engage in the attempt with an election outcome lower on their own list of preferences than what the outcome would have been had they voted honestly. For the particular method known as Ranked Pairs, that salutary deterrence is shown to exist in elections with any number of candidates.
