All of the following methods have some major defects in choosing a majority consensus winner for a single office. Plurality- Obviously a plurality winner might not be the choice of a majority. Major insincere voting is probable for the first (i.e. only) choice. Example 1992 U.S. President election 43 C 39 B 12 P > B 6 P > C 100 C won but B might have been a consensus majority winner. Top 2 in Runoff- One of the losers might be a majority winner if head to head votes were used. Example 45 A > C 42 B > C 10 C 3 D > C 100 C/A 55/45, C/B 58/42, C/D 97/3 Approval- does not rank choices. Counts more choices than needed to get a majority. Can cause a first choice majority to lose. Example-- 26 A > B > D 25 A > C > B 25 C > B > D 24 C > D > B A 51, B 100, C 74, D 75 B wins using Approval, A has a majority of first choice votes. Instant run-off - repeatedly drops candidate with lowest number of votes A dropped loser might be a head to head (Condorcet) winner. Example- 47 D > F 42 B > F 11 F 100 (regular) Bucklin- adds earliest choices until there is a majority Major insincere voting is probable for second choice especially and later choices to a lesser degree. Example 35 G 25 H 20 J > H 15 K > H 5 M > G 100 H gets 60 adding the 1st and 2nd choices. The J, K and M voters may really want to vote for each other as second choices. Folks should add other defective methods.
