Talk:Bucklin voting
I'd like to conjecture that when the Bucklin winner is not in the Smith set, at least one or perhaps all of the Smith set candidates are ranked above or equal to the Bucklin winner on at least one of the ballots that contributed to the Bucklin winner's victory.
A much more general and harder-to-prove conjecture would be that this applies in some form to the Smith set of winner sets of certain Condorcet PR methods and the winner set of Expanding Approvals Rule. BetterVotingAdvocacy (talk) 18:38, 21 February 2020 (UTC)