Experiment: how would the primaries have gone if we had had ranked choice voting?

[Recursion]

That one intrigues me. I think RCV is perfect for a primary situation, particularly because it allows people's displeasure with a candidate to be factored in.

In RCV, rather than voting for one person, you rank your preference of all (or some subset) of the candidates. Higher preferences are worth more.

So keeping it simple, if I had had a ranked choice ballot among Clinton, Sanders, and O'Malley, I would have ranked them O'Malley, Sanders, Clinton, which would have given three votes to O'Malley, two to Sanders, and one to Clinton (the system doesn't have to be that simplistic; just making it plain). If Webb and Chafee were still in the race, I wouldn't have ranked them because I don't want them to receive any votes at all (or then again the ballot could limit me to only ranking 3; there's a lot of ways this could work).

There's something called Condorcet's Paradox which points out that first-past-the-post voting (the kind we use) allows for some incredibly paradoxical results, in particular the fact that the least-liked candidate in a multi-candidate race can win, as well as rock-paper-scissors victory patterns (Bob beats Alice, Alice beats Charles, Charles beats Bob, even when no individual's preferences are cyclic like that). Every scientific body I know of uses RCV or some version of it because of that.

Obviously GD-P is going to be a bit skewed, but I'm curious what the results of an RCV system would be here.

6 votes, 0 passes | Time left: Unlimited
	Clinton, Sanders, O'Malley
	0 (0%)
	Clinton, O'Malley, Sanders
	0 (0%)
	Sanders, Clinton, O'Malley
	1 (17%)
	Sanders, O'Malley, Clinton
	3 (50%)
	O'Malley, Clinton, Sanders
	0 (0%)
	O'Malley, Sanders, Clinton
	2 (33%)
	0 DU members did not wish to select any of the options provided.
Show usernames Disclaimer: This is an Internet poll