“Given modern polling techniques and the sure knowledge that lots of states (including three of the four biggest) are completely out of play, lots of voters feel freed up to vote for candidates of small 3rd parties who wouldn’t under a single nationwide plurality vote. This hasn’t been the traditional effect, but it wasn’t traditionally possible to know with such certainty that a state was out of play.
“Any non-first-past-the-post system would strengthen third parties compared to any first-past-the-post system (and parliamentarism strengthens it compared to presidentialism). Runoffs, Borda counts, IRV, AV, whatever. But, if choosing between the current system and a nationwide plurality FPP, the current system probably marginally increases third-party voting.
“But that’s all about small third parties– Libertarian and Green-sized. When you have a third-party candidate who starts to hit 20-25%, the math is
going to change a lot. Then you will start to see the dynamic that elected Jesse Ventura (in a single statewide FPP plurality). Under the electoral college, a candidate with 25% support is still sure not to win. Under plurality, that’s a very credible level– given that 34% could be a winning plurality in a three-person race, and 26% in a four-person race. A third-party candidate with the support of Perot ‘92 or Wallace ‘68 does much better under FPP than under the electoral college.”
This makes a great deal of sense to me.