Building on Adam’s thread on Kanye West (not a sentence I expected to write this year), I wanted to look at the spoiler effect of a third party candidate in a first past the post election. We don’t have great data on that counterfactual, but I think that we can look at Maine’s 2nd Congressional District election in 2018 to get some idea of what percentage of not-first-two party vote is persuadable to vote for one of the two major parties.
Maine using Ranked Choice Voting for its federal elections. A voter sees all the candidates who are running and ranks them from 1 to the number of candidates. The voter has a choice to stop ranking at any point. If there is no clear majority winner in the first round, the ballots of the candidate who received the fewest 1st choice votes are re-allocated on the basis of the second choice. This repeats until there is a clear majority winner.
Maine’s 2nd District went to a RCV run-off in 2018. The first round had the Republican incumbent hold a tiny plurality lead. The Democratic challenger won after the run-off. That is not the interesting part. The interesting part is this paragraph from the Portland Press Herald on the 3rd and 4th choice candidates:
In the end, Golden gained 10,232 votes from the ranked-choice retabulations and Poliquin gained 4,695 votes. That 5,537-vote edge allowed Golden to overcome Poliquin’s 2,632-vote lead. Roughly 8,000 of the ballots cast for the independents did not designate an additional choice or did not select either of the major-party candidates.
Roughly 8,000 out of almost 23,000 minor party voters did not choose a major party candidate. These folks were mobilized specifically for the minor party candidate. However over three fifths of the minor party voters were okay with at least one of the major party candidates. We can assume that most of them would likely have voted for a major party candidate in a first past the post system.
So what does that mean?
Some fraction of a third party candidate’s vote is never in play for a major party candidate. However a decent to large chunk of it is in play. And an ideological or identity based third party candidate is going get most of the in-play voters from the major party closest to it. So, a spoiler candidate is likely to get some small percentage of idiosyncratic voters who were never in play and then a good size sliver of voters from the major party closest to it. Looking at Maine, the 2nd District results had a net swing of about 1.8% which in a close race was a decisive swing.