Every election cycle we hear of the “spoiler,” the third-party candidate whose candidacy threatens to swing the election, not by winning, but by siphoning votes away from one of the two major party candidates. Some Democrats are still upset at Ralph Nader, the 2000 Green Party Presidential candidate. They believe he took votes from Al Gore in Florida, handing that historically-close election to George W. Bush.
This year, one of the supposed “spoilers” is Libertarian Party candidate Lucy Brenton in the Indiana Senate race. Brenton is running against Democrat Joe Donnelly and Republican Mike Braun. Many political analysts are saying that Brenton’s candidacy threatens to throw the election to Donnelly. Republicans are blaming her in advance for Braun’s possible loss.
When analyzing potential spoilers, most pundits do very simple math. They look at the polling difference between the two major-party candidates, and if a third-party candidate’s poll numbers are greater than that difference, then he’s a “spoiler.” For example, let’s say Candidate D has 44% support, and Candidate R has 41% support. That’s a difference of 3%. Now let’s say third-party Candidate L has 5% support. That’s greater than 3%, so she’s a spoiler!
But there are some serious problems with this simplistic analysis. Here are some ways the myth of an election spoiler doesn’t match reality.
The first false assumption when designating a “spoiler” is that we can even know with certainty how a voter would vote if the third-party candidate were absent. Typically, for example, it’s assumed that Green Party candidates pull from the Democratic Party, and a Libertarian candidate’s support comes from Republicans.
The Indiana Senate race shows how this assumption falls apart. As I mentioned, Republicans are already blaming Libertarian Lucy Benton for Republican Mike Braun’s potential loss. Yet the data shows something different. There has been only one poll conducted that asked a voter’s preference both with Brenton on the ballot and without her on the ballot. Surprisingly for many, Democrat Joe Donnelly is actually the one hurt by Brenton. In the three-way race, it’s Donnelly at 44%, Braun with 41%, and Brenton gaining 5.5% of the vote. In a two-way race, Donnelly comes in at 48% and Braun at 42%. In other words, with the Libertarian candidate removed, the Democrat gains 4% and the Republican only 1% (and the end result of the race doesn’t change).
Ultimately, it’s a futile exercise to try to predict how voters would vote in the absence of a third-party choice.
Beyond trying to guess how third-party voters would vote in the absence of a “spoiler,” there’s the problem of simple math. If a third-party candidate has 5% support, and a 3% difference divides the two major-party candidates, does that mean the third-party is a spoiler? No. In order to be a true spoiler, the candidate would need to be pulling almost all of her support from just one candidate. For example:
|Candidate D||Candidate R||Candidate L|
Now, let’s say that Candidate L withdrew from the race, and 60% of her support went to Candidate R and 40% to Candidate D. This means that Candidate R gets a 3% bump (60% of Candidate L’s 5%), but Candidate D still gets a 2% bump of his own:
|Candidate D||Candidate R|
|44% + 2% = 46%||41% + 3% = 44%|
In this scenario, only if more than 80% of Candidate L’s supporters would all vote for Candidate R could he have a chance of winning (trust me on the math). But third-party candidates never receive such unbalanced support.
For example, the reason Democrats are upset at Ralph Nader is that they assume his voters would have voted for Al Gore if Nader had not run. However, an analysis of Florida voters shows that in fact, at least 40% of Nader voters would have likely voted for Bush, not Gore. So even a candidate considered to be on the far Left still had support from members of both major parties.
Nader voters were not atypical; third-party candidates usually pull supporters from both major parties. This balanced representation is particularly true of Libertarian candidates, who pull from both traditionally Democratic and Republican voting bases. With what appears to many as an eclectic mix of policy positions, ranging from support for marijuana legalization to opposition to government social programs, the Libertarian Party gains support from both liberals and conservatives.
Choosing None of the Above
Another factor not being considered is that many third-party voters would simply not vote if a third-party candidate were not on the ballot. Again, this is particularly true for libertarian-leaning voters, who already are skeptical of the political process and thus tempted to sit a vote out if only the two major parties are represented. Oftentimes a third-party voter only votes because of the presence of the third-party candidate. In this case, neither major party candidate is impacted by the presence of the third-party candidate.
A Right to Your Vote?
There’s one more problem with the idea of an election “spoiler”: the assumption of a guaranteed vote.
Party partisans can’t imagine not voting a straight party ticket. The very idea of voting for someone else is practically treasonous. This is the assumption behind the very idea of a spoiler: that a voter has decided not to vote for the candidate they “should” vote for in favor of some upstart. Yet every election and every candidate stands on its own. Even if someone has voted straight Republican for 40 years, it cannot be assumed that she will vote for a Republican candidate in the next election. Likewise, even if someone considers himself a liberal, that doesn’t mean he will vote for a Democrat. Candidates must earn every vote they receive.
Political pundits obsess about the implications of every vote: What does it mean for control of Congress? How does it impact the sitting President? If Candidate X wins, does that mean she’ll run for President in the future? But for the average voter, all that matters is how a candidate best represents his interests. So if the Republican Party puts forward a weak candidate, then typically-Republican voters might look elsewhere. Likewise for the Democratic candidate. Talk of an election spoiler is really a way for a party to shift the blame for a loss away from its poor candidate.
It’s clear that most analysis of potential election spoilers is fundamentally flawed. The assumptions made about voters’ “what-if” choices are often mistaken and/or too simplistic. Pollsters and political pundits cannot fit every voter neatly into one of two categories, and the presence of a third-party candidate often allows voter diversity to emerge. There is no such thing as an election spoiler, only winning candidates and losing candidates.