Introduction

Much important American electoral discourse revolves around spoiler candidates. In other words, minor contenders (typically representing third parties) who win just enough votes to be accused of preventing major-party candidates from securing majorities in tight races. 

The use of first-past-the-post voting in most U.S. states makes it easier for observers to blame “spoilers” because one party can win with a mere plurality. If the U.S. had universal ranked-choice voting (a form of instant-runoffs) or operated under a multiparty parliamentary or mixed-proportional system, finding scapegoats would be nearly impossible.

Over the last two decades, Libertarian candidates have been branded spoilers, almost always by Republicans losing high-profile races, more often than contenders from any other minor American political movement. 

This makes intuitive sense, since the Green Party has not received more votes than the Libertarian Party in a presidential race since 2000 ー when Ralph Nader won almost 3% of the vote and possibly contributed to Gore’s loss by winning 22,000 votes in New Hampshire; Bush carried the Granite state with 48% of the vote.

Other than that, the Green Party posted its best recent presidential showing with Jill Stein in 2016. Even then, she got just about 1% of the vote ー coming in behind Libertarian Gary Johnson at 3.3% (incidentally the best showing by a minor presidential candidate since Ross Perot in 1996).

In races for Congress over the last decade, Libertarians have also received more of the average vote than their Green Party counterparts. That’s part of the reason why we decided to focus this article solely on examining the impact Libertarian candidates have had on specific races. In other words, there isn’t enough empirical evidence to build a valid case study addressing independent and Green spoilers.

But that shouldn’t imply that it’s easy to gauge Libertarian-driven effects either.

At the end of the day, it’s impossible to determine with certainty which elections Libertarians spoiled for Republicans and which they did not. To get around this problem, we at Split Ticket have created a set of simple criteria that help us estimate which races Libertarian candidates may have affected the most after controlling for one major caveat: Libertarian voters are highly unpredictable.

Besides the fact that small sample sizes inherently increase variability, Libertarian voters are unpredictable because they tend to be more politically-inclined or well-informed than average (a characteristic of third party voters in general) and are thus more likely to have esoteric ideologies and unique grievances.

These tendencies make more sense when one looks at historical voting patterns. In a Cato Institute analysis of Libertarian vote depth, data from the 2000 presidential election indicated that 72% of voters expressing Libertarian ideological sentiments voted for Bush. The President’s support among Libertarians dropped in 2004 due to “expansive” government measures like the Patriot Act and intervention in Iraq, though he still won a majority of voters identifying as ideological Libertarian.

Those sudden shifts hint at one crucial conclusion: most ideological Libertarians may vote Republican over time, but the number doing so can change rapidly between cycles. Actual Libertarian Party voters ー only a fraction of those in the electorate identifying as ideologically Libertarian ー are even more fickle. 

While it is somewhat safe to assume that at least a plurality of a given Libertarian spoiler candidate’s voters would prefer the Republican to a Democrat in an instant-runoff ー assuming they would vote at all ー there’s no scenario in which it would be statistically sound to lump all Libertarian voters into the Republican basket like some observers decrying “spoilers” do.  

Conclusions

To account for that variability, we created a simple “spoiler test” to gauge the extent to which Libertarians cost Republican candidates in close, high-profile elections over the last two decades. The calculation takes the total Libertarian vote share, assumes 15% would not have voted in a two-party race, gives 20% to the Democrat and the remaining 65% to the Republican. 

Races from 2002 to 2022 for House, Senate, and governor were only considered “potentially-spoiled” if a Democrat won a plurality of the vote and the Libertarian’s share alone could have sufficed to elect the Republican. The lone exception to the rule was Georgia’s 2020 Senate race, in which Republican incumbent David Perdue won a plurality going into a runoff he ultimately lost. Contests that ostensibly fit this category but had added sources of variability like strong Green Party candidates were excluded.

Observations Summary (Total & % Flipped After Accounting For Libertarian)

Of the 29 races Split Ticket observed, 18 (64%) would have flipped after reallocating Libertarian votes based on the methodological ratios defined above. For House contests, which accounted for a larger share of the total observations, the percentage fell below the average at 57%. 

Keep in mind that our calculations are intentionally generous to the Democrats to separate races that were almost certainly spoiled from more uncertain ones. Either way, the fact that 62% of our sample contests did flip using our modest reallocation methods indicates that Libertarian spoiler candidates have cost Republicans prominent races on several occasions. Let’s take a look at some of the candidates who should have won without a Libertarian on the ballot.

Would-Be Winners Sorted By Cycle (2002-2022)

The most striking recent conclusion comes from Georgia’s 2020 Senate race, where our calculations suggest that incumbent Republican David Perdue would have defeated Democrat Jon Ossoff by about 3 points in the first round of voting had Libertarian Shane Hazel not been a factor. Given that Perdue, in real life, led by 2 going into a runoff which he lost to Ossoff, one could blame Hazel for costing Republicans their Senate majority.

2016 also saw Libertarians help defeat Republicans Kelly Ayotte (NH) and Pat McCrory (NC). Ayotte should have beaten Granite State Governor Maggie Hassan, a top Democratic recruit who won by just 1,017 votes ─ the narrowest Senate win of the 2016 cycle. Even Governor McCrory would likely have defeated North Carolina Attorney General Roy Cooper despite the fallout from the Republican legislature’s unpopular, discriminatory bathroom bill.

Had Libertarian votes been reallocated, Republican Dino Rossi almost certainly would have been elected Washington’s governor in 2004 ─ he lost to Democrat Christine Gregoire by just 129 votes out of nearly 3 million cast. In Minnesota, sitting Republican Norm Coleman could have been reelected in a Senate contest so contested that the Democrat Al Franken didn’t swear in until July 2009.

Perhaps the biggest twist in fate involves South Dakota’s 2002 Senate race, which Democratic incumbent Tim Johnson won by just 524 votes. Congressman John Thune likely would have unseated him in an R vs. D race, clearing Democratic Leader Tom Daschle’s path for reelection in 2004 and giving the senior statesman an extra six years in the Senate.

Of the states represented above, Libertarian Party spoilers were most impactful in Montana. Senator Jon Tester and former Governor Steve Bullock owe their initial victories to strong Libertarian candidates siphoning conservative votes away from weakened Republican nominees Conrad Burns and Rick Hill. As far as history is concerned, Libertarians could make or break Tester’s 2024 reelection ─ especially against a Republican like Matt Rosendale. 

A few spoiled House elections also stood out from the pack, including Democrat Yadira Caraveo’s victory in Colorado’s 8th district last fall. She took just 48% against Republican Barbara Kirkmeyer, a state senator who likely would have won had Libertarian Richard Ward not taken almost 4% of the vote. Kirkmeyer was a 2022 WAR overperformer.

Democrat Joe Donnelly’s narrow 2010 victory against Jackie Walorski in Indiana’s 2nd was another notable race. Had Walorski unseated him during the GOP’s midterm wave, Donnelly’s 2012 Senate chances might have been damaged enough to prevent his election.

Twelve Republican candidates in races that initially appeared to be spoiled still would have lost even after allocating the Libertarian vote. Republicans Herschel Walker (GA) and Rick Saccone (PA) are both recent additions to this category. Walker, a problematic Senate nominee to say the least, would not have benefitted as much from down-ballot lag in an R vs. D race as Perdue would have in 2020. In other words, Democrat Senator Raphael Warnock still would have secured an outright victory. won reelection. Saccone, another controversial candidate, would not have won the 2018 special election in Pennsylvania’s 18th district (a solid Trump district won by moderate Democrat Conor Lamb) as a result of our reallocations.

Perhaps most interesting of all, our calculations show that Libertarians did not reelect Democratic Senators Joe Manchin (WV) in 2018 or Jon Tester (MT) in 2012. Rep. Denny Rehberg would have lost by 0.8 points in an R vs. D contest and Attorney General Patrick Morrissey would have come up short by 1.4 points. One race that was conclusively spoiled by the Libertarian was the 2012 gubernatorial race in Montana, where Steve Bullock only won by 1.6 points. Republican Rick Hill would have won by 0.1 points, or just over 500 votes, without Libertarian competition.

At the end of the day, there is no way to definitively denote which races Republicans would have won without Libertarian spoiler candidates. It is also possible that we have underestimated the amount of Libertarian votes that would go to the GOP in R vs. D races. Either way, Split Ticket’s methodology suggests that Libertarian candidates did cost Republicans a majority of the high-profile races in our sample in a meaningful way. Later this year we will write a follow-up article examining the impact of other minor party candidates, including those from the Green and Independent American parties.

Correction: This article was republished on February 10, 2023 to account for the 2013 Virginia and 2019 Kentucky gubernatorial elections, which Republican candidates also would have won without Libertarian spoilers.

—

We hope you enjoyed this post. If you did, please consider subscribing to get our articles in your inbox — it’s free!