Third Parties: Always an Election Away

Courtesy: Slate

Courtesy: Slate

It’s been a common refrain virtually every election cycle: this is the year that a third party will make noise in the presidential election. The 2016 election has been no different. There was speculation about Michael Bloomberg again, the #NeverTrump movement tried to draft various third party challengers, and there was no shortage of speculation that Bernie Sanders would mount a successful third party bid. Taking it a step further, former Labor Secretary Robert Reich advocated for the creation of a New Progressives Party. Then, there are the actual third party candidates that have been getting buzz as players in the election; Jill Stein of the Green Party is thought to be able to take away disaffected Bernie-or-Bust Democrats, but the most breathless coverage has been reserved for Gary Johnson of the Libertarian Party.

Alas, those dreaming of a third party to rise up in the 2016 election are about to have their hopes dashed. Rather than being on the cusp of breaking through, history indicates that Gary Johnson is likely near the apex of his support.


Since World War II, there has been one virtual certainty amongst third party candidates: their support will peak in June or July and then they will steadily lose this support the closer it gets to the election. The only exception to this rule was George Wallace in 1968, who was the only independent candidate to hit his peak after early summer. However, even Wallace finished with a lower result than he was polling in June and July. Of the six most successful third party candidates since World War II, the average independent candidate only received 53 percent of their June polling average by Election Day. Even that is being extremely generous, as George Wallace’s 100% of his June average considerably skews the results. Ross Perot in 1992 is the only other candidate to get a final vote tally of at least 50% of his June average.

Source: Gallup Presidential Election Trial Heat Trends 1938-2008

It’s an easily understood phenomenon. As the candidates for the two major parties are decided, there is a portion of the electorate that is unsure about the two main choices, and during some elections there is still a bad taste left over among the supporters of the candidates who did not prevail in the primaries. At this point in the election, they say that they will support the third party candidate because the third party candidates are more of a blank canvas to project the voters’ desires. However, the closer it gets to Election Day, the more voters have learned about the third party candidates and the more apparent it becomes of the ramifications of supporting a third party candidate in terms of helping a candidate who a voter is completely opposed to verses a candidate they don’t really like but would prefer. The polling, as shown above, also becomes a self-fulfilling prophecy; once it is clear that a candidate cannot win, many potential supporters are not open to “throwing their vote away.”

2016 RACE

The current conventional wisdom seems to be that if Libertarian candidate Gary Johnson can get 10% now, who knows what he’ll get by Election Day. This makes for a better story, but there is simply no historical precedent for a candidate doing as poorly as Johnson is now dramatically improving as the election season wears on. If Johnson and Stein follow the typically pattern and lose half or more of their current support, Johnson would end up with 2.5% and Stein 1.5%. This would be an increase from their 0.99% and 0.36% respectively in 2008, but a far cry from the heights pundits have been predicting.

In fact, there are reasons to believe that the post-World War II trend is far more generous to potential third party candidates than the current environment. The four most successful third party candidates (George Wallace, John Anderson, and Ross Perot both times) benefitted from running in a period of reorganization between the two parties. With Southern Democrats and Rockefeller Republicans in flux between the parties, there were plenty of swing voters that could be convinced to vote for a third party in 1968. And Perot was able to self-fund his way to get attention but had very little to show for it in the end.

As voters have become increasingly hardened in their partisan preferences, there are fewer and fewer voters willing to break with their party for a third party candidate. Even with Hillary Clinton and Donald Trump’s historic unfavorable ratings, the fact of the matter is that most voters have a strong partisan preference, even if they consider themselves independent.


There is no question that this election has been interesting so far, and given the likely nominees at the top of the ticket for the two main parties we would expect this trend to continue. That being said, there is nothing in our data to show that Johnson or Stein has much of a chance when it comes to breaking the pattern of the typical independent candidate.

In our recent poll of 1,000 adults nationwide, we decided to test out the various ways to ask the horse race and see if there was any discernible difference.  Half of the respondents were first asked the horse race with all four candidates named. The results were:

We then asked Johnson and Stein supporters which of the two main candidates they would support if their preferred candidate was not on the ballot this November. Not surprisingly, there was a close divide between Clinton and Trump, with most supporters undecided.

If we add in the supporters who would choose a candidate when prompted with the Clinton and Trump supporters in the initial head-to-head matchup, we get:

The other half of respondents were first asked a two-way horserace between Clinton and Trump, and we saw very similar results to the four-way race.

We then asked this set of respondents the complete list, and as the chart below shows, there was very little difference.

At the end of the day, there is little difference in the results on Clinton and Trump regardless of how you ask the question. That being said, we would certainly argue that Johnson should be included in all polls since he is on the ballot in all 50 states, and Stein should be included in states where she has qualified. However, there is little to show that either candidate will be able to break the pattern that we have seen with independent candidates over the past few decades. Indeed, in our research, 26% of Johnson supporters say there is a better than 50% chance that they will end up supporting Clinton in November, and 13% say the same about Donald Trump. Ideologically, 27% of Johnson supporters consider themselves to be aligned with the Democratic Party or to its left, and 24% feel they are aligned or to the right of the Republican Party. With a majority of his supporters having a natural home with the two parties, it would not be surprising to see his support cut in half by November, matching the average movement for independent candidates.

Further hurting Johnson’s chances to maintain his support is that 46% of his supporters currently do not know enough about him to have an opinion on whether they have a favorable or unfavorable view of the Libertarian candidate. What is clear among these voters is their dislike for both Clinton (82% unfavorable) and Trump (86% unfavorable). Interestingly, there was an even split among Johnson voters on whether they voted in the Democratic (28%) or the Republican (27%) primaries. The vast majority of Johnson supporters who voted in the Democratic Primary supported Senator Bernie Sanders (83%), while among those who voted in the Republic primary, most were supporters of Senator Marco Rubio and Senator Ted Cruz.


Based on historical precedent and our current polling, it seems that Gary Johnson and Jill Stein are not about to achieve anything more than minimal electoral success in 2016. History suggests that they have already reached their peak polling as voters are increasingly voting in lockstep with one of the two major parties. Unfortunately for those looking for breaking free of the two party system, there simply isn’t enough room for a 20% showing by a third party candidate at this time. Given the fact that an independent candidate has not won a state since 1968, it seems unlikely that Johnson would win Utah or Maine and, perhaps most of all, we would not recommend betting on the election being decided by the


For the PDF version of this memo, please click here.