Bernie Sanders has gone from long-shot candidate to a real contender for the Democratic nomination for president.
Were Democrats to make the "democratic socialist" from Vermont their nominee, would he have a chance of winning a general election?
We posed that question to six of the country's top political scientists, and their answers were broadly consistent: Under some unlikely circumstances, Sanders could win a general election. But nominating him would make it significantly more difficult for Democrats to keep the White House.
"[Sanders's] political views are more toward the ideological pole than the average voter's," said John Sides, a professor in political science at George Washington University, in an email. "Absent a very favorable set of conditions, nominating a candidate like Sanders as opposed to a more moderate Democrat creates the risk of a penalty at the ballot box."
The famous social science experiment that shows why Sanders would be easy to beat
But that in turn raises another (perhaps more obvious) question: What turns people off from extreme ideas? Why are voters less likely to support candidates who propose more radical solutions?
That question may be answered by a series of famous social science studies conducted several decades ago by Princeton University professor Daniel Kahneman, according to Bruce Miroff, a political science professor at the University at Albany.
The researchers found that people were much more upset by the prospect of losing some amount of their money than they were made happy by the prospect of gaining the same amount. The upshot: People have a strong psychological fear of loss — even when they know it might result in a better long-term outcome.
"If you offer people the opportunity for gain against the fear of loss, the fear of loss is twice as psychologically powerful as the hope for gain," Miroff said.
This phenomenon is called "loss aversion," and it holds true for political psychology as well as behavioral economics, according to Miroff.
There are many of good examples of this at work in our political system: the revolt against "Hillarycare" in the 1990s, the panic over George Bush's plans to privatize Social Security in the early 2000s, and, more recently, the public souring on Obamacare. (Obama's promise that people who liked their plan could keep it was dubbed the lie of the year.)
This dynamic could hurt Sanders, who proposes policies that promise a big upside — but only through serious disruption that the other side will portray as fundamentally dangerous and risky, Miroff said.
"Once the opposition starts saying, 'That may help some people, but most of you are going to lose what you already got,' the polls start plummeting," Miroff said.
In a general election, for instance, Republicans could effectively (and accurately) portray Sanders's single-payer health care proposal as one that would lead many people to lose what they already know and like. The long-term gains of reducing national health spending and increasing overall insurance rates would be abstract gains for many voters, and thus hard to sell against the fear of loss.
"Anyone who stakes out positions that will affect huge numbers of people — in that, the advantage goes to the opposition, because they can stoke fear," Miroff said.
How voters decide who to vote for in elections
Fear of sudden, dramatic change could impede Sanders in a general election. But just as powerfully, Republicans could also successfully portray Sanders as out of step with the average American's political views, according to the academics interviewed for this story.
There isn't a lot of doubt that this would have a big impact in an election. Political scientists have had a pretty good idea since the 1950s of how voters tend to make their choices: by identifying which candidate fits closest to them on an ideological spectrum.
"They look and identify themselves on a liberal-conservative dimension, and they pick who is closer to them," said Andrew Reeves, a professor of political science at Washington University in St. Louis. "From that perspective, Sanders is positioned fairly far out there on the left."
There is some evidence that this year might be different, and that an unprecedented level of dissatisfaction with American government could lead the public to choose candidates who promise to break with the status quo. But even that force is very unlikely to override our most basic models for how voters act, Reeves said.
"Are [voters] going to abandon someone who is most close to them ideologically to go with someone who will shake things up?" Reeves said. "I don't think there's evidence to that effect yet."
As Vox's Matt Yglesias writes, President Barack Obama won in 2012 even though most voters found themselves more ideologically aligned with Mitt Romney. Sanders would have an even bigger ideological gap to close.
The history of "movement candidacies" in American presidential politics isn't encouraging for Bernie fans
The social science research on voter tendencies is supported by modern American political history, which most of the experts referenced in expressing doubt with Sanders's general election chances.
The two most frequently cited examples were the failed candidacies of George McGovern, the Democratic nominee who lost in a 24-point landslide to Richard Nixon in 1972, and Barry Goldwater, the Republican nominee who carried just six states in 1964.
None of the professors thought Sanders would, if nominated, lose by such huge margins. But they saw the historical comparison as telling of the steep odds facing anyone who breaks sharply from the political consensus, especially if the Republicans nominate a candidate — like Marco Rubio — perceived as within the mainstream.
"I think Sanders-Rubio is McGovern-Nixon," said Seth McKee, a political science professor at Texas Tech University. "I think it'd be a blowout: I'd discount [Sanders] maybe 10 percent."
Jedediah Purdy, a Duke University law professor who has written about American political identity, said that Sanders is trying to pull off something largely unprecedented in so quickly shifting his party's platform.
Purdy framed it like this: Some presidential candidates really can transform the electoral landscape and capture the White House. But to do so successfully, these candidates are normally building on the groundwork laid by similar, prior campaigns.
"Goldwater’s movement campaign and the lessons mainstream Republicans took from it afterward made [Ronald] Reagan’s campaign possible in 1980 by rearranging the whole political landscape," Purdy said in an email.
There's not a lot of reason to believe Sanders could bring about this degree of change with his profoundly different platform, Purdy said.
"Sanders is trying to achieve realignment much more quickly than that," Purdy said. "In terms of the suddenness and degree of his break with the mainstream, he looks like Goldwater in 1964 more than like Reagan."
How much, exactly, would Democrats be hurt by nominating Sanders?
Let's say you're a Democrat who prefers Sanders to Clinton, but you worry that nominating Sanders would throw the presidency to a Republican. Is there a way to quantify the risk you'd be taking in rolling the dice with the less electable candidate?
Seth Masket, a political science professor at the University of Denver, said his best "ballpark estimate" is that Sanders would cost the Democratic Party 2 to 3 percentage points in a general election compared with a more conventional nominee.
"It's not as big an effect as flipping a growing economy to one in recession," Masket said. "It's more like flipping a growing economy to a stalled one."
Miroff, a political science professor at the University at Albany, said he thinks Masket's estimate is likely too conservative.
"I'd say it'd have to be considerably higher than 2 to 3 points. I'm thinking the loss would be in the vicinity of 6 to 10 points," Miroff said.
Republicans would find it easy to tie Sanders to the "socialist" label, Miroff said, adding that only 25 percent of the public trusts the government to carry out policies effectively.
"(Sanders) really has made radical, socialistic statements in the past about the redistribution of wealth and the expropriation of the oil industry," Miroff said. "The full force of a Republican attack would find Sanders to be a convenient target."
Why those head-to-head general election polls are "absolutely worthless"
In defense of their candidate's electability, Sanders supporters have often turned to general election polls that show him doing well in head-to-head matchups with potential Republicans.
Sanders himself has recently embraced this argument, telling ABC News that he was the most electable candidate in part because of a poll showing him beating Donald Trump in a general election.
"Take a look at recent polls in which Bernie Sanders is matched with Republican candidates Trump on down [and] Hillary Clinton is matched with Republican candidates," he said.
But it's regarded as blindingly obvious among political scientists that these findings are essentially illusory, and that general election polls this far out are about as predictive now as a weather forecast for Election Day.
"The impressions people have of the eventual nominees months from now will be so different from today," said McKee, the Texas Tech professor. "That's a nice thing to point to, but what does a head-to-head poll mean in early February? ... It's worthless. It's absolutely worthless."