Everything from the weather to sports scores can nudge voters one way or another.
Next Tuesday's presidential election is likely to be determined by a razor-thin margin, political analysts remind us with increasing fervor and frequency.
While the campaigns strategize every last vote, some surprising factors outside their control could play a role in determining our next president.
There's Sandy, of course, but studies have also shown that everything from cloud cover on election day to the pitch of the candidates' voices to the score of the Nov. 3 football game between the Florida Gators and Missouri Tigers could shift results.
Neil Malhotra, an Associate Professor of Political Economy at the Stanford Graduate School of Business, has studied how mood affects voting. Bad moods nudge us toward change, while good moods make us want to preserve the status quo, he said.
"If people are in a bad mood, they're more likely to want to throw the president out," Malhotra said.
In a 2010 study he coauthored, researchers found that a victory by the local sports team could raise an incumbent's vote by 1.5 percentage points -- and up to 3 percentage points in areas near the most popular athletic programs. The researchers studied data from 62 college football teams from 1964-2008, and conducted a survey during the 2009 NCAA men's college basketball tournament.
"It's usually not enough to shift an election. But if an election is close enough, anything can change it," Malhotra pointed out.
Although their data didn't single people out, Malhotra suspects that the shift occurs among undecided-til-the-last-minute voters.
"Lots of people have more important stuff in life than an election, and they only start tuning in until the very end," he said.
And while the study examined sports to test the theory, it doesn't matter what causes the good or bad mood.
It could just as easily be cloud cover -- or cleaning up after a major storm, Malhotra said. (For the record, the Ohio State Buckeyes defeated Penn State on Saturday, while the Florida Gators lost to Georgia. Both swing-state teams are expected to pick up victories this Saturday.)
Then there's the theory that our inner cave man subconsciously shapes our voting behavior. A recent study helps explain why we tend to favor tall politicians, and those with deeper voices -- someone who could win a fistfight, in other words.
"In evolutionary terms, that question [who would win a fistfight] makes perfect sense," said Gregg Murray, assistant professor of political science at Texas Tech and author of Psychology Today's Caveman Politics blog.
Our predecessors would have wanted physically imposing people on their side, people who would help us defend our food and territory. Subconsciously, we may still hold those values.
The height difference between Barack Obama and Mitt Romney is negligible -- Romney is 6'2"; the president is 6'1" -- probably not enough for voters to perceive a difference, Murray said. But height also has a more subtle effect, Murray said: Tall people, particularly males, tend to think of themselves as qualified and interested in being a leader.
"In a democracy, that's really important in terms of who throws their hats in the ring," said Murray, who recently blogged about his daughter's chances of becoming president given our preference for physically formidable leaders. (Hint: Earning her black belt wouldn't hurt.)
Much of voters' preferences is driven by partisanship, of course. When asked to draw the vice presidential candidates in 2008, which pitted a man against a woman, Democrats tended to draw Joe Biden being taller, and Republicans drew Sarah Palin taller.
Casting these general preferences in the light of evolution helps explain some long-term persistent biases, such as why we prefer male leaders in times of war, even to the extent of preferring a leader with more masculine facial features, Murray said.
"It also helps explain why it's so frustrating to turn [our biases] around," he said.
There's a long list of additional, seemingly random reasons why voters cast their ballots. People are more likely to vote for candidates whose names appear higher on the ballot; more likely to vote for education efforts if they vote in schools; and more likely to vote for bans on same-sex marriage if they vote in a church, Malhotra said.
With those contributing factors in mind, you may start to wonder why billions of dollars are spent on campaigns. And with Sandy in the mix, predicting how -- and if -- people will vote just became even more of a whirlwind.