PSA: Don’t Worry About Exit Polls

At some point Tuesday afternoon or evening, someone on TV will likely start talking about data from “early exit polls” and your favorite panel of pundits will freak out and make sweeping, quickly forgotten declarations about the state of the race. And it’ll be dumb.

Ariel Edwards-Levy, the Huffington Post’s polling expert, explained why in a great piece. I really recommend that you read the whole thing but basically the exit poll data is not designed with your Election Night experience in mind. The early exits are used (along with other data sources) by the decision desks at various TV news networks to call races – that is, to figure out where the wind is blowing and get an early sense of what might be going on. But taken at face value early exits can be misleading.

Exit polls are not, however, designed to help give the American public a sneak peak of who’s likely to win, especially before the election is actually over. Fundamentally, exit polls are just surveys, and are subject to many of the same sources of error as any other poll. And, like any poll, they need to be properly weighted to represent the population they’re supposed to be measuring. In this case, that means recalibrating the numbers to match the actual results of the election, which we won’t fully know until the end of the night. Before that happens, exit polls can present a misleading picture of the race.


As Edwards-Levy also points out, the exit pollsters need to know what the results are in order to weight everything correctly and give us a clear picture of what happened. And after that happens, there’s going to be a big nerdy debate over whether the exit polls are right or whether some other method more accurately captured the demographic contours of the vote.

In my view, the inevitable, predictable, short-lived panic over the early exits comes from a pretty normal human impulse. As Election Day approaches, people want to get some greater certainty about what will happen. But the most reliable indicators (polls) aren’t offering a wildly greater level of certainty than they did the week before. So people channel their anxiety into the first new piece of “information” they have. Sometimes it’s exit polls, and other times it’s the length of the lines they saw at their own polling place. Some of this energy gets channeled into artsy predictions based on early voting (which may someday become a useful predictor of final results, but for now doesn’t add much value to poll-based predictions anywhere except for Nevada and maybe Florida). Basically people are good at finding some piece of data that they can channel their nervous energy into

It may be annoying, but if you want to know what happens next you’re just going to have to wait for results to come in.

Related Content