Anyone watching Tuesday's midterms should be ready to settle in for a long election night -- a plethora of close races, slow ballot-counting in states like Alaska and the potential for runoffs means some races could remain unsettled for months.
While it may be tempting to rely on the early exit poll results that inevitably leak around dinnertime -- don't. The numbers won't tell you much at all.
The network exit polls, conducted every two years by Edison Research for the National Election Pool partnership of ABC, CBS, CNN, Fox, NBC and The Associated Press, are an incredibly useful resource for figuring out who voted and why they chose to support the candidates they did. As a means of making predictions before the polls close, though, they're not so great -- nor are they intended to be used that way.
A brief explanation of why, from 2012:
First, an exit poll is just a survey. Like other polls, it is subject to random sampling error, so differences of a few percentage points between the candidates in any given state sample are not terribly meaningful … Second, the networks almost never "call" truly competitive races on exit poll results alone. The decision desk analysts require very high statistical confidence … They usually only achieve that confidence for relatively close races after the exit pollsters obtain the actual vote results from the randomly selected precincts at which interviews were completed (and from other larger random samples of precincts) and combine all of the data into some very sophisticated statistical models.
Even then, if the models project that the leading candidates are separated by just a few percentage points, as pre-election polls suggest they will be in all of the key battleground states, the networks will usually wait until nearly all votes are counted to project a winner.
Third, the initial results of the exit poll interviews have had frequent problems with non-response bias … Those of us seeing leaked data, however, see neither the running calculations of the precinct errors nor the levels of statistical confidence associated with the vote numbers. We see only precise-looking percentages and are oblivious to the potential for error ... [Exit pollsters now] hold back the data from their news media clients in a sealed quarantine room on Election Day until 5 p.m. Eastern time. The quarantine means that any numbers purporting to be "exit polls" before 5 p.m. are almost certainly bogus.
Before 2012, early exit polls consistently showed a non-response bias favoring Democrats, but that problem may be a thing of the past. Michael Barone, who watched the exit polls closely as a Fox News election night analyst, noted that the early exit poll numbers were "pretty close to on target" in 2012 after a "more educated massage" before release. It's yet to be seen how that plays out in 2014.
Regardless, the growth in early voting has complicated the way exit polls are conducted. Nearly a third of the electorate voted early or absentee in 2012 and were thus unavailable to be interviewed at their polling place in traditional exit polls. This year, Colorado joins Oregon and Washington as states voting entirely by mail, and half a dozen states cast more than half their votes early or by absentee ballot in 2012. For the past dozen or so years, in an attempt to account for these individuals, exit pollsters have used telephone polls conducted over the final weekend to interview early voters.
If you want to know who voted and why, seek out the exit poll results later Tuesday night. But if you want to know who’s going to win, you'll just have to wait for the votes to be counted.