FBI Director James Comey sent a(nother) letter to Congress on Sunday afternoon — two days before the finish line of the 2016 campaign — to inform lawmakers that after reviewing emails related to Hillary Clinton that were found on former congressman Anthony Weiner's laptop, the agency stood by its July recommendation against charges.
Will the October-and-now-November Surprise have any significant effect on election results? It will be near-impossible to know the answer before Election Day, if then.
But the narrative will be unavoidable if Donald Trump is elected president — or, really, even if it's just close: The late FBI announcement that it was looking into more emails in the Clinton investigation tipped the scales late in the race, clouding her campaign during the final full week of early voting.
First of all, it's important to note that national polling was tightening even before Comey's Oct. 28 announcement. The Washington Post-ABC News tracking poll conducted Oct. 24 to 27 already showed Trump's deficit declining from double digits to just 2 points — exactly where it is today. Other polls conducted just before the announcement had the margin at two to five points — also down from double digits in some early to mid-October polls.
Second, it's simply very hard to deduce what voters' motivations are. There is anecdotal evidence in recent polls that some people say the announcement mattered to their vote, but it's tenuous, at best.
Monmouth University went at this question directly in a couple of state polls this past week. In Missouri and Pennsylvania, it asked people flat-out whether the Comey announcement changed their votes. In both cases, 4 percent said it had, while about 9 in 10 said it didn't.
So this doesn't seem to have moved a ton of votes. But it's also true that even a 4-point shift in a tight presidential race could prove the difference. If it really did cost Clinton that much support, it would actually be the huge game-changer that some believe it is.
That said, such a result almost definitely oversells the shift.
If you look at the crosstabs in Monmouth's Pennsylvania poll, just 1 percent of Democrats said this changed their vote, while 5 percent of Republicans said it did. But that would be virtually all of the Clinton-supporting Republicans.
If you look at the Post-ABC national tracking poll, 6 percent of Republicans favored Clinton before the Comey announcement, and 6 percent still do. And Quinnipiac polling of Pennsylvania shows Clinton's support among Republicans actually ticked up from 7 percent in mid-October to 9 percent in a poll released Wednesday.
Where this theoretically could have mattered more is among independents. Monmouth's polls show 7 percent of them in Missouri and Pennsylvania say the news changed their votes. Which would be big!
But again, there's not really evidence that this has occurred at the national level or even in Pennsylvania.
The Post-ABC tracking poll before the Comey announcement showed Clinton taking 34 percent of independent voters; by the end of the week, it showed her taking even more — 40 percent — of them. As of Sunday, it was 41 percent.
The picture is similar in Pennsylvania. Polls there have certainly tightened, yes. But they had also tightened before the Comey announcement. And Quinnipiac polling shows Clinton maintaining a very similar advantage among independents. While it was 42 percent to 38 percent in mid-October, a poll released Wednesday (which was conducted mostly after the Comey announcement) showed her up 43 percent to 36 percent.
Many of these numbers and shifts are in the single digits, it bears noting, which means they are within the margin of error or close to it. So none of them should be taken to the bank. But they don't suggest a race that fundamentally changed post-Comey.
Third is the question of what it means to actually change one's mind. Just because 4 percent say the FBI announcement changed their minds doesn't mean that's 4 percent going from Clinton to Trump. It could be people going from Clinton to sitting this one out or Clinton to undecided, which means she could still win them back. It could even be people going from Trump to Clinton, theoretically speaking, because they disagree with the FBI's actions. We just don't know.
People are notoriously bad at figuring out what matters to their votes and reporting it accurately to pollsters. For example, a CBS/New York Times poll late last month shows that 57 percent of Republicans say the Comey announcement makes them less likely to support Clinton. The vast majority of them, though, probably never would have voted for her in the first place.
So the fact that 4 percent of people — pretty evenly dispersed across the political spectrum — said this changed their votes doesn't mean a 4 percent windfall for Trump.
Fourth — and this is probably the most likely area where this matters — is enthusiasm and turnout. Even if the announcement didn't actually change many votes, perhaps it will lead some Clinton supporters to stay home.
Here's the one area where this does seem to have registered rather clearly in the polls in the immediate aftermath of Comey's announcement. Clinton voters did seem to be turned off in the immediate aftermath of the news, as Post pollsters Scott Clement and Emily Guskin wrote Tuesday. The number of Clinton supporters who were “very enthusiastic” about the election dropped from 51 percent to 43 percent.
But even that number completely recovered in late-week Post-ABC tracking polls, and enthusiasm for Clinton was back on par with Trump's — just as it was pre-Comey.
The whole thing was probably a net-positive for Trump when it comes to keeping this race close, and it maybe turned off some Clinton voters who are exasperated by her continuing email problems. And Clinton needed to turn those voters out in the closing days of the campaign.
But it will be very difficult to know whether the Comey announcement is really what changed the race — even if Trump wins.