When FBI Director James Comey told Congress on Oct. 28 that he was reviewing additional emails pertinent to the case of Hillary Clinton’s email server, Clinton had an 81 percent chance of winning the election according to our polls-only forecast. Today, her chances are 65 percent according to the same forecast. The change corresponds with Clinton’s drop in the national popular-vote lead: from a 5.7-percentage-point lead in our estimate on Oct. 28 to a 2.9-point lead now — so a swing of about 3 points against her.
How much of that can be attributed to Comey? And now that Comey told Congress on Sunday that the emails on former Rep. Anthony Weiner’s computer won’t change his earlier conclusions about Clinton, should we expect her numbers to rebound? The cause and effect is hard to sort out. Clinton’s poll numbers were arguably a bit inflated in mid-October amid a very rough period for Donald Trump. And even before Comey, the media seemed eager for one last twist in the news cycle, so Clinton may have been due for a period of greater scrutiny one way or the other — for example, over emails from the Clinton campaign released by WikiLeaks.
Trump should get some credit, as well, for having been comparatively disciplined on the campaign trail. He’s gained about 2 points in national polls since Oct. 28, while Clinton lost 1 point.
Still, if you look at our win-probability graphic, while Clinton’s chances were slightly declining already after she came off her post-debate peak, the rate of decline began to accelerate a couple of days after Comey, once we began to receive some post-Comey polls. Now the decline has leveled off, and her lead has held steady over the past several days. One advantage of having a model like ours that’s pretty quick to detect changes in the polls is that we can potentially make better inferences about the cause of polling shifts. And while it isn’t proof of anything, the pattern is at least consistent with a “shock” caused by a burst of negative news for a candidate, as opposed to a more gradual decline.
In fact, the shift looks pretty similar to a period in July after Comey reprimanded but did not charge Clinton for her email server and testified before Congress about it. That period produced about a 2-point swing against Clinton.
The news may also have had an effect down-ballot. Democrats’ chances of winning the Senate were generally hovering around 70 percent in late October. Today, they’re 50 percent. It doesn’t take a lot to swing the numbers in the Senate forecast because of the large number of competitive races — even a 1-point swing toward Republicans because of higher turnout could affect the odds significantly.
So will the latest Comey letter help Clinton? That’s also hard to say, and any change will really come too late to be picked up on by most polls. It’s also plausible that the headlines themselves aren’t particularly helpful to Clinton, even if the news itself is. The Washington Post’s current web headline, for instance, is “FBI Director Comey says agency won’t recommend charges over Clinton email,” which reminds readers that Clinton was being investigated by the FBI for her email practices. Still, betting markets show Clinton’s probability of winning the election improving by about 3 percentage points on the news.