FiveThirtyEight

This is the tenth article in a series that reviews news coverage of the 2016 general election, explores how Donald Trump won and why his chances were underrated by most of the American media.

Hillary Clinton would probably be president if FBI Director James Comey had not sent a letter to Congress on Oct. 28. The letter, which said the FBI had “learned of the existence of emails that appear to be pertinent to the investigation” into the private email server that Clinton used as secretary of state, upended the news cycle and soon halved Clinton’s lead in the polls, imperiling her position in the Electoral College.

The letter isn’t the only reason that Clinton lost. It does not excuse every decision the Clinton campaign made. Other factors may have played a larger role in her defeat, and it’s up to Democrats to examine those as they choose their strategy for 2018 and 2020.

But the effect of those factors — say, Clinton’s decision to give paid speeches to investment banks, or her messaging on pocket-book issues, or the role that her gender played in the campaign — is hard to measure. The impact of Comey’s letter is comparatively easy to quantify, by contrast. At a maximum, it might have shifted the race by 3 or 4 percentage points toward Donald Trump, swinging Michigan, Pennsylvania, Wisconsin and Florida to him, perhaps along with North Carolina and Arizona. At a minimum, its impact might have been only a percentage point or so. Still, because Clinton lost Michigan, Pennsylvania and Wisconsin by less than 1 point, the letter was probably enough to change the outcome of the Electoral College.

And yet, from almost the moment that Trump won the White House, many mainstream journalists have been in denial about the impact of Comey’s letter. The article that led The New York Times’s website the morning after the election did not mention Comey or “FBI” even once — a bizarre development considering the dramatic headlines that the Times had given to the letter while the campaign was underway. Books on the campaign have treated Comey’s letter as an incidental factor, meanwhile. And even though Clinton herself has repeatedly brought up the letter — including in comments she made at an event in New York on Tuesday — many pundits have preferred to change the conversation when the letter comes up, waving it away instead of debating the merits of the case.

The motivation for this seems fairly clear: If Comey’s letter altered the outcome of the election, the media may have some responsibility for the result. The story dominated news coverage for the better part of a week, drowning out other headlines, whether they were negative for Clinton (such as the news about impending Obamacare premium hikes) or problematic for Trump (such as his alleged ties to Russia). And yet, the story didn’t have a punchline: Two days before the election, Comey disclosed that the emails hadn’t turned up anything new.

One can believe that the Comey letter cost Clinton the election without thinking that the media cost her the election — it was an urgent story that any newsroom had to cover. But if the Comey letter had a decisive effect and the story was mishandled by the press — given a disproportionate amount of attention relative to its substantive importance, often with coverage that jumped to conclusions before the facts of the case were clear — the media needs to grapple with how it approached the story. More sober coverage of the story might have yielded a milder voter reaction.

My focus in this series of articles has been on the media’s horse-race coverage rather than its editorial decisions overall, but when it comes to the Comey letter, these things are intertwined. Not only was the letter probably enough to swing the outcome of the horse race, but the reverse is also true: Perceptions of the horse race probably affected the way the story unfolded. Publications may have given hyperbolic coverage to the Comey letter in part because they misanalyzed the Electoral College and wrongly concluded that Clinton was a sure thing. And Comey himself may have released his letter in part because of his overconfidence in Clinton’s chances. It’s a mess — so let’s see what we can do to untangle it.

Clinton was in a danger zone before Comey’s letter

Clinton woke up on the morning of Oct. 28 as the likely — by no means certain — next president. Trump had come off a period of five weeks in which he’d had three erratic debates and numerous women accuse him of sexual assault after the “Access Hollywood” tape became public. Clinton led by approximately 6 percentage points in national polls and by 6 to 7 points in polls of Michigan, Wisconsin and Pennsylvania. Her leads in Florida and North Carolina were narrow, and she was only tied with Trump in Ohio and Iowa. But it was a pretty good overall position.

Her standing was not quite as safe as it might have appeared from a surface analysis, however. For one thing, there were still lots of undecided voters, especially in the Midwest. Although Trump had a paltry 37 percent to 38 percent of the vote in polls of Michigan, for instance, Clinton had only 43 percent to 44 percent. That left the door open for Trump to leapfrog her if late developments caused undecideds to break toward him. Furthermore, in the event that the race tightened, Clinton’s vote was inefficiently distributed in the Electoral College, concentrated in coastal states rather than swing states. While she had only an 11 percent chance of losing the popular vote according to FiveThirtyEight’s forecast that morning, her chances of losing the Electoral College were a fair bit higher: 18 percent.

Another danger to Clinton was complacency. Several days earlier, the Times had written that she was on the verge of having an “unbreakable lead.” And there was a risk that people looking at statistical forecasts were misreading them and “rounding up” a probable Clinton win to a sure thing. (We’ll take up that topic up at more length in a future article in this series.) But Clinton had actually slipped by a percentage point or so in polls since the final debate on Oct. 19. And the news cycle had become somewhat listless; the most prevalent story that morning was about the trial in the Oregon wildlife refuge standoff. Clinton was in a danger zone: Her lead wasn’t quite large enough to be truly safe, but it was large enough to make people mistakenly think it was.

The Comey letter almost immediately sank Clinton’s polls

News of the Comey letter broke just before 1 p.m. Eastern time on Oct. 28, when Utah. Rep Jason Chaffetz tweeted about it, noting the existence of the letter and stating (incorrectly, it turned out ) that the case into Clinton’s private email server had been “reopened.” The story exploded onto the scene; Fox News was treating Chaffetz’s tweet as “breaking news” within 15 minutes, and the FBI story dominated headlines everywhere within roughly an hour. In an element of tabloid flair, it was soon reported that the emails in question were found on a computer owned by Anthony Weiner, the former congressman, as part of an investigation into whether he’d sent sexually explicit messages to teenage girls.

Few news organizations gave the story more velocity than The New York Times. On the morning of Oct. 29, Comey stories stretched across the print edition’s front page, accompanied by a photo showing Clinton and her aide Huma Abedin, Weiner’s estranged wife. Although some of these articles contained detailed reporting, the headlines focused on speculation about the implications for the horse race — “NEW EMAILS JOLT CLINTON CAMPAIGN IN RACE’S LAST DAYS.”

That Comey’s decision to issue the letter had been so unorthodox and that the contents of the letter were so ambiguous helped fuel the story. The Times’s print lead on Oct. 30 was about Clinton’s pushback against Comey, and a story it published two days later explained that Comey had broken with precedent in releasing the letter. It covered all sides of the controversy. But the controversy was an unwelcome one for Clinton, since it involved voters seeing words like “Clinton,” “email,” “FBI” and “investigation” together in headlines. Within a day of the Comey letter, Google searches for “Clinton FBI” had increased 50-fold and searches for “Clinton email” almost tenfold.

Clinton’s standing in the polls fell sharply. She’d led Trump by 5.9 percentage points in FiveThirtyEight’s popular vote projection at 12:01 a.m. on Oct. 28. A week later — after polls had time to fully reflect the letter — her lead had declined to 2.9 percentage points. That is to say, there was a shift of about 3 percentage points against Clinton. And it was an especially pernicious shift for Clinton because (at least according to the FiveThirtyEight model) Clinton was underperforming in swing states as compared to the country overall. In the average swing state, Clinton’s lead declined from 4.5 percentage points at the start of Oct. 28 to just 1.7 percentage points on Nov. 4. If the polls were off even slightly, Trump could be headed to the White House.

Is it possible this was all just a coincidence — that Clinton’s numbers went into decline for reasons other than Comey’s letter? I think there’s a decent case (which we’ll take up in a moment) that some of the decline in Clinton’s numbers reflected reversion to the mean and was bound to happen anyway.

But it’s not credible to claim that the Comey letter had no effect at all. It was the dominant story of the last 10 days of the campaign. According to the news aggregation site Memeorandum, which algorithmically tracks which stories are gaining the most traction in the mainstream media, the Comey letter was the lead story on six out of seven mornings from Oct. 29 to Nov. 4, pausing only for a half-day stretch when Mother Jones and Slate published stories alleging ties between the Trump campaign and Russia.

It’s rare to see stories linger in headlines for more than two to three days given how quickly the news cycle moves during election campaigns. When one does, some effect on the polls is often expected. And that’s what we saw. The sharpness of the decline — with Clinton losing 3 points in a week — is consistent with a news-driven shift, rather than gradual reversion to the mean.

We also have a lot of other evidence of shifting preferences among voters in the waning days of the campaign. Exit polls showed that undecided and late-deciding voters broke toward Trump, especially in the Midwest. A panel survey conducted by FiveThirtyEight contributor Dan Hopkins and other researchers also found shifts between mid-October and the end of the campaign — an effect that would amount to a swing of about 4 percentage points against Clinton. And we know that previous email-related stories had caused trouble for Clinton in the polls. In July, when Comey said he wouldn’t recommend charges against Clinton but rebuked her handling of classified information, she lost about 2 percentage points in the polls. Periods of intense coverage of her email server had also been associated with polling declines during the Democratic primary.

So while one can debate the magnitude of the effect, there’s a reasonably clear consensus of the evidence that the Comey letter mattered — probably by enough to swing the election. This ought not be one of the more controversial facts about the 2016 campaign; the data is pretty straightforward. Why the media covered the story as it did and how to weigh the Comey letter against the other causes for Clinton’s defeat are the more complicated parts of the story.

The Times thought it was covering President-elect Clinton’s first scandal

Re-read one of those New York Times front-page stories from Oct. 29 — “This Changes Everything’: Donald Trump Exults as Hillary Clinton’s Team Scrambles” — and you’ll be surprised by how strange it is. It begins by describing the Comey letter in dramatic terms, as “the kind of potential turnabout rarely if ever seen at this late stage of a presidential race”:

Everything was looking up for Hillary Clinton. She was riding high in the polls, even seeing an improvement on trustworthiness. She was sitting on $153 million in cash. At 12:37 p.m. Friday, her aides announced that she planned to campaign in Arizona, a state that a Democratic presidential candidate has carried only once since 1948.

Twenty minutes later, October delivered its latest big surprise.

The F.B.I. director’s disclosure to Congress that agents would be reviewing a new trove of emails that appeared pertinent to its investigation into Mrs. Clinton’s private email server — an investigation that had been declared closed — set off a frantic and alarmed scramble inside Mrs. Clinton’s campaign and among her Democratic allies, while Republicans raced to seize the advantage.

In the kind of potential turnabout rarely if ever seen at this late stage of a presidential race, Donald J. Trump exulted in his good fortune.

And yet the same Times article told readers that this rarely-if-ever-seen turnabout wouldn’t cost Clinton the election. She had banked too much of a lead in early voting, the story said, and it came too late in the campaign. Instead, the Comey letter could “cast a cloud over a victorious Mrs. Clinton’s administration-in-waiting”:

With early voting well underway, and Mrs. Clinton already benefiting from Mr. Trump’s weekslong slide in the polls, Democrats’ concerns were tempered — more in the realm of apprehensiveness than panic.

[…]

Mrs. Clinton has an enormous cash advantage — $153 million in the bank for her campaign and joint fund-raising accounts as of last week, compared with $68 million for Mr. Trump’s campaign and joint accounts — which means Mr. Trump has limited means to use the F.B.I. inquiry to damage Mrs. Clinton with television ads.

With more than six million Americans having already voted as of Monday, any efforts by Mr. Trump to claw his way back into contention could come too late. The Clinton campaign says its early voting turnout data points to a Democratic advantage in several swing states, including Florida, Colorado, Arizona and Iowa.

But the specter of an F.B.I. inquiry could cast a cloud over a victorious Mrs. Clinton’s administration-in-waiting. News had hardly spread when exasperated Democrats and donors were ruefully dredging up painful memories of the seemingly constant tug of congressional investigations on Bill Clinton’s White House.

What the heck is going on here? Why was the Times giving Comey’s letter such blockbuster coverage and at the same time going out of its way to insist that it wouldn’t affect the outcome?

The evidence is consistent with the theory that the Times covered the Comey letter as it did because it saw Clinton as the almost-certain next president — and Trump as a historical footnote. By treating the letter as a huge deal, it could get a head start on covering the next administration and its imbroglios. It could also “prove” to its critics that it could provide tough coverage of Democrats, thereby countering accusations of liberal bias (a longstanding hang-up at the Times). So what if it wasn’t clear from the letter whether Clinton had done anything wrong? The Times could use the same weasel-worded language that it often does in such situations, speaking of the Comey letter as having “cast a cloud” over Clinton.

In a sense, the Times may have made a version of the same mistake that Comey reportedly did, according to the very detailed recounting of the FBI director’s decision that the Times published last month. The newspaper’s editors and reporters thought Clinton had the election in the bag. And they didn’t consider how their own actions might influence the outcome and invalidate their assessment. That influence was substantial in Comey’s case and marginal for the Times, as one of many media outlets covering the story. But the media’s choices as a whole potentially mattered, and the tone of campaign coverage shifted substantially just as voters were going to the polls.

“Little Comey” vs. “Big Comey”

One can make a case that the race would have tightened even if Comey had not issued his letter. Clinton had already lost a percentage point or so off her lead in the week before the Comey letter; if she continued at that rate of decline, she’d be down to a 4- to 5-point lead by Election Day. And although polls don’t always tighten down the stretch run — Barack Obama’s lead expanded at the end of the 2008 and 2012 campaigns — they sometimes move more in line with economic conditions and other “fundamental” factors. As of Oct. 28, the polls-plus version of FiveThirtyEight’s forecast, which accounts for these factors, expected Clinton to lose a point or so off her lead before Election Day.

Another complicating factor is that Clinton had a slight rebound in the polls over the final 36 hours of the campaign, with her lead improving from 2.9 percentage points on Nov. 6 to 3.6 points in our final forecast on the morning of Nov. 8 (Election Day). It’s not entirely clear what this uptick represented — it may have reflected pollster herding as outlier polls magically changed their tune. But it also could have meant that the Comey effect was fading as the news cycle moved on to other stories.

So you could postulate that the Comey letter had only about a 1-point impact. Perhaps Clinton’s lead would have been whittled down to around 4.5 points anyway by Election Day because of mean-reversion. And she led in the final polls by about 3.5 points. Yes, she also underperformed her final polls on Election Day, but that could reflect pollster error or undecideds breaking against her for other reasons, this case would say — there was no particular reason to attribute it to Comey.

Nonetheless, Clinton lost Michigan, Pennsylvania and Wisconsin by less than 1 percentage point, and those states were enough to cost her the election. She lost Florida by just slightly more than 1 point. If the Comey letter had a net impact of only a point or so, we’d have been in recount territory in several of these states — but Clinton would probably have come out ahead. I call this the “Little Comey” case — sure, the Comey letter mattered, but only because the election was so close.

*Adjusting for a small Comey effect adds 1 percentage point to Clinton’s vote margin. A big effect adds 4. Hypothetical scenario starts with the 232 electoral votes Clinton actually won, ignoring faithless electors.

Or one could argue for a larger impact from the Comey letter. This is the “Big Comey” case. There was nothing inevitable about the race tightening, it would say, given that the news cycle had been unpredictable and that Trump had a tendency to dig deeper holes for himself while down in the polls.

Representatives of the Clinton campaign made two additional “Big Comey” claims in comments at the Harvard Institute of Politics conference after the election.

First, they said the letter’s impact was larger in Midwestern swing states such as Wisconsin because there were large numbers of undecided voters there, especially among white voters without college degrees. And the Clinton campaign claimed that the second Comey letter — which he issued late in the afternoon on Nov. 6 and which announced that the emails on Weiner’s laptop hadn’t turned up anything new — hurt Clinton because it put “FBI,” “Clinton” and “email” back in the headlines. This is hard to test because the second Comey letter came so late in the campaign that there wasn’t time for polls to pick up its effects.

But it’s plausible that Clinton’s underperformance versus the polls on Election Day had something to do with Comey — either lingering effects from his original letter or new effects from his second letter. The “Big Comey” case might attribute a 4-point impact to him nationally — accounting for the swing between Clinton’s 6-point lead on the morning of Oct. 28 and her 2-point popular vote margin on Election Day — and slightly more than that in the swing states.

My personal views are more toward the “Little Comey” side of the spectrum, since I think there would have been a fair amount of mean-reversion even without Comey. That’s because Clinton and Trump had alternated better and worse months in the polls in a way that tracked with the news cycle. Clinton had been in a strong position in the polls in June, August and — until the Comey letter — in October, while Trump had drawn close to her in May, July and September (and therefore might have been “due” for an uptick in November). This pattern may have reflected some sort of complicated feedback loop in media coverage. After some initial stimulus — say, a strong debate — there was a frenzy of favorable coverage for a candidate and negative coverage for her opponent, with news events framed against a backdrop of rising or falling polls. Then after a few weeks, the reporting on the story exhausted itself, the polls stabilized and the press was eager to look for a reversal of momentum. Comey’s letter came at a time when the campaign press may have been itching for a change in the narrative after several tough weeks for Trump. If not for the Comey letter, perhaps some other story would have blown up in Clinton’s face. Still, this theory is speculative, and those other stories might not have had the kryptonite-like effect that email-related stories had on Clinton’s numbers.

Let’s play the blame game

The Comey letter wasn’t necessarily the most important factor in Clinton’s defeat, although it’s probably the one we can be most certain about. To explain the distinction, consider Clinton’s decision to run a highly negative campaign that focused on branding Trump as an unacceptable choice. One can imagine this being a huge, election-losing mistake: Trump’s negatives didn’t need any reinforcing, whereas Clinton should have used her resources to improve her own image. But one could also argue that Clinton’s strategy worked, up to a point: Trump was exceptionally unpopular and needed a lot of things to break his way to win the election despite that. The range of possible impacts from this strategic choice is wide; perhaps it cost Clinton several percentage points, or perhaps it helped her instead. The range from the Comey letter is narrower, by contrast, and easier to measure. It was a discrete event that came late in the campaign and had a direct effect on the polls.

The standard way to dismiss the letter’s impact is to say that Clinton should never have let the race get that close to begin with. But the race wasn’t that close before the Comey letter; Clinton had led by about 6 percentage points and was poised to win with a map like this one, including states such as North Carolina and Arizona (but not Ohio or Iowa). My guess is that the same pundits who pilloried Clinton’s campaign after the Comey letter would have considered it an impressive showing and spoken highly of her tactics.

Thus, you have to assess the letter’s impact to do an honest accounting of the Clinton campaign. If you’re in the “Big Comey” camp and think Clinton would have won by 5 or 6 percentage points without the letter, it’s hard to fault Clinton all that much. Even given all of Trump’s deficiencies as a candidate, that’s a big margin for an election in which the “fundamentals” pointed toward a fairly close race. “Little Comey” believers have more room to assign blame to Clinton’s campaign, in addition to Comey (and the media’s coverage of him).

But campaign postmortems almost always involve a lot of results-oriented cherry-picking. It’s easy to single out things that Clinton did poorly — her handling of the email scandal, her inability to drive a positive message and her poor Electoral College tactics would have to rank highly among them (although those Electoral College choices probably didn’t swing the election). There are also some things the campaign did well, however. For instance, Clinton got a huge bounce after her convention, and she won all three debates according to polls of debate-watchers. She also made a fairly smart VP pick in Virginia Sen. Tim Kaine. These aren’t minor things; in normal presidential campaigns, preparing for the debates, staging the conventions and picking a solid running mate are about as high-stakes as decisions get. Clinton did poorly in the unscripted portions of the campaign, however, and the campaign went off script in the final 10 days.

If I were advising a future candidate on what to learn from 2016, I’d tell him or her to mostly forget about the Comey letter and focus on the factors that were within the control of Clinton and Trump. That’s not my purpose here. Instead, it’s to get at the truth — to figure out the real story of the election. The real story is that the Comey letter had a fairly large and measurable impact, probably enough to cost Clinton the election. It wasn’t the only thing that mattered, and it might not have been the most important. But the media is still largely in denial about how much of an effect it had.