2016 Election Night

Donald Trump has been elected president of the United States. In an extremely narrow sense, I’m not that surprised by the outcome, since polling — to a greater extent than the conventional wisdom acknowledged — had shown a fairly competitive race with critical weaknesses for Clinton in the Electoral College. It’s possible, perhaps even likely, that Clinton will eventually win the popular vote as more votes come in from California. But in a broader sense? It’s the most shocking political development of my lifetime. We’re going to get some sleep, and then we’ll have much more to say over the next days and weeks about how Trump won and what it means for the country. We hope you’ll continue to join us on a regular basis.

To many people’s surprise, Trump won more Latino voters than Romney did in 2012. Here’s how the firm Latino Decisions found the Latino vote broke out by state, for Clinton and Trump.

• Arizona — 84-12
• California — 80-16
• Colorado — 81-16
• Florida — 67-31
• Illinois — 86-10
• Nevada — 81-16
• North Carolina — 82-15
• New York — 88-10
• Ohio — 80-17
• Texas — 80-16
• Virginia — 81-15
• Wisconsin — 87-10

Trump’s margin among Latino voters in Florida, though thinner than it has been for Republican candidates in past races, likely helped him win that critical state.

https://twitter.com/KellyO/status/796255033306337282

https://twitter.com/AP/status/796253849497571328

Here are the states where Clinton’s likely to beat Obama’s margin from 2012, according to current projections:

• California
• Texas
• Utah
• Arizona
• Georgia
• Massachusetts
• Washington
• Kansas
• the District of Columbia

There are … not a lot of swing states on that list.

There are several really close contests that haven’t been called, so you might be wondering what the recount laws are in those states. Here’s what we know: Florida (recount laws):

• A margin of 0.5 percent or less triggers an automatic recount of machine-tallied votes.
• If that recount brings the margin to 0.25 percent or less, that triggers a hand recount.

Michigan (recount laws):

• A margin of 2,000 votes or less triggers an automatic recount for the top two candidates.
• Candidates can petition for a recount if they believe there was fraud or counting error.

New Hampshire (recount laws):

• Candidates can request a recount if the candidate applying for a recount is behind by less than 20 percent of the total votes cast in towns where the election is contested.

Pennsylvania (recount laws):

• A margin of 0.5 percent triggers an automatic recount.
• Voters can petition county boards for a recount.
• Both voters and candidates can petition courts for a recount.

Johnson is at just 3 percent of the national vote. That may go up slightly, but he’s likely not going to get close to 5 percent. In the end, most voters ended up choosing either Clinton or Trump.

Trump has won one of Maine’s electoral votes. That means he’ll be the first Republican to win an electoral vote in New England since 2000. He’s the first to win an electoral vote in any New England state other than New Hampshire since 1988.

Something to remember: Whatever your feelings about the state of the country right now, it’s fundamentally not that different a place whether the final call is that Clinton has narrowly won or narrowly lost. Add just 1 percent to Clinton’s vote share and take 1 percent away from Trump’s, and she would have won Florida and Pennsylvania, therefore would probably have been on her way to a narrow Electoral College victory.

John Podesta just said that Hillary Clinton has no plan to concede tonight, as many states are still close (though most have been called for Trump). The last time we didn’t get a concession speech on election night was 2004. Kerry conceded the next day.

There was some talk about Trump winning in the Iron Range in Minnesota. Well, Clinton held onto it. For instance, she won St. Louis County (Duluth) by 12 percentage points.

Trump wins Pennsylvania. Our model now gives him a 93 percent chance of winning the election.

One thing I’ve been thinking about here is where the Democratic Party goes next. Three of the last four Congressional elections (2010, 2014 and 2016) were bad for the Democrats, leading to a thin bench. President Obama served out his two terms. The Clinton dynasty is over. But the most obvious alternatives to Clinton — Bernie Sanders, Joe Biden — are also pretty old. It seems that all of the energy in the party is on the left, and it wouldn’t be surprising if the 2020 nominee were someone from the Sanders wing of the party. But who is that candidate? I don’t know. There are a lot of opportunities for talented, up-and-coming, left-wing politicians, beginning with the 2018 midterms.

I wrote this morning about this election’s status as the first since a Supreme Court decision struck down portions of the Voting Rights Act of 1965, prompting states to close hundreds of polling places. Among the states affected were three crucial states won by Trump: Arizona, Georgia and North Carolina.

There’s been an ongoing debate about the sources of Trump support, and it will only heat up after tonight’s stunning result. But I wanted to offer a few quick preliminary thoughts. First, it’s critical to differentiate why candidates win primaries and why they win general elections. And second, in re-reading explanations from Trump’s primary supporters about why they backed him, one thing that stands out to me is their anger at the political system, a point Lynn Vavreck made as well. Any politician who wins a primary is going to gain support for lots of reasons. But as we are figuring out the reasons for Trump’s unexpected success, don’t discount the number of Republicans who saw politics as deeply broken — and saw a businessman and outsider as an answer.

CNN has called Alaska for Trump. Trump looks to be on his way to winning over 300 electoral votes, if current trends hold.

Which means, Harry, that the AP has essentially called the presidency for Trump.

The Associated Press has called Pennsylvania for Trump.

We expect that Trump will eventually finish with about 47 percent of the popular vote which, if he wins the Electoral College, would be the lowest vote share for a president-elect since Bill Clinton in 1992 (43 percent).

Pat Toomey, the incumbent Republican senator from Pennsylvania, will keep his seat. Our model now gives Republicans a 99 percent chance of retaining control of the Senate.

I’ve been watching to see if San Francisco would vote to allow 16- and 17-year-olds to vote in local elections, as they’re allowed to do in a few other countries and a handful of places in the U.S. Prop F needed a majority, and it looks like it’ll come up just short, with 46.3 percent of the vote counted so far.

Jason Kander is probably going to lose the Senate race in Missouri. He’s down by 5 percentage points with 89 percent of precincts reporting. Still, he ran 16 percentage points ahead of Clinton in the presidential race in Missouri. Unfortunately for him, Clinton is losing by so much in the state (21 percentage points) that Kander’s overperformance is not enough.

With several news outlets calling Pennsylvania for Republican Senator Pat Toomey, the best the Democrats could plausibly now do in the Senate is to pick up 4 seats: Illinois (which they won earlier tonight), New Hampshire (not yet called), Missouri (not yet called, but Democrat Jason Kander is trailing significantly) and Louisiana (where they’ll probably have one of the top two finishers, leading to a runoff). Even if Democrats draw that kind of inside straight, however, the scenario would yield just a 50-50 Senate, probably with a President Trump.

We’ll never know exactly how much FBI Director James Comey’s letter to Congress on Oct. 28 about new emails in the Clinton investigation hurt her chances of winning the presidency or Democrats’ chances of winning the Senate. But we do know that her lead in polls fell by about 3 percentage points after the letter — and Democrats likely won’t soon forget that.

Would Bernie Sanders have done better against Trump than Clinton? Many people, including me, thought Sanders would do worse because of his very progressive ideology. But perhaps ideology isn’t as important a factor in voters’ minds. Given Trump’s victory, being an outsider may have trumped an ideology that some might see as extreme.

As of current vote counts, the number of voters who cast ballots for candidates other than Clinton and Trump exceeds Trump’s winning margin — or lead, in races that haven’t yet been called — in many important states, including Arizona, Florida, Michigan, Pennsylvania and Wisconsin. But don’t pin Trump’s win on those voters who eschewed the two major candidates. Not all of them would have voted for Clinton had they been forced to choose only between her and Trump. And some might not have voted at all. Far more Democrats in Florida in 2000 voted for George W. Bush than voted for Ralph Nader.

The Associated Press has called Toomey the winner of the Pennsylvania Senate race. The Republican Party has officially retained control of the upper chamber of Congress with that victory.

Thinking locally here in New York, the split in the vote in the city itself mirrored that in the country as a whole, with Trump making massive gains in Staten Island but getting even fewer votes than a Republican usually does in Manhattan:

Republican results by New York City borough

BOROUGH	ROMNEY	TRUMP
Manhattan	14.9	10.0
Brooklyn	16.9	17.9
Queens	19.9	22.1
Bronx	8.1	9.6
Staten Island	48.1	57.2

Last week, I wrote that the key to understanding the current moment in American politics was that parties are weak but partisanship is strong. The parties have limited control over whom they nominate, but partisans will stick with their party and its nominee regardless. I stand by this: After everything that happened in the primaries, most Republicans lined up behind Trump. What I said works pretty well on the Republican side. But if the Trump victory that looks likely becomes a reality — or even if it doesn’t — there will still be questions about what has happened in the Democratic Party, which generally enjoys an advantage in partisan identification. It seemed likely that a coalition of nonwhite voters, combined with women voters, would offer Clinton — or any Democratic candidate — a solid demographic advantage. I think we may have to ask some hard questions: Were Clinton’s scandals too much for partisanship to overcome? Do even Democratic partisans resist sending their party to the White House for a third term? Is Democratic partisanship intrinsically weaker than Republican partisanship? And then, the hardest question of all: Does partisanship break down when the nominee is a woman?

The marijuana ballot measure in Nevada passed, legalizing recreational pot for adults ages 21 and older. With 54 percent in favor and 46 percent opposed, the results lined up with the polls we looked at.

Three states have been grappling with capital punishment ballot measures. Nebraska reinstated its death penalty, Oklahoma strengthened its, and now we’re down to the most populous state in the union: California. With 24 percent of the vote reporting, California is voting to retain the death penalty, 55-45. California has more than a quarter of the country’s death-penalty population but hasn’t executed anyone since 2006.

Hillary Clinton could still conceivably win the election — or she could lose the national popular vote. But since both outcomes look unlikely, we should start preparing ourselves for the possibility of the second split between the national popular vote and the electoral vote in the last five presidential elections. A coalition of 11 states with 165 electoral votes between them has agreed to an interstate compact that, once signed by states with a combined 270 or more electoral votes, would bind their electors to vote for the winner of the national popular vote — in effect ending the Electoral College. New York just joined this week. It wasn’t enough to affect this election, but maybe today’s result will spur more states to join.

What’s behind Trump’s success? That is a massive question — one journalists, political scientists and, well, everyone will be asking for years to come. But let’s start with one potential answer, however small its eventual explanatory power: trade. Trump has lambasted trade deals like NAFTA and often pointed a finger at trade with China and Mexico as causing the decline in U.S. manufacturing jobs. He’s pitted his campaign against “the globalist elites” — the pro-trade politicians in both parties. As we’ve noted before, though, the trade debate is really a debate about China. NAFTA’s effect pales, in comparison. So let’s focus just on Chinese trade for the moment. Recent research has indicated that trade with China has been more disruptive than previously thought. MIT economist David Autor and co-authors have documented how rising Chinese imports wreaked havoc on competing U.S. industries. In total, their research found the surge of Chinese trade was responsible for the loss of more than 2 million jobs between 1999 and 2011. But, interestingly — and this is where Trump’s electoral map comes in — it had a concentrated geographic impact. States in the Midwest, Appalachia and the Southeast were where Chinese trade hit hardest. Take a look at these maps showing where the U.S. industries were most exposed:

At first look, this map sort of overlaps with Trump’s success. He has won or is currently leading in several manufacturing-heavy Midwestern states; anti-trade sentiment is rife there. Furthermore, Autor and co-authors, in a separate paper, compared rising Chinese imports with the increase in political polarization. And, consistent with this fledgling thesis, they found that areas exposed to heavy trade later elected more conservative (or more liberal) politicians. Globalization might have begat polarization. Now, to be clear: This is not the definitive story; it’s just a thesis to float. But it will be studied for years to come.

The polls are now closed everywhere. We’re left counting votes!

The national exit poll shows Trump making bigger gains among black and Hispanic voters than among whites. But I’d urge at least a little caution. I know that exit polls aren’t supposed to be used for projecting results, but they did an awfully bad job tonight, initially showing what had looked like a near-landslide margin for Clinton. Furthermore, as compared with pre-election polls, Trump clearly overperformed the most in whiter states. So on second thought, maybe that’s a lot of caution and not just a little.

Fox News has called Wisconsin for Trump. There doesn’t seem to be enough votes left in Pennsylvania for Clinton to come back. If that’s right, it means Trump is likely the president-elect.

FiveThirtyEight’s Walt Hickey and I are at ABC News’ studios in Times Square. A crowd gathered outside to cheer and boo when Clinton and Trump win states. The crowd is what you would expect in New York: multiracial, international (as tourists boggle at the American election spectacle) and showy, with one man draped in an American flag. Without stroking our own ego, I find it striking that FiveThirtyEight’s model stood up to the uncertainty of this race while being criticized as opportunistic for not declaring a Clinton victory. I myself, to be honest, thought Clinton had a better shot at a comfortable lead. But while I was wrong, I also remember the passion of Trump’s true believers and the way that people on the knife’s edge of supporting Trump or voting third party or write-in fell on the side of Trump. How do we parse this race? We can look at the influence of the Supreme Court on voter choice, with a fifth of voters marking it as their most important issue and voting overwhelmingly for Trump. We can look at the “change” voters — 87 percent of them picked Trump. But we can also look at the popularity of minimum-wage ballot initiatives, which passed in all four states where they were voted on — Arizona, Colorado, Maine and Washington. We can look at the participation of American men in the labor force, which has dropped from 87 percent in January 1948 to 69 percent this October. How do we disaggregate the effects of race, gender, class, social conservatism and labor economics? Maybe in the end we cannot. But we have to begin to assemble the puzzle called America, which is far more diverse, in its own way, than the crowd gathered in Times Square.

At times, politics seems focused on tangible questions: Who can manage the economy, or how should we tax different groups? At other times, it seems to hinge more on symbols. Symbolic politics is emotion-laden and often grounded in questions of group status; as a result, it is often zero-sum. To make sense of this election, I’ve found myself returning frequently to a chapter by David Sears on symbolic politics. From a wall on the Mexican border to discussions of “deplorables,” from email security to paying for college, so many of the issues that have been foremost in our minds have been less realistic policy proposals and more symbols of group status. The question before us is how do we translate such a symbolic campaign into the concrete guidelines for governance.

Despite the amount of time we spend talking about these two states around nomination time, they don’t get quite as much attention during the general election. Both states have gone blue in four out of the last five elections (before today) — Bush won Iowa in 2004 and New Hampshire in 2000. In retrospect, that Iowa was predicted for Trump and New Hampshire was (and still is at the time of this writing) too close to call should have been a clue to the dynamics of this election. Perhaps partly because of their importance in nomination politics, these are two politically engaged states, with among the highest voter turnout in the nation. New Hampshire, in particular, boasts a brand of maverick-y and libertarian politics. The turn of the race in these states may have been an early clue about the appeal of Trump’s politics, a possible coming Republican wave and, of course, the tendencies of white voters.

Comparing pre-election polls — as based on our adjusted polling average in each state — against current results, as projected by the Upshot including their forecast of the outstanding vote, is pretty interesting. Clinton’s underperforming her polls by about 3 points in the average swing state, but with a lot of regional variation:

Where did polls miss the most?

	CLINTON’S PROJECTED MARGIN
STATE	ADJUSTED POLLING AVERAGE	BASED ON VOTES SO FAR	SHIFT
Maine	+6.9	+2.6	-4.3
Minnesota	+5.9	+0.7	-5.2
Virginia	+5.4	+2.8	-2.6
Wisconsin	+5.4	-2.2	-7.6
New Mexico	+5.3	+7.8	+2.5
Michigan	+4.0	-0.8	-4.8
Colorado	+3.8	+4.9	+1.1
Pennsylvania	+3.7	-0.8	-4.5
New Hampshire	+3.5	+0.4	-3.1
Nevada	+0.7	+1.3	+0.6
North Carolina	+0.7	-3.8	-4.5
Florida	+0.6	-1.4	-2.0
Ohio	-2.0	-8.1	-6.1
Arizona	-2.4	-5.0	-2.6
Iowa	-3.4	-9.5	-6.1
Georgia	-4.0	-5.1	-1.1

Her biggest underperformances are in Wisconsin, Ohio and Iowa. Meanwhile, it looks as though Clinton might actually beat her polling average in New Mexico, Nevada and Colorado.

The Clinton campaign is basically hanging on by a thread at this point. She needs to pull out Michigan and Pennsylvania and then hit one of two scenarios: 1. Win Alaska and New Hampshire or 2. win Arizona. That’s going to be extremely difficult. It’s not impossible, but it’s a straight, if not royal, flush.

Hey friends. Just a word about our plans on the audio/video front. We’re not going to do any more live video updates tonight — the race is still in limbo, and we feel like our best contribution is on the blog itself. Some of our team is going to catch some sleep. We’ll record a podcast tomorrow, mid-morning New York time, and get it up as quickly as possible. Feel free to get in touch by Twitter with any questions you feel like we need to discuss during that show.

Yes, this probably wasn’t a huge polling miss — it looks like Clinton will win the national popular vote by 1 percentage point or so, not the 3 or 4 points she led by. But it’s still a miss, and what looks like a decisive one. Earlier I mentioned some other recent important polling misses worldwide — and readers reminded me of a few I didn’t list. They include the recent Colombia peace referendum, plus general elections in Argentina last year and in Brazil in 2014.

The marijuana ballot measure in Arkansas passed, legalizing medical marijuana for specific debilitating medical conditions. With 53 percent in favor and 47 percent opposed, the results lined up with the polls we looked at.

Democrat Catherine Cortez Masto has won the U.S. Senate race in Nevada. Our model now gives Democrats a 6 percent chance of winning control of the Senate.

Clinton wins Nevada. Our model now gives her a 21 percent chance of winning the election.

In the October 2016 wave of the ongoing Institute for the Study of Citizens and Politics study, we asked respondents to place Trump and Clinton on a 7 point ideology scale. On average, voters put Trump 0.64 points to their right — and put Clinton a whopping 1.88 to their left. Put differently, the average voter saw Clinton as decisively to their left, while ranking Trump’s views as closer to their own. They also had more uncertainty about exactly where to place Trump. Should he win, one of the challenges he will likely face is that he may be forced to clarify just how conservative he is through the process of governing.