Donald Trump will be the next president of the United States.
That's remarkable for all sorts of reasons: He has no governmental experience, for example. And many times during his campaign, Trump's words inflamed large swaths of Americans, whether it was his comments from years ago talking about grabbing women's genitals or calling Mexican immigrants in the U.S. illegally "rapists" and playing up crimes committed by immigrants, including drug crimes and murders.
But right now, it's also remarkable because almost no one saw it coming. All major forecasters predicted a Hillary Clinton win, whether moderately or by a landslide.
So what happened?
We don't know just yet why pollsters and forecasters got it wrong, but here's what made this electorate so different from the one that elected Barack Obama by 4 points in 2012. To be clear, it's impossible to break any election results out into fully discrete demographic groups or trends — race, gender and class are interconnected, impossible-to-disentangle phenomena. But, for now, here's what the exit polls, as reported by CNN, do tell us about a few of those trends that handed this election to Trump.
Men backed Trump
Women's voting preferences haven't shifted much since the time when they preferred Obama by 11 points. However, men swung more heavily Republican, by the latest exits. In 2012, men preferred Mitt Romney by 7 points. This year, they preferred Trump by almost double that.
The exact counts are still likely to shift a bit but, right now, 24 percentage points separate America's women from America's men in the latest exit polls — that is, women preferred Clinton by 12 points, while men preferred Trump by 12 points. If that ends up being the final tally, it will be the largest gap measured by exit polls since at least the 1950s.
Non-college-educated whites, in particular, love Trump
Trump specifically appealed to whites during this election, and they clearly preferred him at the polls — right now, his margin of victory among them is roughly the same as Romney's in 2012, which was 20 points.
But there's a massive fissure between college- and non-college-educated whites. Romney won non-college-educated whites by 26 points, according to polling data provided to NPR by the Pew Research Center. Currently, exit polling shows Trump's margin among that group to be roughly one-and-a-half times that.
Meanwhile, college-educated whites appear to have preferred Trump slightly, according to current exit poll figures (which, once again, could still shift some). That would mean Trump moderately underperformed Romney, who won this group by 14 points, according to the data from Pew.
However, that would still essentially be a win for Trump, considering that at one point, it looked like Trump might be the first Republican in decades to lose white, college-educated Americans.
Clinton failed to perform in key counties
Clinton underperformed Obama — sometimes heavily — in key counties in swing states. For example, Obama beat Romney by more than 381,000 votes in Wayne County, Mich. — by far the state's most populous county. Clinton beat Trump there only by around 287,000 votes, meaning there's a more than 90,000-vote gap between her win and Obama's win there.
In Cuyahoga County, Ohio — home to Cleveland — Trump drew about 4,500 fewer voters than Romney did in 2012. However, Clinton drew almost 37,000 fewer, giving her a much smaller win there than Obama scored in 2012.
In Milwaukee County, Wis., where it appeared Trump had a much bigger early lead (as we reported earlier), the gap closed considerably as the hours wore on, to around 6,700, by the latest data. Likewise, Clinton trailed Obama's margin in Philadelphia County, Pa., by around 10,000 votes. Those aren't huge gaps, but they do reflect the broader national trend of Clinton losing a bit here, a bit there ... adding up to a loss.
Conservatives don't seem married to traditional conservatism
Right now, the numbers for conservatives are looking about the same this year as they did in 2012 — that is, around 80 percent of them went for the Republican candidate in both elections. And in both, they made up just over one-third of the electorate.
But, then, Trump is not at all the kind of conservative that Romney was; indeed, he's not at all a traditional conservative. Trump has rejected free trade agreements, called for up to $1 trillion in government spending on infrastructure, and introduced a tax plan that could balloon the debt by $7.2 trillion in one decade, by one estimate.
That makes it pretty clear that Trump voters weren't driven by far-right ideology (unless many self-proclaimed conservatives had big changes of heart since 2012). Trump's populist, overtly masculine, anti-PC appeal helped him vault past Clinton.
Those leery white evangelicals? They weren't so leery after all
Right now, the polls show that 81 percent of white, self-described evangelicals voted for Trump. That doesn't look too different from 2012, when 78 percent of white, born-again Christians (a term that pollsters often use in place of, or in conjunction with, the term "evangelical Christians") chose Romney.
All of which is to say that despite white women evangelicals' apparent reluctance to choose Trump, and despite some evangelical leaders' arguments against Trump, this group in the end didn't shift much at all.
Trump did well among voters who didn't really like him
You'd think that when people see a candidate unfavorably, they'd vote against that candidate.
That didn't happen this year — at least, not the way it did in 2012. In 2012, 94 percent of voters who saw Obama unfavorably chose Romney. Likewise, 92 percent who saw Romney unfavorably chose Obama.
But as Amy Walter pointed out at the Cook Political Report Tuesday morning, 77 percent of voters who saw Trump unfavorably voted for Clinton; 15 percent of those people still voted for him. The numbers were slightly less stark among Democrats — only 82 percent of people who saw Clinton unfavorably chose Trump; 11 percent chose her anyway.
Update: This post was updated at 8:32 AM with the section about Trump and Clinton's favorability numbers.