Everyone’s still groggy from the late night so forgive me for essentially just tossing numbers at you here instead of spinning them out into something more coherent.

· Remember that final RCP map showing the race surprisingly close, with Clinton winning 272/266? For all the well-deserved attacks on polling and pollsters last night, that map wasn’t far off. Trump will end up winning the electoral college “bigly” but the margins in the states that decided it for him were tantalizingly tight. RCP had Pennsylvania going blue by 1.9 points; it ended up going red by 1.1. That’s 20 EVs. Florida, 29 more EVs, went Republican by 1.4 points; RCP had it going red by less than half a point. And Michigan, where Trump will probably end up winning by less than half a point, was a mere 3.4-point advantage for Clinton in the RCP average, within the range of an unsurprising miss. That’s 16 electoral votes. If Clinton had been just 1.5 points better in each state, she’d have added 65 more EVs to her 218 — which would have put her over the top.

Wisconsin doesn’t qualify as a “near miss” for RCP’s average, as it had Clinton projected to win by 6.5 points. Trump never led a single poll there. In the end, he won the state by one thin point. If Clinton had been 1.5 points better there, she’d be near 300 electoral votes and we would have spent the day wading through media treatises about Trump’s surprising but ultimately futile strength among the white working class.

· As of this morning, with most votes counted, Clinton actually leads the popular vote nationally by around 160,000 votes. That may change, although it’ll end up very tight either way. What might not change is the fact that … both Clinton and Trump right now have fewer votes than Mitt Romney did in 2012, let alone Barack Obama. O ended up with nearly 66 million votes, Romney with slightly less than 61 million. Clinton and Trump are each sitting at a little more than 59 million. Huh. But note: These vote totals will rise. Trump may end up passing Romney (and Clinton?) in the end. But it wasn’t a blowout overperformance vote-wise.

· The most notable exit poll result of the night: Trump ended up slightly outperforming Romney among Latinos and blacks.

bl

Romney lost blacks 93/6 and Latinos 71/27, a net advantage for Dems of 87 and 44 points, respectively. Clinton’s advantage over Trump was 80 points and 36 points, respectively. A lot of ink will be spilled interpreting those results over the next four years. For amnesty fans, the fact that Republicans lost badly among Latinos again, if not as badly as in 2012, will be proof that they still have lots of work to do to get right with that group before it inevitably becomes a major electoral force. For border hawks, the fact that Trump improved on Romney’s showing with Latinos is proof that immigration doesn’t matter much and that an effective working-class pitch will cross racial lines.

As for Hillary’s diminished advantage with blacks, Sonny Bunch wonders: Will Democrats ever dare nominate a white candidate again?

· Trump only slightly outperformed Romney with whites, winning 58/37 versus Mitt’s 59/39 margin four years ago. Working-class whites, specifically, though, hugely outperformed their numbers four years ago, and that made all the difference for Trump in the Rust Belt states where blue-collar rural white voters are populous. Romney won working-class whites by 26 points in 2012. Trump?

wwc

That’s a 39-point advantage. (And note that he ended up winning white college graduates after all, a demographic that had tilted towards Clinton in poll after poll this year.) Whites without a degree weren’t a larger share of the electorate this time, mind you; Trump’s gains came solely through winning a vastly larger share of them than Romney had. If Trump had been even a few points weaker among that group, maybe those very tight races in Michigan, Pennsylvania, and Wisconsin end up tilting blue after all. Nate Cohn noted the phenomenon of Rust Belt and midwestern counties that had been competitive or gone blue in 2012 going heavily for Trump this time:

The Wyoming River Valley of Pennsylvania — which includes Scranton and Wilkes-Barre — voted for Mr. Trump. It had voted for Mr. Obama by double digits.

Youngstown, Ohio, where Mr. Obama won by more than 20 points in 2012, was basically a draw. Mr. Trump swept the string of traditionally Democratic and old industrial towns along Lake Erie. Counties that supported Mr. Obama in 2012 voted for Mr. Trump by 20 points.

The rural countryside of the North swung overwhelmingly to Mr. Trump. Most obvious was Iowa, where Mr. Obama won easily in 2012 but Mr. Trump prevailed easily. These gains extended east, across Wisconsin and Michigan to New England. Mr. Trump won Maine’s Second Congressional District by 12 points; Mr. Obama had won it by eight points.

These gains went far beyond what many believed was possible. But Mr. Obama was strong among white working-class Northerners, and that meant there was a lot of room for a Democrat to fall.

You can see that effect in action on the Times’s nifty election map. According to Harry Enten, the worst polling misses last night were — surprise — in states where whites without a college degree made up most white voters. And that’s not the first time polls have undercounted them. This map from Sean Trende, who anticipated the rise of the “missing white voters” shortly after the last election, was making the rounds on Twitter last night:

In the end, Trump needed something of a perfect storm to win. He needed Hillary to underperform with her black and Latino base; he needed white college graduates, who had been leery of him all year, to come home; he needed his own base of working-class whites to break massively towards him after being a bit more divided between the parties in 2012; and he needed them to show up in the right states, specifically, the Rust Belt. And that’s just what he got.