The demographics of Trump’s win

Oh, I suppose this is something I’ll come back to 1000 times or so, but for now, Brownstein’s early take:

Trump held the traditionally Republican states—he won all of the states Mitt Romney won in 2012—and did exactly what his campaign had predicted for months: battered through the Democratic defenses in the Midwest.

At the same time he repelled her push into the Sunbelt. As I wrote on Election Day:

The worst-case scenario for her is that Trump’s blue-collar blitz narrowly pushes him past her in some of the Rustbelt states she needs, while she cannot advance quite enough among minority and college-educated white voters to overcome his non-college-educated, non-urban, religiously devout coalition in Sunbelt states like North Carolina, Florida, Nevada, and Colorado, much less Arizona and Georgia. Transitioning between her party’s past and future, Hillary Clinton’s nightmare is that she might be caught awkwardly in between.

For Trump the key to that pincer move was his remarkable success among white working class voters.

As polls had predicted for months, the Trump coalition was centered on white voters without a college education. Exit polls posted on showed him crushing Clinton among those voters by enormous margins almost everywhere, particularly in the South. Trump beat Clinton among non-college whites by 18 percentage points in New Hampshire, 21 in Colorado, 22 in Arizona, 24 points in Wisconsin, 31 points in Michigan, and 35 points in Missouri. The margin swelled to enormous margins in Southern states: 34 points in Florida, 40 points in North Carolina, fully 64 points in Georgia. Even in states where Clinton ran well overall, like New Jersey, Pennsylvania, Wisconsin, and Washington, Trump’s margins among blue-collar whites were enormous.

In several cases, those showings represented significant declines for Clinton relative to Obama in 2012. According to the exit polls as of around 10 p.m., her share of the vote among non-college whites, relative to Obama’s showing in 2012, fell 14 points in Maine, 13 points in Michigan, 12 points in New Hampshire, 11 points in Colorado, 10 points in Wisconsin, nine points in Pennsylvania, and six points in Florida…

Overall, the national exit poll showed her improving among college-educated whites over Obama in 2012, but only by three percentage points, and losing them narrowly to Trump. (As a result, the record of no Democrat ever winning most college-educated whites remained intact.) Despite strong performances among minority voters, that left her with too narrow a coalition to withstand the Trump blue-collar surge…

Yet this did not prove enough, in enough places, to withstand the non-college and non-urban surge for Trump. Blue-collar counties in the key Rustbelt states turned sharply toward Trump. In LaCrosse, Wisconsin, Obama won 58 percent of the vote in 2012; Clinton dropped to 44 percent in results as of around midnight; in Racine she fell from Obama’s 51 percent to just 37 percent. Even in Pennsylvania, her vote share in Erie fell to 47 percent, down from Obama’s 58 percent; in Lackawana (Scranton) she dipped to almost exactly 50 percent, down from Obama’s 63 percent last time. In Macomb County outside Detroit, renowned as the birthplace of the Reagan Democrats, Clinton skidded from Obama’s 52 percent to just 41 percent as of midnight.

In an election that became virtually a cultural civil war between two Americas, Trump’s side proved much more enthusiastic and united than Clinton’s. And it has now propelled America into an unexpected, and perhaps, unprecedented, experiment.

And, damnit, we’re all (Finnish friends, etc., aside) in this beaker together.  Ugh.


Via a Waldman tweet:

Wow.  Obviously, that most unpopular candidates ever really did have an effect on turnout.  What an amazing drop-off in eight years.  Definitely want to see more good analysis of this.


A million different ways to look at this election.  In the end, I’ll always remember this one.  Given perhaps the most abnormal (and not in a good way), Republican voters chose to vote as if they had a normal nominee.  Tons of Republican elites with real integrity told them otherwise, but, alas, that seemed to matter not a wit.  (From CNN exit poll page)


That’s really not very different from Romney.  Among the more surprising/disturbing results of this election is that R’s were more loyal to Trump than D’s were for Clinton.

A short counter-factual

Imagine what Republicans and Fox, etc., would be saying today if the last two Democratic presidents to be elected had both been initially elected while losing the popular vote.  Of course the results are legitimate, but the system desperately needs to change.  All voters matter.

That said, a couple of things

Soon, I’m just going to take a benzo and go to bed, but I did feel compelled to share a few good demographic snippets from 538 tonight:

1) Damn, those non-college whites.  This chart is the story, as far as I’m concerned.


2) And, this, I find just shocking.  Okay, I get understand Black voters dropping off a little from Obama, but Hispanics??

There’s going to be a lot of talk about white voters after the election, but looking at the exit polls, that’s not the full story. A big part of the story is that Clinton underperformed Obama with blacks and Hispanics. Clinton is winning only 88 percent of the black vote. Exit polls in 2012 had Obama at 93 percent. Clinton is only at 65 percent among Latinos. Obama won 71 percent of them.

3) Damn, that education gap among white people– especially white women!

But a deeper dive into exit poll numbers reveals something pretty fascinating, which is a split in the vote of college-educated white women and non-college-educated white women.

Non-college white men 72% 23% T+49
Non-college white women 62 34 T+28
College white men 54 39 T+15
College white women 45 51 C+6
Gender and education

Data based on preliminary exit poll results


College-educated white women voted for Clinton 51 percent to 45 percent, but non-college-educated white women voted for Trump 62 percent to 34 percent. That difference is nothing but stark and something we saw inklings of in October, when I wrote about how many Republican women were willing to overlook Trump’s history of sexual harassment allegations and derogatory comments about women. Partisanship is a hell of a drug.

4) This tweet:

5) I’ve been doing a lot of thinking about the nature of modern partisanship and how it is seemingly overwhelmingly cultural/identity politics these days.  It would seem Democrats really need to change this dynamic.  A longer post on that to come.

I’m sorry

I’m sorry I don’t have more cogent and interesting things to share with you tonight.  I’m finding a detached numbness best serving my mental health.

I’m sorry that my country sucks.  I’m sorry that my state sucks.  I’m sorry that a plurality of Americans thinks the most obviously unfit man, perhaps ever, to run for president somehow deserves the office (if nothing else, recall the absolutely unprecedented number of Republican newspapers and conservative thinkers that endorsed Clinton).

I’m sorry that racists, xenophobes, anti-semites, and misogynists will now obviously feel empowered.

I’m sorry that our media treated Trump like he was never going to win, so, “locker room talk” aside, they never had to seriously question his breathtaking unfitness for office.

I’m sorry that polling is clearly in crisis.  I study public opinion for a living.  This stuff matters to me.  I believe in what we do as social scientists.  And I still do.  But when it comes to predicting elections?  Well, damn.

And some point I’m sure I’ll want to write a lot.  That point is not tonight.

%d bloggers like this: