Congratulations to Democrats on a very successful turnout effort, I guess.

As hard as this is to believe given how highly touted the Clinton GOTV effort was, it makes some sense once you realize where the fatal error probably lay. Democrats assumed that white women and working-class whites who had voted for Obama would stick with Clinton this time for the most part, especially in the Rust Belt. Michigan, Wisconsin, and Pennsylvania were part of the blue firewall, weren’t they? Well, there you go. No need to even visit Wisconsin. Their loyalty was taken for granted. So when the time came to call the troops onto the battlefield, the call went out — without fully realizing how many of them had already switched sides.

Volunteers for the Clinton campaign in Pennsylvania, Ohio and North Carolina have reported that when reminding people to vote, they encountered a significant number of Trump voters. Anecdotal evidence points to anywhere from five to 25 percent of contacts were inadvertently targeted to Trump supporters.

This is a big deal because when voters are engaged by a volunteer they are significantly more likely to cast a ballot in an election. To make matters worse, because Republicans had a non-existent ground game in many areas this cycle, this powerful reminder from a Clinton volunteer to get out and vote might have been the only personalized GOTV communication these Trump voters received.

The campaign’s text messaging GOTV effort may have been the worst offender. Volunteers reported as many as 30% of the replies they received from voters they were urging to get out were Trump supporters.

Is that true or too good to check? Consider that the piece was written by two advisors to Bernie Sanders, Hillary’s rival for the nomination. And as it so happens, the authors have a financial stake in this theory: They co-wrote a book criticizing the modern Democratic approach to organizing for national elections. Clinton, they note, followed the “small organizing” model, which relies on collecting commercial data about voters and using it to inform expensive predictive algorithms about which voters are leaning which way and who’s likeliest to turn out. The idea is to be as efficient as possible with your GOTV resources and target only the people whom you believe might bank a vote for you. The problem, say the authors, is that if the data or the algorithm is wrong, you’ll either leave a lot of gettable votes on the table or, worse, you’ll end up inadvertently turning out … your opponent’s voters. True to Sanders’s populist spirit, they prefer the “big organizing” model where campaign volunteers do lots of door-knocking and actually ask voters directly whom they support and whether they think they’ll end up voting. There’s no substitute for shoe-leather GOTV work. Conveniently, the anecdotes above above about Team Hillary grossly missing their targets on who was likely to turn out not only seem to confirm their thesis but are unsourced. Doesn’t mean they’re wrong or making it up, but it’s worth noting. If only because it’s high-larious to think that Hillary Clinton, the Smartest Woman In The World, inadvertently did Trump’s ground-game work for him.

The larger point, though, that Clinton badly underestimated how far working-class whites had drifted from the party, is obviously correct. Here’s something from a Politico piece this week that’ll haunt Democrats until the end of time:

And some began pointing fingers at the young campaign manager, Robby Mook, who spearheaded a strategy supported by the senior campaign team that included only limited outreach to those voters — a theory of the case that Bill Clinton had railed against for months, wondering aloud at meetings why the campaign was not making more of an attempt to even ask that population for its votes…

But in general, Bill Clinton’s viewpoint of fighting for the working class white voters was often dismissed with a hand wave by senior members of the team as a personal vendetta to win back the voters who elected him, from a talented but aging politician who simply refused to accept the new Democratic map. At a meeting ahead of the convention at which aides presented to both Clintons the “Stronger Together” framework for the general election, senior strategist Joel Benenson told the former president bluntly that the voters from West Virginia were never coming back to his party.

What would Bill Clinton know about winning blue-collar white voters, right? This too seems a little too good to be true, as it’s exactly the sort of thing you’d expect Clintonworld to leak in order to shift blame away from themselves to the dunderheads and amateurs who ran the campaign. But Bill did note on the campaign trail this year more than once that the, ahem, “coal people” no longer had much use for his party. That proved to be exactly right. If he was willing to say that publicly, it’s hard to believe he didn’t say it privately as well, especially given Trump’s high support all year among whites without a college degree. An elite, liberal, Ivy-League-educated establishmentarian who, by the way, was running to become the first woman president should have obviously required a more, not less, dogged sales effort to working-class whites, but evidently Hillary thought that the “Clinton” brand and her party affiliation were enough. And that’s how we got President Trump.

A fascinating gloss on all this comes from pollster Patrick Ruffini, who made his point in a series of tweets last night. (You’ll have to scroll through his timeline to read it in full as I don’t think it’s collected in any one place.) Ruffini realized in the course of doing research on primary voting this year that the best way to predict a vote for Trump was to look at not just the demographics of an individual voter but the demographics of his community. If you were white without a college degree and lived in a place with lots of other whites without degrees, you were more likely to vote Trump than you were if you lived in a place surrounded by white college grads. Which makes sense:

If Team Hillary had had that insight, that there’s likely to be a cascade effect among undecideds in communities that are already predisposed demographically to vote Trump, they surely would have realized the peril in neglecting rural Rest Belt areas. As it is, they may have looked at polls of those areas, seen Trump leading narrowly with, say, 10 percent undecided, and concluded that those undecideds would break softly enough for him that Clinton could still win statewide. If, however, they were wrong about that — if undecideds broke heavily for Trump because of the social effects Ruffini describes — then the Rust Belt could be gone in a flash, as ended up happening. This is where you should be thinking of yesterday’s post about all the late-deciders in Michigan, Wisconsin, and Pennsylvania who ended up voting Trump instead of Clinton. Community effects might explain a lot of that. Team Hillary apparently missed it. Oh well. She’ll have to console herself with her national popular vote win during her long retirement.