This quote, from a spectacular Bloomberg story about the Donald Trump campaign’s final electoral push, neatly summarizes the giant gamble that the campaign is taking: “There’s really not that much of a difference between politics and regular marketing,” an unnamed senior official told Bloomberg’s Joshua Green and Sasha Issenberg.
But before we get to that, let’s unearth the buried story-within-the-story.
“We have three major voter suppression operations underway,” another campaign official told Green and Issenberg. The campaign is targeting young white liberals, young women and black voters with negative ads focused on Hillary Clinton’s politics, Bill Clinton’s past and comments Clinton made in the 1990s about black criminals.
The reason the Trump campaign wants to suppress Clinton’s turnout is obvious. Since May 1, shortly before Trump clinched the nomination, he’s led Clinton in the head-to-head RealClearPolitics polling average for eight days. Out of 179.
Over the past month, Clinton has averaged 48 percent on that metric; Trump has averaged 42.9. More importantly, Trump has been above 44 percent on only 17 days. Clinton's been under 44 on only 24.
Trump’s got a ceiling. Clinton’s got a floor. He’s tried (sort of halfheartedly) to expand past his ceiling. He did, once, right after the Republican convention, when he surged into a lead for four days. That’s the second lead on that first graph. It didn’t stick. The other time he had a lead was after Clinton’s numbers collapsed in late May. That’s the model Trump is aiming for on Election Day: If he can’t go up, drag Clinton down. If 20 percent of voters stay home, who cares? Elections are about getting more votes from people who actually vote, not about necessarily capturing the broad will of the American people.
This also makes clear a point that FiveThirtyEight’s Nate Silver has been hammering on for the past week: Turnout models that are driving the polling could be incorrect. Turnout this election cycle could end up being a lot Trump-friendlier than pollsters assume — which is an argument that Trump’s team has been making for a while. Clinton is very likely to win at this point, assuming that the electorate looks the way pollsters think it will. If it doesn’t — if Trump reshapes it, for example — those polls will turn out to be wrong. That’s what happened in 2012: Pollsters assumed the electorate would be a lot Mitt Romney-friendlier than it was.
Articulating the operating rationale shouldn’t detract from the fact that this is a highly unusual thing to cop to. Normally campaigns try to play down the extent to which they’re using tricks and hacks to win, preferring to emphasize their policy chops and their policy vision for America. Trump’s people have continually clutched their pearls about how Clinton’s team is running a negative campaign, asking ironically why Clinton’s folks won't focus on issues. Their path to victory now, though, is to do precisely the opposite — and to talk to the press about it.
To be fair, it may be the only way Trump can win. Perhaps more important, though, is how they plan to make this happen.
From Bloomberg’s story:
In San Antonio, a young staffer showed off a South Park-style animation he’d created of Clinton delivering the “super predator” line (using audio from her original 1996 sound bite), as cartoon text popped up around her: “Hillary Thinks African Americans are Super Predators.” The animation will be delivered to certain African American voters through Facebook “dark posts” — nonpublic posts whose viewership the campaign controls so that, as Parscale puts it, “only the people we want to see it, see it.”
This, explicitly, is Trump’s get-out-the-vote operation — or, sometimes, don’t-go-vote operation. It’s online, Facebook posts and social media. It’s an unusual way of going about it, but Trump’s team thinks it will work.
Quote No. 1, from an unnamed source: “If you’re running a burger shop, you have to let people know that your burgers are good and get them into your shop to buy them. It’s pretty similar with voting: You have to find out what people want and then convince them why your product is the right one.”
Quote No. 2, from Brad Parscale, the campaign’s digital lead (who’s never run a campaign before): “I always wonder why people in politics act like this stuff is so mystical. It’s the same s--- we use in commercial, just has fancier names.”
On the surface, those arguments make sense. It’s just marketing! What’s the big deal?
The trick, though, is that it isn’t just marketing. In normal advertising, you need to get whoever to come and buy your product. Maybe it’s targeted to particular demographic groups, but you just want people to come buy your burgers (to use the example in quote 1). In elections, though, you need particular people to come vote: People who are registered to vote and who support you. You can broadly target demographic groups hoping to hit those folks — this is why campaigns buy television ads — but the trick to turnout operations is to drill deeper and target specific individuals directly.
You can target individual people with Facebook, as we have noted in the past. You can load in a voter file and send ads to very targeted groups. That becomes much more expensive, but you can do it. (In September, Trump was outspending Clinton on digital ads.) Trump doesn’t necessarily need to dive that deep for his suppression efforts, but for turning voters out, he’d need to.
What’s not known is if Facebook nudges will actually affect how and when people vote. Facebook ran an experiment in 2010, showing that social cues on the social network could boost voter turnout. But that’s different from getting particular people to take action or not based on a targeted ad. (Green and Issenberg note that the effect of trying to suppress the vote could just as easily be to inspire people to go to the polls; it’s simply not clear how people will react.) This isn’t about Trump trying to sell his brand to as many people as possible. It’s about Trump trying to sell his brand to particular people.
The new Post-ABC tracking poll serves as a reminder that an attempt to suppress Clinton’s vote could work: A lot of voters view her negatively and are anxious about having her as president. It’s just not clear whether running Facebook ads — or using a new-media marketing strategy — can get that done. Early-voting numbers so far don’t indicate any big drop-off among likely Clinton supporters, though such a pattern would probably be hard to detect at this point.
“Growing the digital footprint has really allowed us to take his message directly to the people,” one source told Bloomberg, noting that people were so passionate on Reddit that the link-sharing community had to change its rules governing user behavior. We don't know which anonymous source this was, or how high up he was, but this falls into the same trap as the burger-store problem above. Lots of enthusiastic Reddit fans and Twitter followers and people at rallies have gotten Donald Trump that 42.9 percent support in the polls.
So this is the Trump campaign’s bet, laid on the table by a group of people who haven’t run campaigns before. Given the inability to increase Trump’s support, they’ll try to drag down Clinton’s, betting mostly on the campaign’s online organizing.
We’ll close by noting that this is not happening in a vacuum. Clinton's team has online followers and Facebook ads, too — as well as a traditional get-out-the-vote operation. Can Trump show young women a Facebook ad about Bill Clinton’s affairs and get them to stay home even as Clinton volunteers are knocking on their door to get them to go vote?
I guess we’ll see.