If you wanted to know why Kellyanne Conway admitted last weekend that Trump is behind, here you go. It’s because he is, according to the campaign’s own data.

Silver’s “polls-only” model gives Clinton an 84.7 percent chance of winning this morning, with a projection of 333 electoral votes.

Despite Trump’s claim that he doesn’t believe the polls, his San Antonio research team spends $100,000 a week on surveys (apart from polls commissioned out of Trump Tower) and has sophisticated models that run daily simulations of the election. The results mirror those of the more reliable public forecasters—in other words, Trump’s staff knows he’s losing. Badly. “Nate Silver’s results have been similar to ours,” says Parscale, referring to the polling analyst and his predictions at FiveThirtyEight, “except they lag by a week or two because he’s relying on public polls.” The campaign knows who it must reach and is still executing its strategy despite the public turmoil: It’s identified 13.5 million voters in 16 battleground states whom it considers persuadable, although the number of voters shrinks daily as they make up their minds…

Trump’s data scientists, including some from the London firm Cambridge Analytica who worked on the “Leave” side of the Brexit initiative, think they’ve identified a small, fluctuating group of people who are reluctant to admit their support for Trump and may be throwing off public polls.

Still, Trump’s reality is plain: He needs a miracle.

If Trump’s internal numbers showed that he was poised to win the election because, for instance, public polls were systemically underestimating the number of “hidden” Trump voters out there, I assume we’d be hearing a lot about that from Conway and via leaks to the media. The great danger for both parties now is complacent voters not turning out, whether it’s Democratic complacency at the idea that Clinton will certainly win or Republican complacency at the idea that Trump will certainly lose. If the campaign’s polling strongly undermined the idea of a likely defeat, Conway et al. would have every reason to trumpet it to energize the right. Instead it sounds like they’ve identified a smallish number of hidden Trump voters who might make the difference if the race ends up very close. But it’s not close enough, for now, to matter much.

So what do you do to make it closer? Bloomberg’s reporters note that neither Trump nor the RNC has done much to register and turn out the many, many millions of working-class white voters who don’t typically vote, which is criminal negligence insofar as those people have the numbers to hand the election to the GOP in a walk. Failing that, Trump can go on trying to win over undecideds from Hillary, but if he’s as far behind as Silver’s model suggests, he’d have to win those undecideds overwhelmingly to make up the difference. There’s no reason, barring a last-minute bombshell detonated against Clinton, to think he can. (Read Byron York for anecdotal evidence of the trouble Trump’s having with undecideds. Those who are sympathetic to what his candidacy represents may have given up due to suspicions about his temperament.) There’s one other way to narrow the gap — convince some of Clinton’s voters to stay home, pushing down her margins and giving those hidden Trump voters a fighting chance of making up what’s left of the difference. One of Trump’s advisors was surprisingly frank about that strategy:

“We have three major voter suppression operations under way,” says a senior official. They’re aimed at three groups Clinton needs to win overwhelmingly: idealistic white liberals, young women, and African Americans. Trump’s invocation at the debate of Clinton’s WikiLeaks e-mails and support for the Trans-Pacific Partnership was designed to turn off Sanders supporters. The parade of women who say they were sexually assaulted by Bill Clinton and harassed or threatened by Hillary is meant to undermine her appeal to young women. And her 1996 suggestion that some African American males are “super predators” is the basis of a below-the-radar effort to discourage infrequent black voters from showing up at the polls—particularly in Florida.

I don’t know why the advisor would reach for a loaded term like “voter suppression” when what he’s describing sounds like simple attack ads designed to weaken support for Clinton among her base. (One strategy calls for posting a video on Facebook targeted at black Democrats highlighting Clinton’s “super predators” comment.) Typically “voter suppression” is used to refer to techniques for stopping people who want to vote — especially blacks — from being able to vote. Team Trump is trying to convince Clinton’s base that they shouldn’t want to vote for her at all. That’s called … campaigning; if Team Hillary runs an ad about the “Access Hollywood” tape that’s targeted at Republican women, is that “voter suppression”? Why Team Trump would insist on using that phrase knowing how it’ll be used against them by the left, I don’t know — but another Trump advisor, presumably the same one quoted in today’s piece, used it a few weeks ago in an interview with this same media outlet to describe the campaign’s strategy. It’s dumb, but it has a certain trollish “I don’t care what the media thinks” bravado that Trump’s inner circle seems to relish.

But never mind that. The important question is, is there evidence that the three groups named in the excerpt are in fact being suppressed? Well, here’s some data from the NYT’s polling of three key swing states:

With margins like that, a lot of blacks would have to be “suppressed” for black turnout to drop to the point where Clinton is in danger. How about “idealistic white liberals,” i.e. Bernie Sanders fans, though? Is there evidence that they’re being alienated by the Wikileaks stuff? Well, although not every young left-leaning adult is a Sanders fan, they overlap enough that it seems meaningful that Clinton’s numbers among millennials are suddenly soaring — better than even Obama’s in 2012, if you believe some polls. Gary Johnson, whose strongest support has come from young voters, has also seen some of his base peel off recently in national polling, which is probably no coincidence. It seems likely that some millennials have moved from Johnson to Clinton in the home stretch, sold on the idea that this really is a binary choice after all. And as for young women, any damage done to Hillary by educating them about Bill Clinton’s sins has probably been mitigated by the tape and the sexual-assault allegations against Trump. Or, contrary to conventional wisdom, maybe gender isn’t the prime dividing line for them. Maybe young women, many of whom are college grads, prefer Clinton for the same reasons that most other college grads do, just as working-class white voters prefer Trump. Either way, there’s little evidence in polls that Clinton has begun to fade with women voters. Her margins among that group are almost always bigger than Trump’s among men.

One question from all of this, noted by a Twitter buddy: Why on earth would Team Trump telegraph their strategy this way, let alone give it a charged name like “voter suppression”? All that’s going to do is inflame liberals and hand them an easy talking point to rally these groups. Case in point, here’s Sanders tweeting about the quote this morning to his 3.7 million followers:

It’s especially stupid given how much Trump has invested in the idea that dirty pool is being played with the election results. Now Democrats have an easy counter in the media, that “voter suppression” of liberals is happening by design, through the Trump campaign’s own admission. I don’t get it.

Update: The RNC emails to say that the bit above about neglecting the gigantic population of unregistered white voters without a college degree simply isn’t true:

The RNC has engaged in massive voter registration efforts and has built multiple universes to target these exact voters.

We’re obviously targeting many other groups and demographics but that line from the Bloomberg story is completely false.