Trump is toxically unpopular. He still might win in 2020.

President Trump is historically unpopular. His overall approval ratings languish at around 37 percent, and his specific actions often poll even lower. (On the eve of the Joe Arpaio pardon, just 21 percent of Arizonans favored such a move.)

As Trump moves ever-further into the toxic swamps of the far-right, his numbers keep sinking. If this keeps up, Trump will end up having to bow out of the 2020 race or face the prospect of losing to any halfway competent Democrat in a landslide ... right?

There's just one problem with this analysis: It fails to take into account the new reality of American politics — and in particular the extent to which the combination of polarization and negative partisanship has begun to change the rules of the political game in a way that could decisively benefit President Trump in the only poll that matters: the one held on Election Day 2020. President Trump may well lose that contest. He could decide not to run. He could resign before then. He could be impeached and removed from office. But he might also survive, stick around, run, and win another term in 2020 — even with historically low approval numbers.

Subscribe to The Week

Escape your echo chamber. Get the facts behind the news, plus analysis from multiple perspectives.

SUBSCRIBE & SAVE

Sign up for The Week's Free Newsletters

From our morning news briefing to a weekly Good News Newsletter, get the best of The Week delivered directly to your inbox.

From our morning news briefing to a weekly Good News Newsletter, get the best of The Week delivered directly to your inbox.

Sign up

Here's how it could happen.

For the sake of this thought experiment, let's bracket outlier events. Let's assume that Robert Mueller's investigation digs up lots of dirt, and several members of the 2016 Trump campaign end up indicted, but that there is no recommendation of felony charges for the president himself. Likewise, let's forecast that the midterm elections of 2018 lead to Republican losses in the House of Representatives but that the GOP remains in the majority, protecting the president from impeachment. Let's also assume there is no massive terrorist attack or major war that Trump can use to rally substantial mainstream support to his side.

Assume, in other words, that Trump's approval remains about where he's been stuck since mid-May, through weekly and sometimes daily displays of erratic behavior and a constant deluge of terrible press: between 36 and 38 percent.

Trump won the 2016 election with 46.4 percent of the popular vote. He was famously able to prevail over Hillary Clinton's 48.5 percent by narrowly carrying three states (Pennsylvania, Michigan, and Wisconsin) that in recent election cycles have been safely Democratic. Without those narrow victories that gave Trump the edge in the Electoral College, he would have lost. So let's also assume that to pull off a win in 2020, the president will need to do a bit better in the popular vote than last time: a solid 47 percent. (Of course, without viable Libertarian and Green Party candidates in the race, he would need to do even better than that, but there's no way to predict third- and fourth-party challenges this early. So let's just presume that Trump would need to win somewhat more votes in 2020 than he did in 2016.)

How would a Trump re-election campaign accomplish this?

We won't be able to answer that question with great precision until we see which Democrat ends up running against him. But we can answer in general: The Trump campaign (and the RNC, and Fox News, and Rush Limbaugh, and Breitbart, and the rest of the right-wing media complex) will work to convince Republican-leaning voters that however much they dislike (or have ambivalent feelings about) Donald Trump, they should hate and fear his opponent far more. "Come home, Republicans!" will be the message. "Yes, it's been a messy four years, but at least Trump isn't one of them!"

How will "them" be defined? If the election were being run right now, we'd see loops of protest footage showing antifa thugs beating up innocent protesters or taking a sledgehammer to a statue of Christopher Columbus interspersed with clips of some left-wing activist or professor praising the anarchists for their moral clarity and courage. Never mind that Trump's opponent in the presidential race, along with the Democratic leadership of the House and Senate, would almost certainly denounce such violence. Some leftist somewhere would say something stupid, and the right would fasten onto, highlight, and promote it endlessly for political gain. "These masked vigilantes display the murderous heart of the left for all to see! If the person running against President Trump denies it, it's only because he's playing politics. In truth, they're enemies of all you hold dear. Vote for Trump or America as you know it is over!"

The U.S. is a country of roughly 320 million people. All the Republicans need is for a tiny handful of them to regularly say or do something that can be used to portray "the left" as morally repulsive as well as uniquely threatening to "ordinary Americans." That's how negative partisanship works. Pushed far enough, it can become unnecessary to make any positive case at all. (Hillary Clinton tried this in 2016, but the strategy failed because she proved to be nearly as unpopular as her opponent.)

That will be the game plan: Demonize the other side so completely that just enough people vote not so much as Republicans but as Anti-Democrats. Combine that push with targeted acts of voter suppression in heavily Democratic districts of key swing states, and the effort just might deliver a second term to Donald Trump.