Nobody serious is officially calling the presidential contest over. But as the summer nears its end, if Donald Trump is not yet toast, he’s certainly looking toasty.
The win probabilities for Hillary Clinton among the major forecasters are getting awfully high, ranging from 83 percent by FiveThirtyEight and DailyKos to 93 percent by the Princeton Election Consortium. The confidence in a Clinton victory among those willing to put down money on it is only a bit lower: Predictwise shows prediction markets currently assessing her odds of winning at 80 percent.
All these numbers, moreover, confirm what most observers who have not pledged allegiance to Donald Trump have believed all along: He is a political exception that proves all the rules, an accidental nominee whose strategy for winning (doubling-down on the GOP’s old-white-voter appeal and then just winging it) never made a lot of sense. Hillary Clinton has her weaknesses, the story goes, and that’s kept the outcome in some doubt for a while. But in the end Americans are not going to elect a crude, self-promoting real-estate developer and entertainment figure with aggressively eccentric views to the most powerful political office in the world, right?
That’s probably true. But you don’t have to cherry pick a few flawed, favorable polls (the way his campaign likes to) to understand Trump’s still in the game.
His secret weapon is that old devil polarization. The exceptional reluctance of an overwhelmingly partisan electorate to cross habitual party lines these days is limiting opportunities for Hillary Clinton to crush Trump in the polls or in November.
As a New York Times piece last week that held up an amber signal to landslide enthusiasts noted, it’s now been 32 years since a presidential candidate won by a double-digit margin (the traditional benchmark for a landslide). The reason is simple enough, though complex in its making: The ideological sorting-out of the two parties that has been under way since the 1960s has produced major parties that are far more coherent, loyal (even tribal), and hostile to the opposition than anything we’ve seen since the New Deal or perhaps the 19th century. This reality has sometimes been obscured by the increasing percentage of Americans who self-identify as independents. These are not by and large the automatic swing voters of yore; as a vast body of research literature has shown, the majority are stone partisans in their voting habits who simply prefer for a variety of reasons to self-identify as free of party bonds.
So the number of voters truly up for grabs in presidential elections has steadily declined, and with it the floor under major-party-candidate performance has risen.
The argument for a Clinton landslide really does come down to the idea that Donald Trump is so offensive to Republican voters that he will lose a ton of them. And that seems plausible given the almost endless cavalcade of Republican elected officials and conservative opinion leaders who have signaled in large and small ways that it is all right to dislike Trump, and perhaps even pass up the opportunity to vote for him. You really have to go all the way back to Barry Goldwater — the worst-performing GOP nominee of the 20th century — to see anything quite like it among Republicans.
But here’s the thing: It’s not all that clear that actual Republican voters will follow unhappy elites out of Trump’s voting column. And many of those who do will not vote for Hillary Clinton.
The most thorough examination of partisan crossover voting in recent weeks has been by Mark Blumenthal, examining Survey Monkey’s very large sample for its online tracking poll. According to his analysis, there are remarkably few defections from the partisan ranks, not just among self-identified Democrats and Republicans but even among independents leaning Democratic and Republican. He shows, for example, Hillary Clinton only capturing 5 percent of self-identified Republicans and only 3 percent of Republican-leaning independents. That is not the stuff of landslides.
Other recent surveys that break down partisan voting patterns vary, but do not generally show the kind of wholesale rank-and-file Republican defections to Clinton that elite Republican unhappiness might suggest. The percentage of Republicans currently planning to vote for Clinton ranges from a high of 12 percent in a Quinnipiac survey that gives Clinton a seven-point lead (in a four-way race) to a low of 2 percent in an Economist/YouGov tracking poll showing her up by 4 points.
It is true that Trump’s own margins among Republicans are held down by defections to Gary Johnson. But as Blumenthal’s analysis also shows, Johnson is pulling votes from both major candidates more or less equally, and when Jill Stein supporters are taken into account, the minor-party vote is hurting Clinton more than Trump. To look at it another way, if a good 15 percent of the electorate is out-of-bounds to the major-party candidates, the odds of someone like Trump winning with relatively low support levels go up. And as FiveThirtyEight’s Harry Enten noted last week, the minor-party vote (or at least Gary Johnson’s share of it; he doesn’t look at Stein’s) is looking pretty resilient and may not undergo the usual late fade.
All of this analysis is separate, of course, from the possibility of a “black swan” event late in the campaign — really bad economic news, a major terrorist attack, a negative revelation involving Clinton — that could boost Trump’s odds. And then there are the debates. Mitt Romney was looking pretty toasty in 2012 before the first presidential debate gave him a major boost in the polls. Yes, he still lost, but nobody was talking about an Obama landslide down the stretch.
So Trump-haters had best avoid irrational exuberance, and maybe bookmark the piece by Vox’s Dara Lind exploring the many ways a President Trump could create a living hell for his detractors by abuse of his executive powers. They are not out of the woods just yet.