Want these election updates emailed to you right when they’re published? Sign up here.
This weekend was a letdown for those of us hoping for new, high-quality national polls to test how last week’s debate affected the campaign. The only new national polls we received were updates to the USC Dornsife/Los Angeles Times and UPI/CVoter tracking polls, both of which have actually moved slightly toward Donald Trump, but they still contain a mix of pre-debate and post-debate data. Meanwhile, an ABC News/Washington Post poll showed Hillary Clinton expanding her favorability rating gap with Trump — her numbers are bad, but his numbers are worse. But that poll apparently didn’t actually ask respondents who they were voting for.
So forecast-wise, we’re in the same place that we were on Friday: It’s pretty clear that the debate helped Clinton, but there’s some doubt about the magnitude of her bounce. It’s plausible that she’s gained only 1 or 2 percentage points, increasing her lead over Trump to about 3 points overall. That’s what our models have accounted for so far, enough to make her a 67 percent favorite according to our polls-only forecast and a 64 percent favorite according to polls-plus. Or her bounce could prove to be larger than that, especially given that Trump has woken up to — or stayed up all night tweeting about — a host of bad stories in the week since the debate.
There was one poll that caught our eye, though, and it was from New Mexico. The survey, from Research & Polling Inc. for the Albuquerque Journal, showed a competitive three-way matchup, with Clinton at 35 percent, Trump at 31 percent, and former New Mexico Gov. Gary Johnson at 24 percent. Because New Mexico hasn’t been polled much, the survey had a fair amount of influence on our forecast, reducing Clinton’s chances of winning New Mexico to 82 percent from 85 percent in the polls-only model.
Most of the time, Trump would be the beneficiary of a Clinton loss in New Mexico. But the model also assigns Johnson, the Libertarian Party’s presidential nominee, an outside chance — 2 or 3 percent — of winning the state. That could lead to an Electoral College deadlock that looks like this:
In this map, via 270towin.com, Clinton has 267 electoral votes, Trump has 266, and Johnson has New Mexico’s five. With no candidate possessing an Electoral College majority, the election would go to the House of Representatives, with Clinton, Trump and Johnson all eligible to receive votes.
Each candidate might be able to make a claim to legitimacy, of sorts. Trump might argue that the outcome showed that voters had profoundly rejected the status quo — and what could be a bigger rejection of the status quo than a President Trump? But more importantly, he’d have a sympathetic audience, since Republicans are likely to control the majority of congressional delegations. Clinton would probably have won the popular vote in this scenario, since she’s more likely to win the popular vote while losing the Electoral College than the other way around, according to our forecast. And Johnson might try to position himself as some sort of compromise choice.
It’s a somewhat plausible map too, with Clinton winning all her “must-win” states except New Mexico, which she loses because of Johnson’s native-son status. In the Albuquerque Journal’s poll, Clinton led Trump by 10 percentage points in a two-way matchup — about what you’d expect in New Mexico given the state’s demographics. But the poll showed Johnson (and Green Party nominee Jill Stein) taking disproportionately from Clinton’s support instead of Trump’s.
But plausible is a long way from likely. It’s not far-fetched to think the Electoral College would be close enough that New Mexico would make the difference, and it’s not totally crazy to think that Johnson could win his home state. But for both to occur together is quite a parlay.1 In 20,000 simulations of our polls-only model this morning, cases in which neither Clinton nor Trump received a majority of electoral votes and Johnson received at least one came up just 30 times, putting the chances at 0.15 percent. Most of those did involve Johnson winning New Mexico, sometimes along with Alaska (probably his second-best state).
A somewhat more common deadlock scenario is Trump and Clinton each getting 269 electoral votes without Johnson getting any. The chances of that are about 0.4 percent.