Election Day is still months away, but the current polling is painting a very bleak picture of President Donald Trump’s odds of reelection.
On Wednesday morning, the New York Times released its latest national poll conducted in connection with Sienna College and it showed presumptive Democratic nominee Joe Biden with a massive 50-36 lead over Trump. As Nate Cohn, the Times’s polling guru, noted in discussion of the poll on Twitter, that lead is so large it’s essentially invulnerable to assumptions about the demographic composition of turnout. Trump in the Times poll has an extremely narrow 1-point lead with white voters, wins the 50-64 age bracket by 1 point, and is actually losing those over 64 by 2 points. With numbers like that, basically any level of youth and nonwhite turnout where Biden enjoys huge advantages would be good enough to put him over the top.
The Times poll is particularly bleak for Trump, and even more worrisome for him, its underlying methodology, which involves “weighting” the sample to party registration and not just demographic factors, is a relatively Trump-friendly approach. And while some of Trump’s other polls are better, none of them are exactly good.
But that doesn’t mean the election is over. If this past spring has taught us anything, it’s that things can change fast.
Trump is way down in national averages
There are different ways you can average polls together, and at the moment, they give similar answers.
Of course, the real election is decided in the Electoral College, but with numbers like these, the Electoral College can’t save you. FiveThirtyEight has the most detailed state-by-state breakdowns, and they show Biden comfortably ahead in Arizona, Wisconsin, Michigan, Pennsylvania, and Florida, augmented by narrow leads in North Carolina, Georgia, and Ohio, and all tied up in Iowa.
Which is just to say that the sheer scale of Biden’s lead suggests a conversation about how many Senate seats Democrats can pick up in the context of a blowout rather than the specifics of the Electoral College.
But it’s important to note that this doesn’t mean Trump’s Electoral College advantage has gone away. And even though 10 points is a huge lead, the race is in some ways closer than it looks.
It wouldn’t take much to put the outcome in doubt
Those who recall the 2016 election have naturally developed some doubts about poll-based predictions of big Trump defeats. So to understand what’s similar today and what’s different, it’s helpful to remember exactly what went wrong in 2016.
Polls overestimated Clinton’s national lead in the popular vote but only by a modest amount. The real error made by many forecasters (though, notably, not by FiveThirtyEight) was to model the odds of state-level polling errors as totally independent from one another.
Consider current polling that says Biden is narrowly favored in North Carolina, Georgia, and Ohio while tied in Iowa. If you treat these as four independent contests, it starts to look extremely unlikely that Trump could sweep all four states. If you toss a fair coin four times, it will only come up tails 6.25 percent of the time. And since Trump’s odds in three of those four states are worse than a coin flip, that translates into even worse odds for Trump. But another way of thinking about it is that if polls are generally overestimating Biden by a bit, then he is probably being overestimated everywhere. So the real question is, “Just how likely is it that Biden is being slightly overestimated?”
Modest polling errors are common, so it would only be moderately surprising to see it happen. Any time a baseball at-bat ends in a hit, it’s a moderately unlikely outcome but not exactly an earth-shattering surprise.
The bad news for Trump is that winning North Carolina, Georgia, Ohio, and Iowa wouldn’t be nearly good enough for him. He needs to win Florida, Arizona, and Pennsylvania to carry the day. And right now, Biden’s leads in those states are big enough that it would take a very large and genuinely rare scale of polling error for that to happen.
The good news for Trump is that his Electoral College edge remains large. In the FiveThirtyEight average, he’s losing nationally by 9.9, but he’s only losing Pennsylvania by 5.8 percentage points. That means that if Trump could cut his national polling deficit down to 5 or so — which could be easily enough achieved by reminding right-of-center voters who are currently undecided that they have fundamental disagreements with Biden on policy — he’d be within “normal polling error” range in the pivotal state.
Even then, he’d be favored to lose, and it’s certainly possible that Trump’s numbers will get worse in the future rather than better (especially if the economy worsens when emergency measures expire in August), but the point is just that his large deficit is hardly insurmountable.
Correction: An earlier version of this story misstated Biden’s lead over Trump in the recent New York Times/Siena College poll. It is 50 to 36.
Support Vox’s explanatory journalism
Every day at Vox, we aim to answer your most important questions and provide you, and our audience around the world, with information that has the power to save lives. Our mission has never been more vital than it is in this moment: to empower you through understanding. Vox’s work is reaching more people than ever, but our distinctive brand of explanatory journalism takes resources — particularly during a pandemic and an economic downturn. Your financial contribution will not constitute a donation, but it will enable our staff to continue to offer free articles, videos, and podcasts at the quality and volume that this moment requires. Please consider making a contribution to Vox today.