Biden’s to lose
Welcome back to Bournbrook’s regular series on the state of the US presidential race, your impartial guide to the campaign and what you can expect as election date draws near.
With only three days until voting ends, today marks the penultimate instalment of US Election Watch. On Tuesday, you can expect a final overview of the race, where the candidates stand in each of the battleground states and what to look out for as results come in. To further help you navigate your way through election night, Bournbrook will be hosting a live-blog featuring our reactions and analysis. Today, however, I want to cut through the noise and look at the big picture. To that effect, what do we know about the race at this late stage?
We know that Donald Trump is likely to lose. There are plenty of competing narratives, with people on both sides of the partisan divide desperate to tell you that this isn’t over. That may well be true, but here are the facts.
The president is far behind, as Joe Biden leads nationally by about nine percentage points. Biden’s is the largest lead of any candidate since 1996 and the largest for a challenger on record. In terms of state polling, the Democrat is ahead in enough places to secure 357 electoral college votes, a victory comparable to Barack Obama’s 2008 landslide.
We know that there is effectively no time left for Trump to mount a comeback. We also know that, despite an eventful election year, the polling has proved remarkably stable. In 2016, the president managed to defy expectations due in part to a late shift among undecided voters. Not only are far fewer undecided this time around, but over half of voters have already cast their ballot. As of this morning, national turnout is at sixty-three per cent of the 2016 total. It is even higher in a number of battleground states. Arizona, Florida, Georgia and North Carolina (among others) are at over eighty per cent of 2016 turnout. Texas, meanwhile, at over one hundred per cent. We know that most democratic votes are already in the bag, that there is no reason to believe that Trump will win over the few remaining undecideds, or that it would save him even if he did.
We know that the president is unpopular. We know that no one with his approval rating has ever been re-elected. Joe Biden, meanwhile, is not just more popular than Trump, or less despised than Hillary Clinton. A clear majority of Americans view him favourably, a rare case for a politician in our age. We also know that a clear majority disapprove of Trump’s handling of the coronavirus, which a plurality of voters considers their most important issue. There is no getting around it. Whether you look at the national polling, the state polling or even the congressional district level polling, it all looks terrible for the president.
We know that Trump’s remaining hope of victory lies in a systematic polling error larger than we saw in 2016. According to analysis by Nate Cohn at the New York Times, a 2016 size polling error in Trump’s direction would still leave Biden with 335 electoral votes. To win, Trump would need polls to be off by at least one extra percentage point in Pennsylvania, by less than one per cent in Florida and Georgia and by at least another 2 points in Arizona. Again, this is on top of a 2016 magnitude error in the president’s direction.
We know that the polls will be wrong, in the sense that they are never perfect. If they were, we wouldn’t need to hold elections. What we don’t know is which candidate they are currently overconfident in. We are always fighting the last war in politics. Since the polls underestimated Trump in 2016, most assume they will likely again in 2020. But there is no guarantee of that.
Consider that most pollsters have made adjustments to their methodology to account for 2016 (in particular, many more now weight by education). They know full well the damage that would follow to their credibility should they miss another Trump upset, and could even overcompensate too much as a result. To that effect, if the polls underestimate Biden to the same extent as they did Obama in 2012, we will see the biggest electoral landslide since 1988 (and the largest for a Democrat since 1964).
We know that this is a very different race than Trump vs Clinton. Back then, the Democrat was awfully unpopular and widely distrusted. Not so this time. In 2016, Trump was an outsider and a change candidate. In 2020, he is an unpopular incumbent. Four years ago, Trump won due to support from independents, white women and voter over the age of 65, a coalition that has now turned against him. Last time, major news events tended to favour Trump (e.g. Wikileaks, the Comey letter and the Clinton email scandal). This time, they have turned Americans against him (e.g. civil unrest and the coronavirus).
We should know better than to discount Trump completely. At the same time, with three days left to go, this thing is clearly Joe Biden’s to lose.
