Ten days

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. 

On Thursday, Trump and Biden met on the debate stage for the final time. The event broached a range of topics, from COVID to race relations to climate change. It was, overall, a significantly more watchable affair than what we witnessed three weeks ago (admittedly a very low bar). This time, Biden was met with fewer interruptions, in part since candidates’ microphones were muted for each opening statement and also because Trump’s advisors likely told him to take it easy. The result was a much sharper Joe Biden than we saw at the first debate and a more civil Donald Trump. 

Opinion polling of debate watchers offers two key takeaways. First, Trump fared notably better than last time, as 39% of voters thought the president won, compared to 28% for debate one (according to CNN). Second, a clear majority still favoured Joe Biden’s performance, whether by 14% as per CNN or 19% according to YouGov. Will any of this matter? Probably not. Ultimately, Trump needed a major breakthrough here and he did not get one. 

Three weeks ago, I wrote that the president was running out of time. Now, with only ten days until voting ends, his position is increasingly dire. Over fifty-four million people have already voted in the 2020 election, almost 40% of the 2016 total. Unless something changes fast, or the polls are off by a large margin, Donald Trump will be a one-term president. Furthermore, the Republican Party at large could be in for a historic drubbing. 

The state of the race 

If this week’s debate is to make a difference in the polls, we won’t know it for a few days. For now, Joe Biden leads Donald Trump by 9.7% in the FiveThirtyEight national polling average, a Trump gain of 0.8% since last week. Despite a minor tightening, Biden’s lead remains the biggest for a challenger (at this late point in the race) since Bill Clinton in 1992. The polling averages in battleground states convey a similar picture, of mostly minor gains for Trump with Biden still clearly in the lead. 

Data: FiveThirtyEight

With so little time left, just how much can the race feasibly change? Below is the Real Clear Politics national average going back to 2004, both ten days out and on election day, alongside the actual result (apologies for using different aggregators but FiveThirtyEight didn’t exist before 2008, and hasn’t published averages before 2016).

Data: RealClearPolitics

As we can see, the average difference between polls at this point in the race and on election day is only 1.1%. Meanwhile, the average polling error between the election day average and the actual result is 1.4%. This means it would be immensely difficult for Trump to close the gap nationally. 

However, Trump does not have to make up all of Biden’s nearly ten-point lead. After all, the polls in battleground states are noticeably closer. At the moment, Biden would survive a 2016-scale polling error in Trump’s direction (the Democrat would still flip Arizona, Michigan, Pennsylvania and Nebraska’s 2nd District, getting him to 280 electoral votes). 

Basically, Republicans need two things to happen over the next ten days: Biden’s battleground lead to be cut by a few more points followed by a 2016-like polling error in Trump’s direction. Or, to put it another way, Biden’s odds of winning the presidential election are probably whatever the odds you assign to him maintaining his lead in Pennsylvania, the most likely ‘tipping point’ state. Indeed, according to the FiveThirtyEight forecast, Trump’s likelihood of winning the election is exactly the same as his odds in Pennsylvania (13%). 

With that being said, if the race does not tighten further, or if the polls are broadly right, or if they are instead off in Biden’s direction, Trump will surely lose. And, in doing so, he will likely drag other Republicans down with him. 

Down the ballot 

Forty-two per cent. That is the current share of the national popular vote expected to go for Donald Trump, as per the FiveThirtyEight average. It is also the same amount expected to support Republican candidates for the House of Representatives. This is not a coincidence. 

In 2018, Democrats regained control over the lower chamber of Congress. That was despite an electoral map which, due to intense partisan gerrymandering, heavily favoured the Republicans. Two years later, they are all but certain to remain in power. While Republicans need a net gain of twenty-one seats in order to flip the chamber, the Democrats are actually expected to increase their majority. The Cook Political Report rates nine Democratic seats are as a ‘toss-up’, alongside seventeen won by Republicans two years ago. 

The congressional map still favours the GOP, but not to the same extent as in 2018. Last year, a court in North Carolina struck down new boundaries proposed by the state’s Republican-controlled legislature (which were themselves drawn up after the previous map was ruled as unconstitutional). The result? Two new congressional districts where the Democrats are clearly favoured. 

This year’s election is expected to see historically low levels of ‘split-ticket’ voting, whereby one backs more than one party (e.g. A Democrat for president and a Republican for Congress). To that effect, where Donald Trump is expected to lose ground, Republicans might well go down with him. Even in states which the president is still likely to win, Biden’s relatively strong performance might help his party’s candidates over the top in their individual districts. In Texas, current polling has Trump down eight points relative to his 2016 margin and six Republican seats are under threat. 

While control of the House is almost certainly a foregone conclusion, Republicans are fighting tooth and nail to defend their majority in the Senate. It is, after all, their most effective check on a potential Biden/Harris administration, capable of blocking legislation as well as judicial nominees. Right now, the party controls fifty-three seats, a working majority of three (in a 50/50 vote, the vice president assumes the role of Senate tie-breaker). 

Assuming Kamala Harris is the vice president come January 21st, Democrats need a net gain of three seats to win control of the upper chamber. In reality, this means flipping four, as Republicans are heavily favoured to defeat Democrat Doug Jones in Alabama. As things stand, Democratic challengers lead in five states: Arizona, Colorado, Iowa, Maine and North Carolina. What do they all have in common? Joe Biden also leads in every one of those states. 

To be clear, Biden is not necessarily the one carrying Democrats to victory. In Arizona, Iowa and North Carolina, the party’s Senate candidates poll better than their presidential nominee. However, in Democratic-leaning states such as Colorado and Maine, a highly divisive president is not what the likes of Cory Gardiner and Susan Collins need. And if Biden has a very good night, Democrats could end up with as many as fifty-five seats, with gains in Georgia, South Carolina, Montana, Texas and Alaska. While this is not a very likely outcome, it is about as likely as Trump winning the election.