By Adam Bennett
For the first time in a long while, US-based liberals could be forgiven for feeling optimistic. President Donald Trump, having looked all but secure in his position under a year ago, is now trailing Joe Biden consistently and clearly in the polls. Crucially, Biden appears to have clawed in front not just nationally, but in several key ‘swing states’ which tend to prove decisive in the US’ electoral system.
Democrats, however, would be wise to keep the champagne bottles corked for the time being. Whilst the phrase ‘the most unpredictable election in a generation’ is becoming overused to the point of cliche, there are a plethora of variables bubbling under 2020’s biggest electoral event.
When commentators describe an election as ‘unpredictable’, it’s generally meant to indicate that the race is guaranteed to be close-run. This time, however, it means that anything from a Republican landslide to a Democratic one can be considered a possibility. Like most things connected to Donald Trump, the reasons for this are complicated.
Always a deciding factor in democratic elections across the globe, the nature of precisely who shows up to vote will again be pivotal this November. This year, however, there are reasons to believe it will be harder to predict than ever before.
Firstly and most obviously is the Coronavirus pandemic. Research indicates that Democrats are more likely than Republicans to behave cautiously due to the threat of Covid, so it may be that more Democrats will choose to stay home. This effect could be particularly pronounced if said Democrats believe Biden is guaranteed a win due to his healthy polling, and therefore that a trip to the polling station is not worth the health risk.
Having said that, there is scant evidence from other recent elections to indicate that Covid-19 will ensure low voter turnout. South Korea saw 66.2% of its electorate vote in its April elections, whilst New Zealanders notched an even more impressive 72% turnout a few weeks ago. Whether Americans will behave any differently is, at this stage, a complete unknown.
Further complicating matters, however, is the President’s public undermining of the postal service as a means of voting. For obvious reasons, voting by mail will have been an attractive choice for voters looking to reduce their exposure to Covid-19. Donald Trump, however, has described universal mail-in voting as a ‘risk to democracy’, and blocked a proposed package to provide the USPS (United States Postal Service) with more money in order to properly handle an expected surge in postal votes. Will US voters choose to abstain from voting rather than place their ballots in a beleaguered postal system? Again, only time will tell.
One thing all current polls seem to agree on is that Joe Biden is more popular than Donald Trump. Winning an election in America, however, is more complicated than an outright popularity contest. Hilary Clinton discovered this to her own loss in 2016 when she failed to defeat Donald Trump despite winning a higher number of total votes cast.
The reason this is such a problem for Democrats is that, in modern times, their voter base is piled up high in cities but much more disperse amongst the rest of the US. Presidents in the US are decided not by the sheer number of votes a candidate wins, but rather by how many points in the ‘Electoral College’ they can secure, with each state giving a certain number of points to its winner. To win said points, a candidate needs to have a higher share of the vote than his or her opponent. Ultimately it’s not votes that win Presidencies – it’s points.
This means that if Joe Biden beats Donald Trump by 51% to 49% in, say, Florida, he will win 29 electoral college points. If he beats Trump by 99% to 1%, he will still win the same 29 points. In this sense, we can start to see how piling up your supporters in one geographic area is bad for a candidate – hence Democrat anxieties despite the polling lead.
The Margin Of Error
The classic caveat to any conclusion drawn from polls is the margin of error (MoE). It’s often said that the polls in 2016 were ‘wrong’, and many people have chosen to disregard them on that basis (including, it must be said, the President himself). But the 2016 polls were only wrong if you disregard the MoE.
Taken as a whole, polls in the week leading up to the 2016 election had Hilary Clinton leading Donald Trump by around 3.5 percentage points. In the event, she won the popular vote by 2.1 percentage points (but lost the election due to how those votes were dispersed geographically, handing the all-important Electoral College to Donald Trump). That means that the polls were out by around 1.1 percentage points – well within the MoE, which will typically sit at around 3% either way, depending on the polling company.
It’s when we start to apply that same MoE to 2020’s state-by-state polling that the Biden camp can start to feel truly nervy. In those all-important swing states, the ones on whose outcome elections are traditionally won and lost, Joe Biden’s lead starts to look a lot less solid.
In Florida, the crucial state in which the outcome was contested so vociferously in 2000’s election, Biden’s lead stands at 2.7%. Within the MoE. In North Carolina, won by Trump in 2016, it’s 2.7% again. Within the MoE. In Arizona, another Trump 2016 win, he trails Biden by 2.8%. Within the – well, you get the idea.
What’s more, it’s far from unheard of for polls to move substantially even just a few weeks out from election day. So somewhere like Pennsylvania, where Biden enjoys a 5.2% lead, is not immune from slipping into the MoE.
All To Play For
The only certainty, then, is uncertainty. In a ‘normal’ election year, there are factors that can dramatically flip a race on its head. In 2020, those factors are both more numerous and more volatile than perhaps ever before. Between now and polling day, the best advice is to assume nothing.
Adam Bennett is the editor-in-chief of The International. You can find him on Twitter here.