Staring down the known unknowns. Trump v Biden 2020

Thought Leadership

On the face of it, the social media performances of Donald Trump and Joe Biden point to a US 2020 presidential election victory for the incumbent president. As I argued during the 2016 election, social media allows us to observe millions of people interacting with the candidates, beyond the confines of traditional polling of a thousand or so individuals.

As we know, polling was spectacularly wrong about Trump in 2016, and social media trends closer to the truth. And if we look at social media data now, in Trump v Biden, that same logic indicates a crushing landslide for POTUS 45.

However, there are two major known unknowns this time round: COVID-19 and mass postal voting. Both of these factors lie outside established knowledge, making predictability a thing of the past.

Let’s look at the current state of social media play.

Trump has by far the larger audience: 137 million followers across Facebook, Instagram, Twitter and YouTube; vs 15 million for Biden. Moreover, Trump posts more often: 8.9 thousand times in the year-to-date, vs 4.4 thousand times for Biden. And Trump has gained an astonishing 1.10 billion social media engagements (e.g. likes, comments, shares) from those posts, compared to 1.93 million for Biden.

Of course, a social media engagement can be negative, such as trolling or angry commentary. But again, Trump is far ahead of Biden on explicitly positive engagements (e.g. likes, loves or positive reactions): 882 million for the former; 156 million for the latter.

And lastly, social media users shared Trump’s posts 108 million times, vs 22.4 million for Biden, with sharing often seen as a proxy for advocacy.

In short, on social media, Trump talks to more people more often than his opponent, and those people are vastly more engaged and enthusiastic. We saw the same imbalance in social media performances during Trump v Clinton in 2016, favouring the former, in contrast to almost all the polls.

And compared to 2016, Trump’s social media figures have increased across the board. Explicitly positive reactions, for example, have more than doubled since the last election across a comparable timeframe.

Before turning to the unknowable impacts of COVID-19 and postal voting, it’s worth noting some instructive findings.

First, whatever you think about Trump, his social media positioning is laser-focussed and consistent with his general persona. Trump’s most engaged-with posts are attacks on Joe Biden’s mental fitness; Biden and Kamala Harris’ ‘lies to protect the radical leftist mob’; the designation of ANTIFA as a terrorist organisation; and on the positive side: a message of love for his deceased brother Robert; a happy birthday message for his wife Melania; and a simple posting of the American flag.

So, Trump generates strong support through either attacks on ‘the enemy’ (including news media, Democrats, ANTIFA, Black Lives Matter, protestors, radicals and ‘the China virus’); and through praise and love for what you might call ‘flag and family’.

Trump’s sometimes confusing and contradictory statements, and his coarse humour, are able to sit on top of this essentially binary framework: you are either with me or against me. This simple communication structure, put to hard and often flamboyant use, has been internalised by millions of people, many of whom vociferously support Trump and attack his enemies.

In a phrase: it’s working very well.

Second, a look at Biden’s most engaging social media posts paints an alternative worldview. Instead of Trump’s ‘us versus them’, Biden’s most effective social media communication is ‘us for each other’. It’s less about Biden himself: the most engaging posts call out his choice of VP Kamala Harris; his continuing close relationship and debt to Barack Obama; and his blessing of US civil rights leader and statesman John Lewis at his funeral.

Of course, we are still talking about American politics; attacks on the opponent are par for the course. Biden denigrates Trump on his handling of COVID-19; mocks some of his statements and even makes a good joke: You won’t have to worry about my tweets when I’m president.

But other than Trump, Biden doesn’t really attack anyone: not Republicans, not protestors, not particular communities or organisations. Biden called himself an ally of the light in his acceptance of the Democratic Party’s nomination for President, and his communication is often a testament to that ambition.

Trump divides; Biden thanks. Trump is Good vs Evil; Biden is one for all; all for one. Trump is America first; Biden is America together. Both candidates resort to deeply familiar frameworks, but the bigger picture shows that anger and division are winning the day. At least up to now.

All of what I’ve said is based on observable fact. What we don’t know is how COVID-19, a formidable foe to Trump’s mere words, will develop in the American voter mindset; nor what mass postal voting will mean for a Republican President.

The pandemic is particularly virulent in America, and the impact of it on the sitting Administration hasn’t been properly measured. Opinion polling shows various degrees of blame being attributed to Trump, but a terrible indictment of his Presidency could be coming at the ballot box, particularly if the virus continues to spread.

Nationwide postal voting is another as-of-yet untested variable; its impact can only be observed in real time, when it actually happens.

Trump fears the postal vote, and attacks it frequently, as seen in his opening statements to his acceptance speech for the RNC Presidential nomination. This fear is presumably based on the assumption, one that Biden acknowledges openly, that postal voting means millions of voters from poor, disadvantaged or minority backgrounds will cast their ballots, many for the first time ever.

In other words, Trump is tacitly acknowledging that Biden speaks for those communities, and is more likely to gain their support.

The big unknowns about postal voting are whether Trump will succeed before the election to break the US postal service (and his efforts are certainly bent that way); whether it will actually result in an increased electorate; and whether that larger electorate sways Democrat.

Those considerations, and the unpredictability of COVID-19, mean the outcome of the 2020 US Presidential election is exceptionally hard to call. The data seem to herald a Trump victory all over again, but I sense that untested variables and circumstances-yet-to-come will have an outsize impact this time.


This article was written André van Loon, Senior Research & Insight Director at We Are Social in London.