Summary

• Trump has won in the head to head battle between his personal Facebook page and Biden‘s across all engagement metrics. This is seen across total engagement and when averaging engagement to remove Trump’s big advantage from increased levels of posting.
• Biden has gone backwards compared to Clinton, and Trump is only matching his 2016 performance across some parameters.
• Biden’s campaign has seen average shares actually decrease as the election drew closer.
• In 2016 Clinton got more video views than Trump, currently Biden is nowhere in comparison to either 2016 or Trump in 2020.
• Trump’s campaign has seen average numbers of video views decrease as the election drew closer but seen all other engagement forms increase.
• Overall the data shows that Facebook is Trump dominated this election, a factor that wasn’t the case in 2016 as Clinton performed well.

All data gathered via @crowdtangle, analysis undertaken by myself.

---

Ridin’ with Biden or on the Trump train

I believe organic campaign engagement is important. It shows general campaign health and offers a glimpse into the activity levels of supporters and how they are working to get the message out. This report’s data thus shows that on Facebook it was Trump who has more successfully energised his base. Trump has managed to get his Facebook virtual members to spread his messages out to their friends and family more. However, as the data later shows, this base may well be small, permanently engaged and not as influential in getting their subsequent Facebook networks to engage than one would expect. Thus, although Trump may be activating a key base, does it matter? At the moment we cannot tell, but if we wake up with President Trump again, given how polls have failed to show clear Trump supremacy, organic engagement will be part of that story of victory as the data shows Trump ahead.

The data presents some serious questions when it comes to organic campaigning. We still know so little about how campaigns use social media, especially as an organic viral tool. It is amazing that across the one-month campaign period, precisely when you would think engagement would skyrocket, we do not see that occur for Biden in shares. Something is up with the Democrat campaign on Facebook for this to be a fact.

The engagement battle from 2016 to 2020 and between the two parties candidates

Comparing 2016 to 2020 shows clear trends, in some ways the Republicans have progressed whilst in others they are static. In contrast Biden has failed to reach Clinton’s levels of engagement.

Graph 1 shows that in engagement the trend across the two elections is not all Trump supremacy. In fact, when averaging the engagement in Graph 2 (to essentially get engagement per post) you see that compared to 2016, Trump has gone backwards in some aspects (centrally in likes). In contrast, compared to 2016, Trump’s average shares have increased. Shares are more powerful, but given the nature of a like versus a share (especially the higher interest and agreement threshold needed for someone to actually click share), are we seeing a small core Trump supporting base more enthused, but a wider support network less keen to engage?

Given the less powerful act of liking has seen a drop off, this potential for Trump’s Facebook campaign to be an echo chamber should worry the Republicans. Although the campaigns’ virtual members may be active, it is possible that the average public are not listening to fired up Trump supporters. These trends will only become clearer with further analysis.

Although the data for Trump is not brilliant it is still strong, this is in contrast to Biden’s campaign. Biden has has failed to improve upon Clinton’s 2016 performance across any measure, with likes and shares most notably down. His campaign has failed to energise virtual supporters, with this potentially due to reduced numbers of Democrat activists on the Facebook platform.

However, ignore Facebook at your peril.

Examining total video views in Graph 3 we see Trump domination in 2020. Biden has serious issues in gathering organic views, his supporters are either not sharing enough, people are not interested or his supporters networks are tiny. Whatever the reason, it is not a good sign. This Trump domination is actually only a 2020 phenomenon as Clinton did very well in 2016, actually getting more video views for her page than Donald Trump. In contrast, Biden has seen video views at a quarter of Clinton’s 2016 figures.

Averaging the data to account for posting frequency as in Graph 4 does not help the situation for Biden, he is well behind Trump and Clinton. Trump has outperforming his 2016 video views, but the real story here is Clinton’s impressive video reach and success in getting her messages spread across the Facebook network.

Pre campaign period vs. one-month campaign period – is there change?

To understand more clearly what has been occurring across the campaign, including more recent trends, engagement and video views are compared across two periods. Firstly we examine the data across the pre-traditional campaign period from the 26/6/2020 to the 4/10/2010, and a one-month (~) campaign period from the 5/10/2020 to the 1/11/2020. The choice of these dates is due the previous analysis I had undertaken for this Conversation article.

Firstly examining average video views in Graph 5, we see that there has been no boost in video views for Trump and only a tiny increase for Biden. People seem to not be flocking to engage with the two leader campaigns as the election draws closer. This is fascinating and suggests that only a small core of consistently active supporters are engaging with these pages online.

The data also shows some remarkable stability in engagement as in Graph 6. There has been a boost from the election drawing nearer but it is not colossal. In fact there are even negative trends. For example From the 26/06/2020 to the 04/10/2020 Biden’s Facebook page was averaging 4,585 shares per post, but from the 5/10/2020 to the 1/11/2020 he only averaged 4132 (a decline of 11%) – see Graph 7.

In contrast we have seen Trump’s engagement grow across the board as the election draws closer. He gained 33% more shares on average across the latter period covering October, a surprising figure given his page’s lack of increased video views. This trend in shares is therefore likely seen for non-video content, or that those sharing are being somewhat ignored. Nevertheless, the data is clear, Trumps virtual members are the more engaged and are activating at greater rates as we got closer to election day. In contrast Biden’s audience although also increasing in engagement has shown reduced levels of enthusiasm, especially when it comes to sharing his content.

This reduction in shares for the Biden campaign should worry the Democrats. If you cannot engage your core supporters online to greater activity just before an election, then this shows something is wrong. They managed it in 2016 with Clinton. So what has changed? We will see. This same factor should also worry the Republicans, as although a small core are clearly more keen to share, are the wider community interested? Both leaders are showing signs of voter apathy and fatigue.

Perhaps the real story is not of one of great enthusiasm for either candidate, but actually the classic story of this election being the choice of a lesser of two evils.

Finally, in terms of content approach, we see in Graphs 8 and 9 that 2020 approach has shifted dramatically from 2016 for both parties. Video has arisen to be a core part of their Facebook campaigns.

Examining the two campaign periods, it is clear that as we got closer to the election Biden increased his use of video, while the Trump campaign actually reduced the form, instead favouring photo and links. This somewhat explains the declining video views we saw from the Trump campaign, but also suggests a changing tactical approach for pre-election day.

Does this matter?

The data suggests the Trump campaign have been more successful at activating their supporters online. I will return to this conclusion post result.