A few days after the riot at the Capitol on Jan. 6, 2021, I picked out a bit of data that seemed as though it offered some insight into how so many supporters of Donald Trump had become convinced that the 2020 election was stolen.
Skip to end of carousel
Sign up for the How to Read This Chart newsletter
Subscribe to How to Read This Chart, a weekly dive into the data behind the news. Each Saturday, national columnist Philip Bump makes and breaks down charts explaining the latest in economics, pop culture, politics and more.
End of carousel
Part of it was Trump’s incessant claims that it had been. Those claims were amplified by his supporters, including those who saw an opportunity to build followers or vacuum up donations. They also were echoed in the broad right-wing media universe, as well, on putative news sites and by commentators who stood to gain the personal benefits that would flow from aligning with Trump’s arguments.
At the time, there was an active Twitter account that grabbed data about Facebook usage from the site Chartbeat, which tracked social media metrics. Called “Facebook’s Top 10,” it showed the points of origin of the 10 most popular link posts (that is, posts with links) on any given day on the social media behemoth.
Between Election Day and Jan. 6, about a third of the 10 most popular link posts were to right-wing sites or activists, including the Trump campaign.
What’s important to understand, though, is that this wasn’t exceptional. It was often the case that right-wing sites such as those run by commentators Ben Shapiro or Dan Bongino saw some of the most popular content. It had been clear since 2016 that social media news about Donald Trump — accurate or exaggerated — fared better than other political content did. The prominence of right-wing voices seemed to reflect that.
Advertisem*nt
But it was also partly because of decisions made by Facebook. Reporting in 2020 suggested that the company’s tweaks to its algorithms, the code that decides what people see, were weighted to advantage right-wing sites — in part because objective efforts to uproot misinformation disproportionately resulted in right-wing sites being demoted. NPR wrote a story in 2021 noting how Shapiro had built a media empire on top of his Facebook success, thanks to the site’s algorithm and thanks to conservative content driving more of the engagement Facebook sought.
Eventually, though, Facebook decided to scale back how much political content was making its way into people’s feeds. The effect on the top 10 link posts was dramatic. In 2021, links to Shapiro’s page made the top 10 list 235 times. In 2022, he was there 175 times. In 2023? Sixteen times. Bongino went from 173 appearances in 2021 to 30 last year.
(The automated Top 10 account, created by the New York Times’s Kevin Roose, stopped updating last June a few months after Twitter changed how publishers could automatically publish to accounts.)
Advertisem*nt
This history is worth revisiting now because of a report from the Atlantic. Written by Washington Post veteran Paul Farhi, it notes that conservative and right-wing media sites have seen huge drops in traffic.
“The flow of traffic to Donald Trump’s most loyal digital-media boosters isn’t just slowing, as in the rest of the industry; it’s utterly collapsing,” Farhi writes. Who’s to blame? “The obvious culprit,” he writes, “is Facebook.”
The British site PressGazette published data looking at how Facebook’s referral traffic — that is, the links from Facebook to other sites — has dropped across the media landscape. The Washington Post has seen referrals drop, as have right-wing sites such as Fox News and the New York Post.
The effects are particularly stark for smaller sites, according to the PressGazette’s analysis. The smallest sites in its analysis now saw Facebook referral traffic at only 2 percent of the level seen in March 2018.
That analysis didn’t include American right-wing commentary sites, but it’s safe to assume that the effects are similar.
Advertisem*nt
Farhi’s report prompted some musing about whether the effects of Facebook’s shifts were generally a function of Facebook’s audience. After all, Facebook is understood to be more popular among older Americans than other social media sites, and less popular among younger Americans. Older Americans, as you likely know, are more likely to be Republican.
Analysis from Pew Research Center, though, suggests that Democrats and Republicans are about as equally likely to use Facebook. In part because older Americans use social media less than younger Americans in general, they are also less likely to report using Facebook. In analysis from Pew published this week, Republicans were more likely than Democrats to say that politics doesn’t belong on Facebook.
That broader data, measured after Facebook’s shifts, probably masks that a subset of conservative Americans were very active on Facebook in and around 2020 and very actively engaged in political commentary. That’s obvious from measures like the Top 10 account.
Again, it isn’t only right-wing sites that saw traffic evaporate after Facebook tweaked its algorithm. But it is likely the case that some of those sites were more reliant on traffic from Facebook and that the decline in referrals had a disproportionate effect.
What that means for information- (and misinformation-) sharing in the 2024 election remains to be seen.
