Something remarkable has been happening for the past six weeks: Donald Trump is no longer utterly dominating television coverage of the 2016 presidential race.
The timeline below shows the total number of mentions per day of each of the major candidates of both parties on national television networks Aljazeera America, Bloomberg News, CNBC, CNN, Comedy Central, FOX Business, FOX News, LinkTV and MSNBC since June of this year (when Donald Trump formally entered the race). Within a week of entering the race Trump (seen in light blue) utterly dominated television mentions of the candidates. However, starting Sept. 20, in the aftermath of the second Republican debate, Trump’s total daily mentions drop precipitously and have never recovered.
Converting from raw mentions to the percent of all mentions garnered by each candidate, the timeline below shows that since his entrance into the trace Trump accounted for 40-50 percent of all television mentions of candidates from either party. Since October he has accounted for just 20-30 percent of all mentions, with Clinton, Bush, Carson, and Rubio all enjoying greater coverage. Fiorina’s bump in coverage after the second GOP debate has largely faded, while Carson and Rubio both seem to be on the upswing.
The Internet Archive and I also used audio fingerprinting technology to break up the entire debate into soundbites, track how often each soundbite was excerpted across major U.S. domestic television networks over the following 24 hours, and create an interactive visualization of which comments went “viral” on television. You can access the interactive visualizations of the undercard and prime debates yourself and filter by candidate, keyword, channel, and show.
Lindsey Graham was the overwhelming winner of the undercard debate, garnering 77 percent of the excerpts from the debate over the subsequent 24 hours being his statements. His most viral remark, excerpted 23 times, was an attack against Clinton and Sanders, while his second most popular was his “clenched fist or an open hand” remark on how he would work with the Chinese government.
Graham’s upward trajectory in the debates offers an object lesson that debate performance does not necessarily equal poll performance. In the first undercard debate his statements were the third-most excerpted, in the second they were 52 percent of all statements excerpted, and in the third they accounted for even more. Yet, Fox Business announced that due to his poor polling numbers he is being cut from next week’s GOP undercard debate. On the other hand, Santorum and Jindal, whose comments in all three undercard debates have earned little subsequent airplay, will both continue on to the fourth debate.
Though he won the prime debate, Trump continued his decline in viral soundbites, from 32 percent of excerpts of the first debate, to 27 percent in the second, to 19 percent in the third. In fact, Rubio’s responses accounted for almost as many airings, at 17 percent. The four most viral statements were each from a different candidate and all were negative: Cruz’s attack on the moderators (this was later echoed by Christie), Rubio’s attack on Bush, Bush’s attack on Rubio, and Trump’s attack on Kasich.
This time there was also substantial variation in the attention each network paid to each of the candidates. Rubio beat Trump for first place on Al Jazeera, FOX News, MSNBC, PBS affiliates, Telemundo, and Univision. Bush was the most popular on Bloomberg, Huckabee on FOX Business, and Cruz on NBC affiliates.
What can we learn from all of this? Perhaps most significantly, television networks appear to have finally tired of Trump, even as overall coverage of the election as a whole has reached its highest levels of 2015. Even his debate performances appear to be slipping, with his remarks accounting for a steadily decreasing percentage of the excerpts airing in the days after each debate.
Conversely, even as Graham’s edge in the undercard debate has increased dramatically, he has been bumped out of the debate schedule entirely. This suggests that either debate performance is not linked to poll performance, or that the excerpts that television networks choose to focus on from the debates are not indicative of how average Americans see the debates in terms of winners and losers.
Either way, with Trump continuing to fall in both overall mentions and debate excerpts, it leaves open the question of whether we are seeing the end of the Era of Trump or whether winning and losing on television no longer means winning or losing at the polls and we are instead seeing the end of the Era of Television in American politics.
Kalev Leetaru is a senior fellow at the George Washington University Center for Cyber and Homeland Security. He thanks the Internet Archive’s television news archive for the use of its data in this analysis and Roger MacDonald, Trevor Von Stein, Kyung Lee, Jake Johnson and the rest of the television team for all their work in preparing the transcripts and running the fingerprinting analysis.