2020

Twitter Got the Democratic Debate Wrong

July 1, 2019
Featured Image
And this is only HALF the field. (Photo by Joe Raedle/Getty Images)

President Trump demonstrated his mastery of the news cycle this weekend by grabbing the spotlight back from the Democratic presidential candidates and all he had to do was step onto North Korean soil to glad hand the world’s most murderous dictator. Total galaxy brain. So many libs owned.

But before we move on totally from the Democratic debates, I want to underscore something structural about the way we actually watch and analyze these things. Because it’s pretty obvious that Twitter got the debate wrong.

If you listened to Twitter last week, you learned the Kamala Harris totally DESTROYED Joe Biden. Her attack on him showed him to be completely unprepared. A paper tiger. Glass jaw. Etc.

Twitter got it 100 percent, totally, completely wrong.

Democracy Corps did a post-debate survey of Democratic voters and found that Biden’s net favorability with African-Americans went up by 18 points.

Read that again: Post-debate, Biden’s net favorability with African-Americans went up by 18 points.

This wasn’t just one poll: A Morning-Consult/538 poll showed that Biden’s total favorability changed barely at all from pre- to post-debate—going from 76.5 percent to 75.6 percent, a drop that’s essentially meaningless. And when voters in that survey were asked what they thought of the candidates’ debate performance, Biden was one of the four candidates in the second debate who got a thumbs up.

The other three candidates from the second debate who Democrats said did well? Kamala Harris, Bernie Sanders, and Pete Buttigieg. Which was exactly the read I had.

The reason I bring this up is not to because I’m brilliant, but because it was obvious watching the debate that Biden, Harris, Sanders, and Mayor Pete were the class of the field. And the only way you could have not understood that is if you were having your judgment manipulated by Twitter.

Podcast
Joe Palazzolo and Michael Rothfeld on Trump's Fixers

In my Friday newsletter (it’s called the Triad; you should subscribe; it’s awesome) I wrote about an experiment I did with Twitter and debates a couple cycles ago:

I ran an experiment on myself where I watched one presidential debate with Twitter open on my computer and one debate with no Twitter.

I was pretty shocked by what happened.

In the Twitter debate, I was reacting in real time and listening to what others were saying and my view of what happened largely conformed to what the general mood seemed to be.

In the non-Twitter debate, when the event was over and I checked in with Twitter, I found that I had a very different view of what had happened than the online consensus.

The difference was so stark that I repeated the experiment a couple times and got the same results.

Here’s what happens: On Twitter, people watching the debate form a tentative thought about what’s happening. And then the incentive structure of Twitter pushes them to double-down on this notion over and over in the most over-the-top manner possible. So that they can harvest all of those sweet likes and re-tweets.

For example, take the first Obama-Romney debate.

Obama came out a little shaky. On Twitter, Democratic partisans got nervous, then Republican partisans sensed blood in the water and an hour into the show you have 100,000 voices on Twitter seeing who can say “OBAMA IS COLLAPSING” in the most colorful way possible with the best gif.

But that was one of the debates I watched without Twitter and I’m here to tell you: Obama didn’t collapse in that debate. Obama was fine. What happened is that Mitt Romney was amazing. Romney had an absolutely electric performance that night. And he was never able to match that again.

You’d never know it from reading Twitter, though.

What Twitter does with a presidential debate is create a consensus view and then turn it into a superlative. Notice that you Twitter rarely ever comes away from one of these debates with two opposing verdicts on the outcome. Somehow just about everyone winds up agreeing.

I would argue that watching a debate with Twitter on distorts your view of what’s actually happening onstage.

This is kind of like the opposite of the Kennedy-Nixon, TV/radio split—because lots more people are watching the debates without Twitter than with it. The difference is that the people watching it with Twitter all work in media. So they wind up writing takes that are odds with reality because they’ve got their Twitter goggles on.

Twitter is a binary system where things are always either The Best or The Worst ever. Always. But the real world doesn’t work like that


Let me give you the obvious non-binary results from last weeks debates: Biden did well and proved that he is not a paper tiger. Kamala Harris did great and showed a ton of raw political talent, living up to the promise her supporters have been holding out for. (Which is why her numbers jumped.) Pete Buttigieg had a successful night and showed that he can play in the big leagues. Bernie Sanders gave his people what they wanted.

In the first debate, Elizabeth Warren dominated. Cory Booker got noticed. And Julian Castro succeeded in getting attention by setting himself on fire. We can argue about who helped themselves the most, but all seven of them ought to be happy. (Even if it only really matters for four of them.)

So next time the Democrats debate (July 30 and 31, in Detroit Rock City) turn your Twitter off while you watch.

You’ll thank me.

Jonathan V. Last

Jonathan V. Last is executive editor of The Bulwark.