Penn’s Program on Opinion Research and Election Studies (PORES) and Robert A. Fox Leadership Program hosted a conversation with Steve Kornacki, national political correspondent for NBC News and MSNBC. Moderated by PORES Director John Lapinski and attended by more than 100 members of the Penn community, Kornacki shared insights on his career path, his takeaways from the 2020 election, and what polling and election nights might look like in the future.
Introduced by Lapinski as the “Go-to data guy,” Kornacki’s fame for his role in MSNBC’s election night coverage reached new heights this year, from the flurry of memes on Twitter, having his office cleaned by Jimmy Fallon, and even being named one of People Magazine’s “Sexiest Men Alive” in 2020, much to his embarrassment, he said during the virtual event.
To kick things off, Kornacki shared his career path thus far, from his start in print media covering state politics in New Jersey, then working on Capitol Hill for Roll Call before joining The New York Observer. During his time covering politics in New York, Kornacki said he would get calls from cable news networks looking for guests to talk about politics, and then, after becoming a dedicated MSNBC contributor in 2012, Kornacki got the chance to play a bigger role on election night, something he’d long been interested in, as the network’s “board guy” in 2014. “I feel lucky in that I developed a sense of confidence that this is the thing I should be doing and is the thing I know how to do,” he said. “I’m very grateful I got that chance and have been able to build on that.”
Lapinski, who knows Kornacki from working at NBC’s Decision Desk, asked what was most surprising about the 2020 election and primary season. Kornacki discussed the memorable, yearlong lead-up to the 2020 Iowa caucus and the frustration at the delays in getting results back that night. And while he wasn’t prepared for such a slow start to the Democratic primary season, what ended up surprising him more was how Joe Biden’s eventual nomination was not immediately clear.
“The thing we’d seen is that the first states winnow the field. You didn’t have to win them, but you needed to be competitive, and if you don’t win you need to show something that inspires confidence, but Biden didn’t do that,” said Kornacki. It wasn’t until the South Carolina primary and the support of African American voters that spurred his campaign forward. “From that point on, it was amazing how the psychology changed,” he said.
For the general election itself, Kornacki said his takeaway from the “slow motion” reveal of votes was how little actually changed. “It was enough of a change that Biden was president, and the demographics of where the pattern changed was clear: suburban metropolitan areas and college-educated white voters,” he said, adding that this was especially true in Philadelphia’s suburbs. “But it was more of a revulsion towards Trump that crossed state lines, and that wasn’t something that only happened in Pennsylvania,” he said.
Kornacki pointed out that a swap of approximately 42,000 votes in Wisconsin, Georgia, and Arizona, would have made the electoral college results 269 to 269. In 2016, he noted, this margin was 77,000 votes in Wisconsin, Michigan, and Georgia. “When I say that little actually changed, the popular vote margin was substantial, but what’s emerged in the last 20 years is a disconnect between those two in terms of how we decide an election,” he said.
Lapinski asked what the 2020 election might indicate for future election nights, and if calling a presidential election by 11 p.m. might ever be possible again. Kornacki said that this all depends on how efficient states can make their election processes, explaining that Florida has systems in place to quickly process mail-in votes. In contrast, this year in Pennsylvania there was a massive increase in mail-in ballots, and no agreement on what to do with them before election day, so there were major delays in getting votes counted. “The thing about elections in this country is that it’s not nationalized; it’s state-by-state, and so the reality is that we will deal with a mishmash, with some states efficient and others not,” he said.
The event was then opened up to questions from students, starting off with how Kornacki prepared for 2020 election “night.” Kornacki told the group that it was “more about caffeine than food” and that he stayed up for two nights straight, but did manage to get some sleep on Friday night since no results would be coming in. “When we got to Saturday morning, if it was going to go a few more days I didn’t know how I was going to do it. Then, at 11:30 a.m. there it was, the checkmark, and I called my boss 10 minutes later and said ‘I’m out of here.’”
For interpreting and communicating new election night data, Kornacki said that the first step is to be aware, and to make the audience aware, of which “buckets” votes are coming from and what types of votes are remaining. For the 2020 election, that meant clarifying whether votes were cast in-person on election day, early in-person voting, or mail-in ballots, since different types of votes would lean towards different candidates. “When I’m trying to communicate stuff on air, I say to people, ‘What are we looking at? Here’s a big batch of votes, were these cast today?’ If so, we will interpret them one way,” he said.
When asked about how to improve polling, Kornacki answered that answering this “billion dollar question” is currently a major area of investment by media organizations but that it might not be solved until Trump’s impact on politics has subsided. Kornacki said that, because polling errors are not widespread, with those from Georgia fairly accurate, it could be a problem that’s specific to how Trump voters interact with pollsters or his overall polarization of the electorate.
For Kornacki, his approach in 2020 to talking about polls was to walk viewers through different scenarios, comparing polling data and actual election results from 2016 to show how projected and actual results could differ. “Whatever happens with polling going forward, you will see a component of that, with polling presented as a range,” he explains. “The number we got is this, but it could be this or this based on different scenarios or how voter turnout might be.”
Reflecting on the “Kornacki Cam” and how it feels to be one of the most recognizable faces on election night, Kornacki said that what stood out for him was how many people were watching and paying such close attention to election results minute-by-minute. “The level of interest, now and the last few years, has been amazing to think about,” he said. “The culture of politics, the culture of elections, the widespread interest—I sometimes have to step back and remind myself because it’s new. It’s great that we have this much interest, and we’ll see where it goes.”
John Lapinski is the Robert A. Fox Leadership Professor of Political Science in the Department of Political Science in the School of Arts & Sciences at the University of Pennsylvania. He is also the faculty director of the Penn Program on Opinion Research and Election Studies and the Fox Leadership Program.