Usually, gubernatorial races are followed by only the nerdiest election watchers. They often involve weird state-level issues, and the results sometimes don’t make sense unless you really dive deep into the candidates’ personalities involved or the long history of the state’s parties. But that’s not the case in the colorful, lawless swamp known as Florida.
That’s because the Florida gubernatorial race is a window into one possible short-to-medium-term future for national politics—and, oddly enough, a potentially important factor in the battle for Senate control. Current Florida Governor Rick Scott is running for Senate against Bill Nelson.
To replace him in the gubernatorial mansion, Republicans have nominated Ron DeSantis, a former House member who is fully on-board the Trump Train. During the primary, DeSantis put out an ad showing him teaching one of his children to “build the wall” with toy blocks and showed one of his infant child in a “Make America Great Again” onesie. At the same time, DeSantis has racked up a conservative voting record—his DW-NOMINATE score (which estimates ideology on a left-right scale based on how frequently legislators vote together) is well to the right of the average Republican and is pretty close to the far-right end of the spectrum.
Democrats nominated Andrew Gillum, the mayor of Tallahassee. Gillum is a black progressive who was endorsed by Bernie Sanders. He favors Medicare for All as well as abolishing ICE (though he wouldn’t be able to do that as governor) and won a somewhat surprising primary upset against Gwen Graham, a more moderate Democrat.
As others have rightly pointed out, this probably isn’t going to play out like some war between moderates or a clash over esoteric state issues that nobody outside Florida cares about. It’s going to be a clash between a Trump-loving Republican and a very liberal Democrat who isn’t a white man. The closest parallel we have to that (although it’s imperfect in a number of ways) is probably Florida in 2016, when Trump himself ran against Hillary Clinton. That combination produced a racially and culturally polarized result.
This map shows the precinct-level results for 2016 (data from the Harvard Dataverse), when Donald Trump barely edged out Hillary Clinton and won the state’s electoral votes.
And it tells a relatively simple story. Clinton won by a wide margin (almost two-thirds of the two-party vote share) in the Miami metro area. Trump and Clinton ran more evenly in the state’s larger cities (note that the definition of cities I’m using here is of CBSAs, which include outlying areas—more here). And when you hit smaller cities, towns, and rural areas you start to see larger Trump margins. That overall pattern translates into what you see in the map—a few really important blue dots (some of which are major cities) and some large (often less densely populated) red swaths.
Trump won Florida in 2016. According to CAP estimates, Trump’s margin among non-college educated whites was three points wider than Romney’s (Trump and Romney got the same overall percentage of the vote among that group, but Clinton’s vote percentage was three points lower than Obama’s), but his margin among college-educated white voters was seven points lower. Trump’s margin was better than Romney’s among Hispanics and blacks, though that may have been due more to dropoff for Clinton than to an increase for Trump (e.g. Trump won 37.9 percent of the Hispanic vote, Romney won 36.1 percent, Clinton won 57.5 percent and Obama won 63 percent). Those changes basically turned a narrow Obama win in 2012 to a narrow Trump win in 2016.
But polls suggest that Republicans may not always win cultural clashes in this way. According to the RealClearPolitics average, Gillum leads DeSantis by about four points. That’s not an insurmountable lead. DeSantis could catch up, or a polling error could throw the race towards Republicans.
But it is a real advantage. And it’s not hard to imagine a swing state like Florida swinging left with a Republican as unpopular as Trump in the White House. Super-charged turnout among college-educated whites, backlash against Trump among swing voters, high black turnout (motivated both by Gillum’s race and some of DeSantis’s recent remarks) and more could help Gillum.
If Gillum does win, there are going to be two major consequences. First, Democrats from the progressive wing will argue that Gillum’s win is yet another sign that the Democrats should run a very liberal, racially diverse slate of candidates in key races.
And Gillum’s run could, oddly enough, have consequences for the balance of power in the Senate. Some of the same polls that have shown a Gillum lead have also showed a tight race for Florida’s Senate race.
That has led some to ask what a Gillum win would mean for Rick Scott: That is, what type of voter casts their ballot for Gillum and Scott? And are there are enough of them for Scott to win even if Gillum loses?
I see three plausible answers.
The first is Hispanic voters.
Recent Republican presidential candidates don’t have a great history with Florida Hispanics. Romney and Trump only got about one-third of this Latino vote in Florida, and exit polls (which have their methodological issues) suggest that John McCain did only somewhat better. But Hispanic voters in Florida aren’t uniformly Democratic. Carlos Curbelo and Ileana Ros-Lehtenien represent highly Hispanic districts. Moreover, both exit polls and geographic patterns suggest that Hispanics helped Marco Rubio comfortably win a second term in 2016, the same year that Trump barely squeaked by.
I strongly doubt that Scott will be as popular with Latino voters as Curbelo, Ros-Lehtenien, or Rubio. But it’s not crazy for him to try to outperform DeSantis (who, again, put out an ad teaching his child to “build the wall” and hugs Trump at every possible opportunity) among that group.
The second is college-educated white voters.
Florida, like many other states, appears to be home to some college-educated white voters who liked Mitt Romney but voted for Hillary Clinton instead.
A lot of these college-educated voters are going to vote for Nelson as a protest against Trump. But if Scott is trying to identify growth areas—people who he might be able to attract that a Trump-ier candidate like DeSantis might not—he’s probably eyeing the sort of white collar suburbanites that past Republicans had no trouble winning.
The third possible answer is that there aren’t enough Gillum-Scott voters, and that the polls will eventually converge on the fundamentals.
According to the latest SwingSeat forecast, Scott has a roughly 30 percent win probability. That’s not great. SwingSeat is making this prediction by looking at a combination of polling data and “fundamentals” (think of presidential approval, past results in a state, etc.). Essentially, the fundamentals think that this race shouldn’t be close at all—that is, it should look more like Wisconsin, Pennsylvania, Michigan and other swing states where anti-Trump sentiment is helping incumbent Democrats maintain a large lead. But the polls say something different, so a Nelson win isn’t a foregone conclusion.
Moreover, the polls have recently taken a slight turn against Scott. On Tuesday, a Quinnipiac poll showed Nelson ahead by seven and a Marist poll put him ahead by three. It’s possible that these polls are just outliers and that the next few polls will show a tie. But it’s also possible that it won’t, and that Scott and DeSantis will both lose.
Finally, it’s possible that both DeSantis and Scott find their ways to a win and that looking at the differences between their coalitions ends up being more of an academic exercise than anything else.
We don’t yet know how either of these races will turn out. Maybe normal events will push things towards Republicans. Maybe a weird, unexpected event (e.g. the results of an FBI investigation in Tallahassee) will scramble the calculus in Florida in a weird way. But the point is that if Scott or DeSantis looks like they’re in trouble, the other one will have to find a creative way to win.
