Florida - Persuasion or Turnout...or both?

In the never-ending quest to simplify Florida, one of the ongoing debates about winning the state is whether Florida is a state won by winning persuadable voters, or whether it is all about turning out one’s base. 

I remember when I started with Obama, I got a ton of advice – most of it unsolicited (much was helpful), though a significant portion went something like this:  “Steve, nothing matters but I-4…Steve, if you don’t maximize the Jewish vote, you can’t win…Steve, field is dumb, it is an air war state…Steve, TV is dumb, it is a field war state…Steve, you have to do better with absentees...Steve, don't waste money trying to convince Democrats to vote by mail...Steve, you have to watch your floor in North Florida, or you can’t win…Steve, you have to take Obama to Condo X, or you won't win...Steve, you have to pay for bus benches in Miami, or you can’t win.”  You get the point.

Here is the secret – all of it matters.  Florida is neither a persuasion state, or a turnout state.  It is, in my honest opinion, both.  It doesn’t matter if it is a Presidential cycle, or a midterm year, Florida is a state about managing margins, everywhere.

Avid readers of my blog (thank you to all three of you) have read me refer to Florida as a self-correcting scale.   The bases of both parties do a nice job of balancing – or cancelling themselves out, almost regardless of population or demographic shifts.   

Before we go any further – it is important to note that this phenomenon is almost exclusively a result of my party losing vote share among non-Hispanic whites.  If we were winning non-Hispanic whites at a level anywhere near Obama 2008, based on the demographic shifts in Florida, we would be a leaning to likely Democratic state.  At the same time – if Florida wasn’t experiencing demographic changes – and the Republicans weren’t losing share among voters of color – particularly Hispanics, we would be a predictably Republican state.   Functionally, if either party can broaden their own coalition, Florida quickly gets less competitive.

But these two factors have largely cancelled each other out – hence the self-correcting scale.

Let’s review quickly how Democrats and Republicans win Florida. 

Because I am a Democrat, let’s start there.  Democrats earn their votes in a handful of counties, specifically: Leon, Gadsden, Alachua, Hillsborough, Orange, Osceola, Palm Beach, Broward, and Dade.   Winning Democratic candidates typically do a few other things: win Pinellas, win St. Lucie, win a few North Florida counties like Jefferson, maintain reasonable margins counties like in Duval, Sarasota, Volusia, and Seminole.    They also maintain a reasonable floor in North Florida, suburban/exurban counties around I-4, and the Fort Myers media market.

For Republicans, their math is a little different – they win a lot more counties, but by relatively smaller margins.  Their win comes from winning in places like Pinellas and St. Lucie, and running up the score in places the Panhandle, the First Coast, the suburban and exurban counties around I-4, and in southwest Florida. 

I’ve written extensively about this dynamic in Presidential cycles.  You can read my primer on Florida here, or my 2016 debrief here and here, but in short, I would argue there was a lot of misreading of the Obama wins in Florida.  Yes, they were driven by significantly increasing the margins in the Democratic base counties over Kerry, and growing them in 2012.  But here’s the thing – that alone wouldn’t have won the state.  In both 08 and 12, Obama generally kept the margins in check in the GOP counties – and he won the few battleground counties that exist in Florida. 

Take Obama 12 and Clinton 16 – both races decided by a roughly one percent margin.   For all the chatter about a “less than enthusiastic” Democratic base, Clinton won the base Democratic counties by more than Obama did.  Her problem wasn’t turnout.  Her problem was Trump winning the few battleground counties, and setting new records in both share of the vote and actual vote margins in those places where they must run up the score to win, and where we need to keep it in check.  

I can read your mind – “That’s interesting Steve, but this is a midterm cycle, and you know it is different.”

Yes, it is – and no it isn’t.

Yes, it s different because the electorate is smaller, and at least in the last two cycles, been more Republican (a fact impacted by two consecutive midterm waves for the GOP), which was a change from 06, where turnout marginally leaned Democratic (and Dems won 2 statewide races). 

But there are a lot of similarities between the Presidential and midterm cycles.   Both Republicans and Democrats still need to carry their margins in the same counties as they do in Presidential cycles.  While the vote totals are different in individual regions and counties are different, the functional roadmaps for winning isn't.

Rick Scott won two elections by a point, however, the shape of those wins was quite different, and in those differences lies the path to how the Democrats can win in 2018.

In 2010, the Democratic struggles were a creature of three real problems:  Hispanic drop-off from 2008, lower participation among white Democrats particularly in Central Florida, and wave of GOP and GOP-leaning NPA voters who saw voting for the GOP as a way to send a message to President Obama.  From a math standpoint, this led to lower than necessary margins in South and Central Florida base counties.   But here is the thing, Scott ran up some very large margins in parts of the state, Sink kept him in check in many others.  In fact, she kept him in check by more than enough in many GOP counties to have a winning coalition if the Democratic counties had performed well.   But they didn't.   The lesson of Sink:  Florida isn’t alone a persuasion state.

Crist’s math in 2014 was quite different.  Crist ran on a far more progressive platform than Sink, with a fairly robust turnout operation – certainly not the size of Obama, but the largest in midterm cycle history for Florida Democrats, and as a result succeeded to run up the score in the base Democratic counties, winning the three South Florida counties by almost 100,000 more votes than Sink.   He also did well enough in the “Crist counties” – the stretch from Pasco through Sarasota, where his brand is most established, winning those counties by a total of almost 2.5%, where Sink lost them by a half of a point.   

But the floor fell out for him in North Florida.  Despite North Florida shrinking as a percentage of the electorate from 2010 (20%) to 2014 (19%), Crist lost the region by 8% more than Sink did, netting Scott’s margin roughly 107,000 more votes, more than wiping out the gains Crist made in the base Democratic counties (97,000 votes).

One other way of looking at it, Crist won the base Democratic counties by 92,000 more votes than Sink did.  He lost everything else by 95,000 more votes than Sink.   The lesson of Crist, as was also the lesson of Clinton:  Florida isn’t alone a turnout state.

If Clinton has her margins in the base counties, plus Obama’s elsewhere, she wins by a point or two.

If Sink had her math, plus Crist’s margins in the base counties, he wins by about a point. If Crist has his margins, plus Sink’s margins only in North Florida, he wins by almost a point. 

2018 will be different yet.  The Democratic nominee will benefit from an electorate that is more diverse, meaning the base county margins should rise, and I think there is a lot of room for growth in the Orlando urban core.  However, at the same time, they will be unlikely be able to count on some the margins Crist won in his corner of the state, and will have to contend with areas where the GOP population is growing.  The questions aren't as simple as how do we turnout more voters, but also have to include questions like how do we keep Duval looking more like it did for Obama, Clinton, and Sink than it did for Scott in 14 or Rubio?    

For Republicans, they must deal with the fact demographics are changing in a way that helps the Democrats, and that 2018, unlike 2010 and 2014, will almost surely not be a very good Republican year, as we've seen in each of the competitive special and off-cycle elections this year.

I believe that in Scott/Nelson, as well as in the Governor's race, Florida starts this year somewhere around 47-47 -- maybe even 48-48, and we will be fighting over the path to that remaining 150,000 votes or so that a winning candidate will need.   Some of those votes are found by increasing turnout, others won and lost in the persuasion fight.  The candidate who wins in 2018 won't find those votes by getting just one of those things right, they will succeed in building the right answer to a puzzle.  That is just how Florida works these days.