The US Senate race in the key swing state of Florida just got a lot more exciting when Republican Gov. Rick Scott announced that he is running. Scott is not expected to face any major primary competition in his bid to take on Democratic Sen. Bill Nelson.
A win by Scott would make the Democratic Party's chances of taking back the Senate very difficult given that they already need to take two Republican seats and hold all their vulnerable seats in states that heavily went for President Donald Trump in 2016 (Indiana, Missouri, Montana, North Dakota and West Virginia).
So just how worried should Democrats be?
Scott has never lost an election thanks to his ability to self-fund his own campaigns, and he'll be the toughest competition that Nelson has faced since he was elected to the Senate in 2000.
But in a year in which Republicans are swimming upstream because Trump is unpopular, Scott may finally have met his match.
Nelson seems to be holding a lead in higher quality polls taken since the beginning of the year. An average of a University of North Florida and Quinnipiac University poll gives Nelson a 47% to 42% advantage. These polls used live interviews and called cell phones. Importantly, they're also transparent about their methodologies. Polls that don't meet this "gold standard" have the race tied, but over the past three cycles (and in the 2017 special Senate election in Alabama) inclusion of these surveys have actually lessened the predictive power of polling averages. Still, it's noteworthy to point that there does seem to be the split in the polling.
Now, it is true that polling this early is just that: early. You'd rather be ahead than behind at this point, though. When you examine polling since 2006, Nelson's 5 point lead in the early Senate polling (January to June) of an election year translates into winning about 70% of the time. Now that's far from a certain win for Nelson. Scott's not out of it by any means. His chance of winning given the gold standard polling is slightly greater than the chance of flipping a coin twice and having it land on heads both time.
History suggests Scott may have a bit more of an uphill climb than the polling finds at this time. It is extremely difficult to beat Senate incumbents (like Nelson) in a midterm when their party is in the opposition. That is, voters like to punish the party that is in the White House.
When incumbents like Nelson have run for reelection in midterms when their party is not in charge of the White House, they have won an astounding 96% of the time since 1982. Incumbents like Nelson haven't lost a single election in the last three midterms, when the president at the time had approval ratings at or even slightly above Trump's current approval rating nationally.
One factor that Scott does have working in his favor is that Florida tends to be a little more Republican leaning than the rest of the country. Trump won it by a little more than a point, even when Hillary Clinton was taking the national popular vote by about 2 percentage points. Even in states that lean toward the other party in presidential elections, however, the opposition party incumbent has won 93% of the time in midterms since 1982.
A big question though is whether to treat Scott like a normal challenger in trying to evaluate how the race is going to develop. Scott will be able to spend a ton on the race, more people can form an opinion of Scott than Nelson at this time (according to Quinnipiac) and Scott has a higher approval rating than disapproval rating (again according to Quinnipiac). Indeed, an examination of past Senate races finds that governors make some of the strongest challengers to incumbent senators.
If one were to take away the incumbency advantage that we would give to Nelson, it makes a big difference in our perception of the race. While incumbents of the opposition party representing states that lean towards the president's party have won over 90% of the time in midterm elections since 1982, the opposition party (i.e. Nelson in this case) has won only a little more than 60% of the time in races where there are no incumbents. Or put another way, Nelson's chances go down considerably when no candidate holds an incumbency advantage. A race in such a situation would probably best be considered a tossup.
My guess is that Nelson would probably still be considered a small favorite even in this situation given how unpopular Trump is and that he barely won Florida in 2016.
Statistician and FiveThirtyEight contributor Dean Strachan looked at every Senate race from 1990 and 2016 and examined how the following factors best explained outcomes: the presidential approval ratings, the state's partisan lean (as measured by the past presidential vote), incumbency and whether it was a midterm or presidential year election.
If Scott is treated like a normal challenger, Nelson wins more than 95% of the time. That looks a lot like what the basic statistics above also suggested. A Scott win in such a circumstance should rarely happen.
Nelson, however, wins just a little more than 70% of the time against Scott if neither candidate is counted as the incumbent. That's somewhat better than we'd expect Nelson to do looking at all open seats without taking into account the president's approval rating or exactly how much redder Florida is than the nation. That percentage chance is quite similar to what the polls have at this time.
The bottom line is that Nelson should probably be considered a slight favorite even with all of Scott's advantages over a normal challenger, which should bring a smile to Democrats' faces because they probably need to hold onto Florida to win the Senate.
Thanks to his time as governor and checkbook, Scott may be able to throw out the normal playbook when it comes to Senate challengers, which should bring a smile to Republicans' faces.