Nice Job Florida!!! Take a bow!!!

[Awsi Dooger]

34. Can't afford minor mistakes in a state with opponent so energized and organized

The poll models were wrong in Florida. That is part of the angst here and elsewhere. If we properly realized those were coin-flip races then it would still hurt but we wouldn't be claiming theft, or looking to blame.

Instead, the poll models continually assumed greater Democratic enthusiasm and turnout in Florida. That accounted for the Nelson and Gillum leads. But it was wrong. The GOP is incredibly well organized and energized here. As soon as the early votes started rolling in you could see the Republican turnout was greater across the state and specifically in the vital question of how elderly and how white the electorate would be. Young voters increased their participation but not nearly to the degree of other states, or toward necessary level in Florida.

Here is a related segment of Marc Caputo's latest blog post:

https://www.politico.com/newsletters/florida-playbook/2018/11/19/did-brenda-kill-bill-desantis-scott-win-again-fraud-probe-puts-fl-dems-on-d-the-lone-democrat-nikki-fried-348669

"WERE THE POLLS WRONG? — So polls told us that Nelson was marginally leading Scott, and that Gillum was marginally leading DeSantis by a little more. But both Democrats lost. Were the polls wrong? Yes and no. The key word above is “marginal.” In most polls, these leads were within a point or two of a margin of error and either sides’ candidates could win. But still, the toplines for most surveys were wrong to the degree they showed the Democrats more ahead than the Republicans. Why? Modeling. In the end, the story of this Florida midterm was like others: More Republicans turned out than Democrats. But many of the polls had more Democratic turnout than Republican turnout. (One caveat: Some of the random-digit dialing polls ask about party identification, which isn't the same thing as party registration.) Still, if the voter-list polls had better turnout models (which Associated Industries of Florida did), the results might’ve been less of a shocker to many.

WE WERE RIGHT (WE THINK) — One thing the polls collectively did seem to get right were the party breaks for the candidates (i.e., X, Y, and Z percent of Democrats, Republicans and independents were respectively voting for Gillum or DeSantis). As we noted Election Day morning when we took the public polling’s party breaks and averaged them, Gillum and Nelson were on the path to victory if 1) Democrats could hold Republicans below a 2 percentage-point turnout advantage overall and/or if 2) Democrats did not lose Election Day turnout to Republicans by 4 percentage points or more, as had happened in the past. It looks as if history repeated itself on Election Day. We don’t have final numbers, but it looked like an R+4 electorate on Election Day. And so Republicans won the big races."