Bronough Street, we have a problem.
That problem for Democrats is that Republicans have figured out early voting. The one “ace up the sleeve” for Democrats in Florida has been offset by an aggressive early voting campaign by Republicans. In fact, 7,152 more Republicans have voted early compared to Democrats. When Democrats make small gains in Broward, Miami-Dade, Hillsborough and, yes, Duval Counties, Republicans are making significantly larger early vote gains in Sarasota, Okaloosa, Santa Rosa, and almost every other Republican-leaning county in the state. Remember when I said at the beginning of the election that the Florida Panhandle was close? Not anymore! Escambia is still interesting, but Okaloosa and Santa Rosa Counties are on track to perform better for DeSantis than they did for Trump in 2016.
With this being the case, this weekend’s early vote effort might be the make-or-break moment for Democrats. Usually, Democrats have been able to run up their margins during weekend early voting. However, as of right now, they need to have a stronger-than-usual weekend.
Also, when it comes to “increased turnout”, do not associate that with Democratic performance. All indicators show that the increased turnout is actually because more Republicans, or independents in Republican-leaning areas, are turning out to vote. Therefore, the usual narrative doesn’t always apply.
Turnout Summary: As of this morning, 2,140,758 voters have turned out to vote, nearly approaching 50% of the total turnout in 2014. The overall turnout rate is at 17.63%. Republicans currently have a 21.34% turnout rate while Democrats have a 18.77% turnout rate. Independents have a 11.33% turnout rate. The Enthusiasm Gap is at -5.01%, moving only .03% toward the Republicans since yesterday. However, the rate has been stagnant for a few days now. The lack of mobility in the Enthusiasm Gap helps the Republicans.
Bottom Line: As of this morning, a generic Democrats statewide candidate would receive 48.03% of the vote, while a Republican candidate would receive 48.82%, a .79% gap. This is a .19% increase in support toward the Republicans since yesterday.
Details: As was mentioned yesterday in the “Details” section, we see that the Enthusiasm Gap barely moved compared to the increase in support for Republicans. This is indicating that independents are shifting toward the Republicans for a second day in a row. The way this is determined is by the projected NPA/Minor party voters’ vote choice in the 2016 presidential and US Senate elections*. Simply put, NPA/Minor party voters are turning out in higher numbers in strongly Republican counties, like Sumter and Lee Counties, than in Democratic counties, like Orange and Osceola.
What to watch: Today, I will be updating the electoral map for the projected vote throughout the day, so keep refreshing every hour or two. This is the “make-or-break” time for Democrats. If Democrats don’t pull out to a lead in projected voter…or heck, even registered voters turning out, then Election Day is going to be rough…very rough for Democrats (unless something changes, something I will address on November 5th).
Here is a link to the most recent statewide projection model. (Note: DeSoto and Jefferson County projections will be fixed today. The reason why they have not been updated is because the Supervisor of Elections in both counties have not responded to emails about previous election results. Because of this, the results for DeSoto and Jefferson could be less accurate, since we have to take into account Election Day vote).
*Many ask why I am using the 2016 election to determine vote choice. The reason is simple, the electorate has changed. We are in a post-Trump electorate. In fact, one can make the argument that we are finalizing a realigning of the electorate that has been going on for years, but that many political scientists fail to accept. Also, I feel that we have officially moved away from the 1932 realignment, as our politics is no longer based on the New Deal coalition. One important note, however, is that 2016 is in no way being used to determine voter turnout…that would just be stupid.
The Florida Squeeze 
U r correct. Must be big weekend for Dems (soul’s to the polls etc) but the offset is those Counties now just starting to get counted.
Today it is a +6 to +8 win for GOP (maybe more)
D Congressional prospects sinking
LikeLike
Did you see Target Smart Modeling?
LikeLike
I have not. Do you have a link?
LikeLike
Curious why you have Sarasota as basically a 50/50 split right now. Trump won by 11% or over 24k votes and R’s have outnumbered D’s by almost 6k ballots so far in VBM/EV but you have it basically even. You think NPR’s will tilt that much towards the D candidate?
LikeLike
1. 2016 is a presidential year, turnout is different.
2. Model projection is based on the likelihood of vote choice based on party registration. Party turnout is not static, and changes from election to election.
3. History shows that VBM NPAs/Minors in Sarasota lean Democratic.
4. This is a projection of votes that have already been cast, and does not predict the November 6th result.
LikeLike
Agree with “housebroken dad” here. Turnout IS different but facts are facts, and its 36k Rep and 30k Dem right now. You’d have to be assuming independents are breaking 70/30 to the Dems just to get back to more or less even (assuming all the Rs and Ds vote in line with party registration, which I know doesn’t happen).
Your model is either assuming that, or some signiifcant cross-party vote .. or a bit of both.
LikeLike
My model measures vote choice based on previous results. Simply looking at only voter turnout by party is elementary, as well as flawed because it isn’t measuring vote choice. Again, NPA/Others vote.
And yes, with those who VBMed, the NPA/Others in Sarasota County broke heavily with the Democrats. Extremely heavily, by 80%. However, 62% of NPA/Others in early vote in Sarasota voted for Republicans. Sarasota is an odd county like that. Don’t know why, but it is. No other county has this split. This split happened in 2016 as well. Wish it could be explained.
LikeLike
How do you know whom a person voted for previously given its private and isn’t shared? I’ll stick with the 90-95% Rs and Ds vote their own party.
LikeLike
Okay. Good luck with that. Interested to see how you predict how Liberty County will vote.
And yes, a vote is private. It is called inferring from previous data. That is called political science, my friend 🙂
LikeLike
In 2016 VBM, 32,797 registered Republicans voted. Trump received 35,114 of the 2016 VBM votes. Assuming all Republicans voted for Trump, this leaves 2,317 of Trump’s votes coming from other parties or NPAs. As far as Democrats, only 28,660 voted by mail. However, Hillary received 38,864 votes in 2016 VBM. Assuming all Democrats voted for Hillary that leaves 10,204 votes for Hillary coming from other parties or NPAs.
Since we already assigned all Republican votes to Trump and all Democratic votes to Clinton, we thus assign these votes to the NPA/Minor registered voters who voted by mail, which is a total of 12,521. For Hillary, who received 10,204 of these votes, that is 81.7% of the vote, and Trump with only 18.3%. Thus, simply:
TrmpRes = (Trmp16-RegR)
ClntRes =(Clnt16-RegD)
IndSpltClnt=ClntRes/(ClntRes+TrmpRes)
IndSpltTrmp = 1-ClntRes
Of course, there is more to the model which is the reason there are differences and also allows for third party votes, as well as more accurate projections in North Florida, but this is the basic assumption that the model tests. So yes, the math works. Also, the model does not conflate VBM and EV, and treats them separately.
LikeLike