In 2016 the Sunday before the election I posted a diary here that looked at the Florida Early vote
I am more optimistic today than I was in 2016. Let me explain why.
Let’s start with some basic numbers:
|2016 Net — exit poll
|2020 net — polling
In 2016 there was virtually no difference between their parties in their support among the own
partisans. In addition Trump actually won independents in Florida. This is NOT the case in 2020. The polling is consistent (these numbers are based on over 15 Florida polls. Democrats are more unified that Republicans AND independents are supporting Democrats. Both of these trends mean that simply adding up the ballots by party IS not predictive of tomorrow’s results.
This table describes the EV in Florida as of this morning. In 2016 the Democrats led by roughly 37K, and this year we we lead by about 108K. But that number does NOT reflect the lead the Democrats have banked. In the far right is my estimate of the Democratic lead which uses the 2020 net form polling. I estimate that we currently have a lead of about 414K, of 4.7%. This is substantially higher than my estimate in 2016 (which was about 2%) and is higher than the Obama early vote lead in 2012 (3.9%).
The question though is IS THIS A BIG ENOUGH MARGIN. In 2016 the republicans voted in substantially greater numbers on election day then Democrats. Moever, the Democratic lead in voter registration is lower than it was in 2016.
This attempts to address that question. I predicted the GOP and Democratic by using the turnout figures from 2016 ( 81% for Republicans and 74% for Democrats). I must confess I think this is too large — I do NOT believe that the Republican turnout will be 7% higher than Democratic Turnout. But even if it IS it will not be enough to overcome Democratic advantages among Independents and the fact that Democrats will win more Republicans than vice versa. In this sense 2020 is completely different from 2016.
This is the projection. I estimate that there are 1.68 million votes left in Florida, and Republicans will win election day decisively (by about 350,000 votes). But even with that advantage (and based on a 7% turnout advantage that is not likely) the Democrats STILL win by nearly 200,000 votes.
This is 1.79%, which is about the same as the polling average. T
The keys to this estimate is are the same as the keys to the election:
1. Better performance among Republicans (in part because Biden is doing better among old people)
2. Significantly better performance among the young. I wrote more extensively about that here.