[SpeedyInTexas]

“10/27/20: REPs - 2,383,218, DEMs - 2,685,500, lead of 302,282 for DEMs, 41.8% to 37.0%”

Gap down to 4.8 points.

Was 1.5 points on Election Eve 2016.

Getting there.

[SpeedyInTexas]

D VBM return rate is 68.1%. Yesterday was 66.8% (only increased 1.3 points!)

R VBM return rate is 64.8%

3.3 point gap. Gap is closing.

We are going to estimate D overall turnout at their VBM return rate on Election Eve morning.

Then assume they add 5 points to that number over last 2 days. Then assume they underperform that number by 5 points with all in person voting (IPEV+ED). [net we estimate all in person voting is 10 points under final VBM percent]

With those calculations we can estimate the size of Trump’s victory in Florida.

Go Team Go

[Rumierules]

How much better is Florida EV data for determining how the election is going than other swing states like NC, Ohio, Wisconsin, etc.? I know the tracking of party identification/voting is helpful.

[BohDaThone]

A small caution. IN 2016, as has been noted, by election day the D-R gap was only 96K in ballots cast. However, the actual Clinton margin from all early votes was about 230K -- largely from a Hillary margin among the independent (not registered either R or D) voters.

Trump overcame that with a 350K margin on election day. This year his % margin on election day may be larger, but the total election day vote may be less.

Things are so different this year that it is hard to say, but you should not count on the Biden early vote margin being as small as the R-D gap in registrants who vote early.