Billy Newby - who has notably been retweeted by Robert Barnes - has been compiling data for the four swing states of North Carolina, Florida, Arizona, and Pennsylvania. Here they are,
North Carolina Voter Registration Data:
View attachment 1654269
Link /
Archive
R+343.49 per day (June 20, 2020 to October 10, 2020)
R+455.86 (October 3, 2020 to October 10, 2020)
Florida Voter Registration Data:
View attachment 1654270
Link /
Archive
R+1127.85 (June 30, 2020 to October 9, 2020)
R+1794.56 (September 30, 2020 to October 9, 2020)
Arizona Voter Registration Data:
View attachment 1654271
Link /
Archive
R+102.24 (April 1, 2020 to October 5. 2020)
R+256.08 (August 4, 2020 to October 5, 2020)
Pennsylvania Voter Registration Data:
View attachment 1654272
Link /
Archive
R+804.58 (June 29, 2020 to October 5, 2020)
R+1184.8 (August 31, 2020 to October 5, 2020)
R+961.29 (September 28, 2020 to October 5, 2020)
All states show an increase in GOP registration since at least the second half of this year. With the exception of Arizona, all states have trended Republican since 2016 and even Arizona is closing that gap. In North Carolina, there is a slow, but gradual trend towards Republicans. Florida and Pennsylvania, on the other hand, have an all out assault. Florida had people registering to vote by the thousands until registration closed and Pennsylvania's increases have been nearly as fast. Worth noting that the Florida's increased Republican voter registration just happens to be in areas where most pollsters miss while almost all of Republican's growth in Pennsylvania - except arguably Bucks County - is in Appalachian Country, which except for Allegheny County is blood red.
All except Pennsylvania have the last bracket accelerating growth, but even Republicans outpacing Democrats at 961 people per day is great for a competitive state. My guess is that Trump's announcement that he had the coronavirus might have slightly slowed his lead since people were worried he could be facing life threatening complications or even die from it. If it affected Arizona's GOP growth too or even reverse it, we won't know until the registration deadline on
October 23rd (
archive). But the last bracket of data for both states is on October 5th, the day Trump returned back to the White House after his brief stay at the Walter Reed Medical Center.
This is important because we have the last bracket of data for Florida and North Carolina that extend to the 9th and 10th respectively, days after Trump returned to the White House
and the Vice President Debate. Notice how Republicans accelerated their gains in Florida and North Carolina in their final voter registration days? That tells me that Trump showing that he conquered COVID-19 and Pence's pulverizing Harris at the Vice President debate in front of
sixty million people had an impact on undecided voters and that they don't want to risk a Harris presidency happening with how old Biden will be if he wins. Time will tell if Pennsylvania and Arizona maintain their pace of growth or not.
What I will say is that after looking at this, it is unlikely Democrats will narrow their voter registration gap in Arizona and
highly unlikely the Democrats will increase their margin in Pennsylvania. In the latter, the best the DNC can hope for is to slow GOP's pace until the deadline on the 19th. It's obvious the DNC are hoping that the extended deadline in Arizona will get more people to register due to COVID-19, but I have a hunch that due to recent events, it might do the opposite.
Also, here's another interesting tidbit
in this NBC News article (
archive) made on October 1st. Granted, it has a negative spin on how voter registration isn't enough to make up for the polling gap, but it does acknowledge it could matter if the polls tighten before Election Day or there are substantial polling error (hint hint). Worth noting that Trump is doing even better in voter registration than he did in 2016.
---
Another thing to note,
Baris will be having a much better idea of how Pennsylvania might vote tomorrow (
archive). While he cannot do a show on it on Sunday, watch out for his Twitter to see if he releases any interesting information.
Yesterday he said that the sample was "very male" and that it could be 51/49 (women/men) or 52/48. We'll see if that holds true when the results come out tomorrow.