Not sure we'll get more "final" national live-interview polls, but here are my average crosstabs (in situations where pollsters polled twice in Oct., "October 2020" includes the first poll and "FInal 2020 Polls" includes the second). 
Most striking: stability of the race and Biden's lead. But a strange/fascinating twist: Biden's weakest point, September, was also his high-point w/ non-whites.
Since then, Trump has narrowed Biden's lead w/ Black & Hispanic voters, even as Biden has stretched his lead overall.
Since then, Trump has narrowed Biden's lead w/ Black & Hispanic voters, even as Biden has stretched his lead overall.
Maybe the latter trend is noise, maybe it isn't, we'll see in a few days. But two things that are impossible to call noise:
- Biden has made remarkable gains vs. Clinton among 65+ voters, esp. in the final month
- We're headed for a historic gender canyon, not gap
- Biden has made remarkable gains vs. Clinton among 65+ voters, esp. in the final month
- We're headed for a historic gender canyon, not gap
Took the @CookPolitical swingometer for one last spin. Here's what it spits out when you plug in the demographic splits in final live-interview national polls (even assuming a 10% turnout bump among non-college whites and 5% bump for everyone else). cookpolitical.com/swingometer
To be clear, this isn't the Electoral map I'd draw if I were entering an office pool, etc. But if the average of final live-interview national polls were right, this kind of popular vote margin wouldn't be out of the question.
• • •
Missing some Tweet in this thread? You can try to
force a refresh
