Meanwhile per Elect Project over 35.5 million people have already voted. The number is actually probably much higher; there is a bit of lag (understandable) between the votes being recorded at the state level and recorded here:
https://electproject.github.io/Early...20G/index.html
It is a bit dangerous to draw too many conclusions even from the party ID of voters, as frequently they don’t track vote choice all that closely. But if polls are any indication a huge proportion of those voting early are democrats this year with a Republican Tsunami expected on Election Day. The question is whether enthusiasm becomes so high that an E-Day comeback is no longer possible.
With that in mind it’s a little bit instructive to look at total turnout as a percentage of the 2016 electorate. In the US overall, that number crossed 25% today. In Texas it’s over 52%, and in Harris County (a district Beto O’Rourke won by 17 points in 2018) the early vote is on pace to exceed the TOTAL votes cast in 2016.
None of which is to say Biden will win Texas, but there are some reasons to think it’s a softer target for him than a place like Ohio, simply because there are actually relatively fewer rural voters in Texas. The rural districts are huge and Trump will win all of them by huge margins, but there are not that many voters there. He won in 2016 partly because he also won suburban districts—if those districts are turning on him, and if turnout soars in urban counties like Harris.... you can see how a narrow Biden victory COULD happen.