Once again, from Twitter:
.
Biden leads 54 to 45 among those that have already voted. For more information on early vote statistics (and for live results on election night) visit https://2020.dataforprogress.org.— Data For Progress (@DataProgress) 10:50 AM – 26 October 2020
In their full sample, there are 452 people who have already voted, and Biden leads among them 54-45, as noted in the tweet. Of the 566 people who have not yet voted but say they will, Trump leads 50-44, with five percent undecided. (Not many third-party supporters in this sample, which is a combination of SMS and web panel.) Since independents are the new hotness, Biden leads among them in the full sample, 45-43.
The previous DFP result was a week ago. At that time, 180 voters from their sample had voted, with Biden leading 57-41 among them. You can make of that what you want. Biden led 47-46 in that poll, with Cornyn leading Hegar 44-41. The main takeaway here is fewer undecideds, and that more of the undecideds are going to Hegar than Cornyn. Indeed, Hegar leads by the same 54-45 among those who have voted, but trails 50-39 with the rest, with 9% undecided. This is the first poll I can think of that suggests she will finish within a point or so of Biden.
Again, we’ll see if this is the end of the polls for this cycle. We sure can’t complain that we were ignored.
My understanding is that DFP has a Democratic lean, so this probably means Trump would be slightly ahead according to most pollsters. However, I have my doubts about the pollsters having good likely voter models if Texas has a turnout over 11 million voters.
I could sit here and say each candidate will get 50% of the vote with + or – 3.4% deviation and by next Wednesday I’d be right also. That’s without doing a poll, congratulations, these polls don’t mean shit.