I know my biases are obvious, but I'm not convinced they haven't impacted it. There's a growing bag of non poll based evidence here to be cautiously optimistic about
- Early voting historically favours democrats pretty heavily and we're seeing excellent D results so far...for example, here's PA
View attachment 22338
Democrats requested twice as many mail in ballots, and have returned their ballots at a much higher rate already.
- We keep seeing survey/polling answers suggesting that Nikki Haley Republicans have defected from Trump in significant numbers. Every #crosstab I've seen so far seems to expect the % of R's voting Trump to be more or less equal to the % of D's voting Harris. Either the Haley Republicans are lying, or the pollsters aren't cooking this into their math.
- A pretty durable staple of #crosstabs over the last 2 months has been Trump doing much better among young black men. That's maybe the single most unreliable demographic to get right in polling, and the single demographic historically least likely to show up a the polls.
- Actual voting results over the last few months lean heavily Democrat including fucking mayoral races in Alaska of all places. I can't help but think this is a likely blind spot that poll weighting just has wrong when we're seeing it so durably in real world examples, including early mail in voting here in the presidential election
- There is a pretty obvious and concerted effort to flood the zone by R affiliated pollsters. Polls aren't cheap, someone is paying for this and they're doing it for a reason. This isn't money you're spending if you're comfortable with where you think the race is.