I seriously doubt Trump's performance improved in a serious way because of Rogan, polls are just illegitimate and have been for many years
Not sure I'd say polling is illegitimate, and not sure if that's actually what you're saying...
The pollster I follow had this thing called as most likely 312 EV weeks ago. And tore apart the Iowa Selzer poll before it was even released. Based on the things I've learned, I'd say the Rogan interview and endorsement probably didn't make the difference, I expect more or less the same result whether those happened or not. But I think the overall podcast strategy moved the needle some.
On the other hand... Nate Silver is a total hack, and the whole FiveThirtyEight 'high quality poll' ratings thing is a complete and utter scam. I wasn't super surprised to learn that as I had read one of Nate's books and came away... unimpressed with his intellect and supposed mastery of statistics.
Most
media polls are junk, even when they get a more or less good-looking end result chances are it was happenstance, and they got there for the wrong reasons. There's a lot of circle-jerk, herding, bad methodology... and the aforementioned 'high quality' scam which punishes any pollster who produces results that are too far right even if they end up being accurate. Most importantly they're just generally underfunded. It costs a lot of money and effort to actually reach the electorate in the post landline era.
I think both R and D probably had pretty good internal polls and neither was too surprised at the result.
I had good fun scanning the Ree threads every once in a while to see what kind of nonsense they were peddling. The funniest was some get-out-the-vote campaigner in PA who wrote (idiotic) essays about how it was 'impossible' for Trump to win there. Looks like he got banned, must have blown a gasket after the fact.