Here is the Michigan Trump+1.8 (rounding up to Trump+2) poll you're referring to. The poll actually came a few days
after the mid-October one that showed up on Realclearpolitics. Oddly enough, the site has not bothered to get around to this poll. James is even further ahead of Peters, at 2.2. All of this was before the first debate. Personally I'm skeptical that Trump is that far ahead in Michigan - I think he's in a dead heat with Biden as of right now and it's possible Trafalgar just reached a different group - but it lines with up him being competitive in that state.
Cahaly needs to poll Minnesota again. In August, Trump and Biden were virtually tied, but so much has changed in the past couple of months and Trump getting COVID-19 likely bumped Biden up a few points there like Baris argued happened in Pennsylvania. That said, Trump getting COVID-19 was the floor and barring a horrible revelation that exceeds Access Hollywood in controversy, he can only go up from here.
It's actually getting people to endorse the Republican Senator challenger in New Mexico? If so, holy moley. Not that I think the person will win, but I have no doubt New Mexico will trend a closer race (but still one that Biden wins) than before. If oil is important in Colorado too, then that also sets the stage for it and New Mexico being competitive in 2024 and 2028 if Trump wins and Republicans keep populism as its platform. Regardless, this was a serious mistake for Biden. He could have written off his Abe Lincoln comment, the corruption, and maybe even being called out in his miserable failure of a crime bill in 1994, but there is no way he can walk back his "ban oil!" comment. Trump could get a double digit win in Texas again just for that and growing Hispanic support.
Speaking of pollsters worth looking into, worth noting that Rasmussen Reports had Clinton at a
two point lead nationally the day before the election (
archive) and turned out to be right on the money. Compare that to 2020. Last week had Biden five points ahead and Baris pointed out in one of his videos that he was approaching the danger zone, which is being four points ahead or less (which according to him makes a Trump win possible). He is now only
leading by three points (
archive) and this was before he got eviscerated on the debate Thursday. Next Wednesday will be the last opportunity we get results from Rasmussen and it will be one way to tell whenever or not undecided voters are breaking for Trump or not.
Barring something shocking or Pennsylvanian Republicans somehow getting a 5-4 majority to overturn the three day ballot counting extension before Election Day, I don't see anything else changing the race. Regardless of the outcome, the outliers will once again be closer to the actual results than most media/university pollsters.