[Political Junkie Too]

That's what I figured. And there's nothing wrong with that, if the polls are accurate.

I should clarify, if you haven't already seen it in other posts of mine.

The model is probabilistic. It uses the margin of error to convert the polls into probabilities of the leader actually winning. The probabilities then go into a Monte Carlo simulation to produce the final results.

A few weeks ago, I did a sensitivity test on bias in the polls by running it through Solver to find the give/take that makes the election a tie in the model. Actually, you saw this already in this post from October 6. At the time, a 2.1% swing from Biden to Trump tied the election. A 3.6% swing reproduced 2016 results.

I should rerun this analysis with today's polls, but the point is still this: if you believe the polls are off by more than 3%, then Trump is winning.

-PJ

[Coop]

Ah yes, I do recall that post now, thank you. I am obviously not using models, polls (for the most part) and statistics. But based on the factors I identified earlier I expect Trump to exceed his 2016 performance.