What about the slow D trend for mail ins? could that be that the same 4 Nov spike got spreaded over more days? Because one would expect that the very last to arrive ballots should trend R the same as rural areas
I just know that "assume a spherical cow" analyzes - ones which assume everything is uniform - are suspect from the get-go. My suspicion was further supported when the model failed to include well-described factors which I think would reasonable cast doubt on such a simple model. And the commentary failed to discuss why those factors were rejected, preferring instead to jump right to a claim of fraudulent voting, which lead me to conclude the author was even less informed than I.
A recurrent problem in data^Wscience is that it's easy to find false signals when trawling through data. As Feynman commented: "The first principle is that you must not fool yourself and you are the easiest person to fool."
What about the slow D trend for mail ins? could that be that the same 4 Nov spike got spreaded over more days? Because one would expect that the very last to arrive ballots should trend R the same as rural areas