For cases as clear-cut as this motorcycle accident, my wild speculation would be 2-5%. But I think there were way, way more "gray area" cases that all got recorded as COVID deaths, e.g., people with advanced cancer who caught COVID and then died not long after.
> For example, I don't know if my dead great grandmother cast fraudulent votes for Biden. Is this evidence of widespread election fraud?
A better analogy would be if there were solid evidence of a provisional ballot cast in your dead great grandmother's name, and there just wasn't any evidence as to whether they counted it or threw it out.
> Have you ever seen a death certificate form, or been part of filling one out? I think there is also room for misunderstanding and miscommunication based on how these work (which I assume varies by state).
I have not.