If that sort of analysis was at all valid we’d decide elections by statistical sampling or even simply based off party identification.

Dr. Ayyadurai is about as partisan an “academic” as you’ll find, basically the right-wing version of the left-wing academic that conservatives (including myself) often decry. Even outside of politics, many in the tech world are familiar with him as the person who insists that he invented email, despite all historical indicators to the contrary. He is very far from a reliable source, unfortunately. I say all of this not to impugn him as a person, but rather to show that any claim from him should be examined with pretty significant critcality.

In addition, your article is factually wrong. It contains a fabrication about what happened in Antrim county. There was no hand count of ballots. There was a simple mis-tally from subtotals, which was fixed via the audit trail. In addition, you suggest that there was some kind of “glitch” in the software, when in fact what happened is that there was human error in not properly updating the machine software. You can find the full details in Claim 4 of this article I authored. Mischaracterizing what happened on the ground so significantly throws your entire analysis into question.

If I have time I’ll dig further into the actual statistical analysis presented here, but from the start you’re not coming from a reliable base.