random county population looks like the population of the state This is probably systemic for any county named after the state.
Sadly there is no possible way with XX.X%Trump XX.X%Biden and Total Votes to determine this. There are input lines that subtract Total Votes. I totaled every subtraction for both candidates, adjusted for rounding errors conservatively, and came up with Biden losing more votes that Trump. It is very simple to program it to give this perception and add in additional votes to Biden down the line. There is no way to determine t hat number using this Edison data. Anyone referencing the Edison data is wasting time investigating a dead end.
I checked their work with the provided sources. It was 100% unverifiable. . TLDR, the switch happened to all candidates and even more so to Biden in many states.
Since then, I've been messaged that they believe vote moving happened to everyone and is being used to obfuscate the steal. I've been invited to help out and I will continue to apply scientific methodology to vet the work being done.
The 10k+ upvote post from yesterday seems to have made the same mistake. I looked at the data and i find more subtractions from biden than I do trump. Im going to dig deeper and set up filters for changes greater than .15%. Not having the vote tallies is really making it hard to analyze.
Rock county was 39k Clinton 31k Trump in 2016. 46k Biden to 36k Trump keeps a similar ratio. I wonder if this is either benign or evidence 2016 had this flipping going on as well. It would look really sketchy if Trump won the county easily in 2016 then lost the county badly in 2020.
Need to vet this heavily. I think he may be finding constant ratios caused by long delays where precinct data doesn't change because no new votes were added. His comment near the end about prime numbers is interesting though.