you can consult the voter registration for mail-in ballots and the return rates for both democrats and republicans, there are no massive discrepancies, sometimes democrats asked for more ballots, but nothing about organic turnout or voter preference pattern would explain these results.
View attachment 1735172
View attachment 1735173
In any election the totals flatline when you arrange by the size of precint, you don't get a discrepancy
Look at Georgia:
When you arrange by size of precint, there is a DIRECT RELATIONSHIP between Trump and Biden, if biden were to win in a fair election, he should have a graph that follows the patterns where the values flatline, biden with a line higher in the graph (indicating a higher share of the vote) and trump with a flat line lower, when you compare both with Jo, you see there was no tampering on her votes, the pattern respects what an election should show, Trump and Biden however show an extremelly unnatural, obviously manipulated diagonal line that is directly related to the size of precints, the chance of that happening naturally makes this basically impossible.
basically after sampling every marble, you will get an even distribution of marbles to the point they reach an equilibrium indicative of the share of marbles in the bag, half for each, a flat line in the middle of the graph.
Basically you are seeing a scenario when Trump always loses and Biden always wins, and the more votes in a batch, the more trump loses and more biden wins, from a random poll of votes from several districts, this should never happen.