Statistically and logistically impossible. These results would not occur without intentionally engineering these results. The test already accounts for geographical difference. Nearer locations have shorter trips and arrive sooner, thus D votes occur with higher frequency earlier, and R votes with higher frequency later.
Yes, also ratios seem very artificial at times. Still analyzing via several methods. Data spikes are odd and batch update ratios to consitant overtime to be natural.
I really want to find similar dataset by county and even district. That would be the holy grail to proove fraud.
You would expect the same trend in other states then. High population areas would have more ballots and take longer to count, but the travel time is way more than the count time, so we don't see that in michigan. 10 people can count to 1000 in only 10 minutes. The fact that detroit took 8 hours to deliver from detroit to detroit means they were doing far more than just counting and putting ballots in boxes to be mailed.
I would think that too, but if I have my facts in order, all those ballots travel the mail system, which effectively randomizes the distribution. They should be ending their journey at a counting center before end of Election Day. The distribution of votes should still be approximately in line with projections; it's the wild deviations that are really suspicious.
I don't think I can speak to that with anything different from what's in the source tweets.
The really wacky points on the left of the graphs are in person votes, those are fine.
Then the mail comes in, shuffled, and it should be generally consistent. The argument is that farther counties arrive to centralized locations later than nearer counties simply due to travel time, and that nearer counties are more Dem, so as time passes within a state, D/R ratio decreases.
The travel time here is from local receipt location to centralized counting location, I believe.
Statistically and logistically impossible. These results would not occur without intentionally engineering these results. The test already accounts for geographical difference. Nearer locations have shorter trips and arrive sooner, thus D votes occur with higher frequency earlier, and R votes with higher frequency later.
Yes, also ratios seem very artificial at times. Still analyzing via several methods. Data spikes are odd and batch update ratios to consitant overtime to be natural.
I really want to find similar dataset by county and even district. That would be the holy grail to proove fraud.
How can we get this?
The news orgs get it somehow. Might be something to talk to local Elections Commission, or local GOP chapter heads.
wouldn't counties with a high D count have more mail-in ballots, therefore taking longer to count and coming in later?
You would expect the same trend in other states then. High population areas would have more ballots and take longer to count, but the travel time is way more than the count time, so we don't see that in michigan. 10 people can count to 1000 in only 10 minutes. The fact that detroit took 8 hours to deliver from detroit to detroit means they were doing far more than just counting and putting ballots in boxes to be mailed.
I would think that too, but if I have my facts in order, all those ballots travel the mail system, which effectively randomizes the distribution. They should be ending their journey at a counting center before end of Election Day. The distribution of votes should still be approximately in line with projections; it's the wild deviations that are really suspicious.
I don't think I can speak to that with anything different from what's in the source tweets.
The really wacky points on the left of the graphs are in person votes, those are fine.
Then the mail comes in, shuffled, and it should be generally consistent. The argument is that farther counties arrive to centralized locations later than nearer counties simply due to travel time, and that nearer counties are more Dem, so as time passes within a state, D/R ratio decreases.
The travel time here is from local receipt location to centralized counting location, I believe.
Curious to hear if I'm missing something