Michigan lets you vote individual candidates, or a "straight party" option.
Each blue dot is a precinct. The more "straight party" Republican votes in a precinct, the more % of Trump votes were siphoned to Biden.
100% this is a computer algorithm, because the data shows that the % of votes siphoned to Biden increases perfectly linearly, the more straight party Republican votes, across all the precincts.
For example, a precinct that had 85% straight republican votes, had 20% of Trump votes siphoned to Biden - leaving Trump with 65% in that precinct.
Wayne County (Detroit) did NOT show evidence of an algorithm. Only Republican precincts had votes siphoned. It looks like the algorithm didn't kick in unless the precinct had 20% or more straight republican ballots.
It also explains how Trump can do better than last election in the places he lost but somehow do worse Republican districts. People really think Trump somehow turned blue voters but lost people who voted for him last election.
A little more analysis, if you want to understand the graphs in the video.
In the video, fast-forward to 36:30 - these are the results for Kent County. What you see here is:
Precincts that voted....
40% Republican: ~5% Trump siphoned off
50% Republican: ~10% Trump siphoned off
60% Republican: ~15% Trump siphoned off
70% Republican: ~20% Trump siphoned off
80% Republican: ~25% Trump siphoned off
The precincts aren't all precisely that amount, but they follow that linear trend very closely. The variation could be attributed to a little variation programmed into the algo, slight differences in timing of the votes being counted, I'm not a statistician so I'm not sure. But the trend and overall is very clear.
I don't think that's the right take. Assuming that the President Only voters voted at an even 50%, irrespective of straight-line outcome, a downward sloping graph is precisely what you'd expect.
Sure but I don't think all precincts would vote 50%. It would be weird if precincts that voted 20 Rep/80 Dem or 80 Rep/20 Dem to all be 50% T. you'd expect them to be similar
TL;DR
Michigan lets you vote individual candidates, or a "straight party" option. Each blue dot is a precinct. The more "straight party" Republican votes in a precinct, the more % of Trump votes were siphoned to Biden.
100% this is a computer algorithm, because the data shows that the % of votes siphoned to Biden increases perfectly linearly, the more straight party Republican votes, across all the precincts.
For example, a precinct that had 85% straight republican votes, had 20% of Trump votes siphoned to Biden - leaving Trump with 65% in that precinct.
Wayne County (Detroit) did NOT show evidence of an algorithm. Only Republican precincts had votes siphoned. It looks like the algorithm didn't kick in unless the precinct had 20% or more straight republican ballots.
Yup, this algo would spit out a number telling folks EXACTLY how many Biden only tickets they'd need.
Holy shit, they hid the siphon in the districts where people wouldn't be looking.
It also explains how Trump can do better than last election in the places he lost but somehow do worse Republican districts. People really think Trump somehow turned blue voters but lost people who voted for him last election.
A little more analysis, if you want to understand the graphs in the video.
In the video, fast-forward to 36:30 - these are the results for Kent County. What you see here is:
Precincts that voted....
40% Republican: ~5% Trump siphoned off
50% Republican: ~10% Trump siphoned off
60% Republican: ~15% Trump siphoned off
70% Republican: ~20% Trump siphoned off
80% Republican: ~25% Trump siphoned off
The precincts aren't all precisely that amount, but they follow that linear trend very closely. The variation could be attributed to a little variation programmed into the algo, slight differences in timing of the votes being counted, I'm not a statistician so I'm not sure. But the trend and overall is very clear.
Did they run Wayne County looking at Joe Biden’s results? Because if it’s a horizontal line, it is a big f*ckin deal.
Yes. And yes.
I don't think that's the right take. Assuming that the President Only voters voted at an even 50%, irrespective of straight-line outcome, a downward sloping graph is precisely what you'd expect.
Sure but I don't think all precincts would vote 50%. It would be weird if precincts that voted 20 Rep/80 Dem or 80 Rep/20 Dem to all be 50% T. you'd expect them to be similar