Go to 37:30 in the video. They address the possibility that there may be a natural bias against Trump as the straight ballot votes increase. But they point out that even if that were true there is no way that the cluster would be so tight and consistent. That is, you’d see much more randomness because there would be many confounding factors like location, affluency, average age, etc. Yet the upper and lower bounds of the plot are unnaturally consistent.
This consistency cannot happen naturally but rather only happen by design
Please let me know if I’m missing something because his presentation seems to be the most likely evidence of fraud, but I seem to be in the minority of that consensus.
He separated the straight down ballot votes from the total votes to calculate the “baseline” for the Xaxis and then the overall total vote for the Yaxis. This wouldn’t necessarily create a negative slope like those tweets suggest.
Pretend 1000 total votes, but 500 of those people just simply selected the option to vote all Republican or vote all democrat. Of those 500 it was a breakdown of 300R and 200D so 60% Republican which he would use on the X axis. The other 500 votes that individually selected candidates should follow a similar trend, so assume 300R and 200D again for 60% Republican. This would be (60%-60%) equaling zero and right on the expected baseline. What his data is claiming is that in a precinct where half the 1000 voters chose to select all one party at an example ratio of 400R votes and 100D votes (80% R) there was somehow a shift in the individual voters in that same precinct that the other half of the thousand voted 200R and 300D (40% R) so the Y axis would be (80%-40%) for a -40% Y axis
That part is incorrect. Check the video at 15:00 for definitions, and then at 19:35 where he says how they are defining the graph. He is computing the Y value by taking the percentage of split-tickets that voted for Trump and subtracting the X value. This post argues that it's not crazy to think the split-ticket percentage is independent of X, which would imply a slope of -1:
No they have been doing this same analysis for 5 years, they know what unaltered elections look like. The main flaw is that they should have shown more 'normal' curves from hand-counted elections, they only mention it but don't show it. The normal ones all look like a random blur.
If you follow the twitter thread, the argument has actually gotten much stronger that the analysis is just wrong. Early part of the thread claims that any sensible model would give a negative slope (which Dr. Shiva incorrectly interprets as a sign that votes are stolen) instead of a horizontal line. Later analysis, based on no model at all, just the assumption that Rep % + Dem % = 100% (no significant amount of third party votes), shows that if you take the exact same set of data and graph it in terms of "Biden votes" instead of "Trump votes" you would get a result showing points clustered around a line with the same slope. A negative slope for Trump means Biden must also have a negative slope, so interpreting a negative slope as evidence that votes were stolen from that candidate makes no sense.
Unfortunately, his analysis seems to be flawed:
https://twitter.com/Clustify/status/1326312800310923265
Go to 37:30 in the video. They address the possibility that there may be a natural bias against Trump as the straight ballot votes increase. But they point out that even if that were true there is no way that the cluster would be so tight and consistent. That is, you’d see much more randomness because there would be many confounding factors like location, affluency, average age, etc. Yet the upper and lower bounds of the plot are unnaturally consistent.
This consistency cannot happen naturally but rather only happen by design
Please let me know if I’m missing something because his presentation seems to be the most likely evidence of fraud, but I seem to be in the minority of that consensus.
He separated the straight down ballot votes from the total votes to calculate the “baseline” for the Xaxis and then the overall total vote for the Yaxis. This wouldn’t necessarily create a negative slope like those tweets suggest.
Pretend 1000 total votes, but 500 of those people just simply selected the option to vote all Republican or vote all democrat. Of those 500 it was a breakdown of 300R and 200D so 60% Republican which he would use on the X axis. The other 500 votes that individually selected candidates should follow a similar trend, so assume 300R and 200D again for 60% Republican. This would be (60%-60%) equaling zero and right on the expected baseline. What his data is claiming is that in a precinct where half the 1000 voters chose to select all one party at an example ratio of 400R votes and 100D votes (80% R) there was somehow a shift in the individual voters in that same precinct that the other half of the thousand voted 200R and 300D (40% R) so the Y axis would be (80%-40%) for a -40% Y axis
That part is incorrect. Check the video at 15:00 for definitions, and then at 19:35 where he says how they are defining the graph. He is computing the Y value by taking the percentage of split-tickets that voted for Trump and subtracting the X value. This post argues that it's not crazy to think the split-ticket percentage is independent of X, which would imply a slope of -1:
https://twitter.com/Clustify/status/1326553412356382723
No they have been doing this same analysis for 5 years, they know what unaltered elections look like. The main flaw is that they should have shown more 'normal' curves from hand-counted elections, they only mention it but don't show it. The normal ones all look like a random blur.
If you follow the twitter thread, the argument has actually gotten much stronger that the analysis is just wrong. Early part of the thread claims that any sensible model would give a negative slope (which Dr. Shiva incorrectly interprets as a sign that votes are stolen) instead of a horizontal line. Later analysis, based on no model at all, just the assumption that Rep % + Dem % = 100% (no significant amount of third party votes), shows that if you take the exact same set of data and graph it in terms of "Biden votes" instead of "Trump votes" you would get a result showing points clustered around a line with the same slope. A negative slope for Trump means Biden must also have a negative slope, so interpreting a negative slope as evidence that votes were stolen from that candidate makes no sense.
https://twitter.com/Clustify/status/1326788638756249600
That is basically the argument that is made elsewhere in the thread: https://twitter.com/Clustify/status/1326553412356382723
It gets better...
https://thedonald.win/p/11Q8Slu4Yw/x/c/1BjFWu2IPo