This is a huge oversimplification and I think misses the point and the elegance of his idea somewhat.
From what I understood of his video, the democrats would have to have known what the real polling data was suggesting would happen in a natural election. Let's say it was 52:48 for Trump. They wanted to make sure to rig the election to make sure they actually got a 55:45 Biden win (for instance) since this is a safe margin of victory that's not too outrageous but large enough that it defies credulity to challenge it.
Let's say that they expected 1 million voters in the county they're looking at. They would leave about 70% of those votes untouched in their natural state to make things look natural. Then for the remaining 30%, they would simply substitute the real voting results with fake ones to make up the margin that they need to win by the desired amount. In my example, the 70% of votes that are untouched would be 364K T, 336K B. To win with a ratio of 55-45 with 1 million total votes, the remaining 300K votes would have to have NO MORE than 86K votes for Trump or 28% of the votes. The remaining 72% were would mostly be given to Biden, but some of them (calculated by a bounded random number generator would be given to 3rd party candidates.)
They couldn't just throw in 216,000 Biden votes into one precinct because it would be super obvious, and they couldn't just sprinkle them on the top of a bunch of different precincts because it would not guarantee them a victory and would put them at risk of overvoting on some precincts of other problems that could get them caught.
So they used this wheel method to come up with different ratios: 1/10, 4/65, some even favoring Trump. A whole bunch of them and mathematically calculated how many votes were needed in each bucket ratio to balance the votes across all of these ratios to GUARANTEE that the final result would be the ratio (72% Biden, 28% Trump) they want once everything was compiled. The guy in this video calls these "virtual precincts"
Then they needed to spread these votes out amongst the REAL precincts. They couldn't just assign fixed ratios permanently to one precinct because 1) it would be too obvious, and 2) a lot of the precincts didn't contain enough votes to make up the numbers. So they targeted mostly low turnout precincts (because they'd need to stuff real ballots in these actual precincts to make the results auditable. Might explain why there are SO MANY Biden only ballots, as well as the abnormally high turnout in some precincts) and added votes at these different fixed ratios in an algorithmically determined pattern. Once one precinct's maximum vote potential was nearly exhausted, the algorithm would stop in that precinct and pop up in another precinct, and another, and another, until that vote ratio hit it's "quota" to be balanced against all of the other ratios. That's why he can observe this repeatable pattern of a county reporting a certain ratio then falling off that ratio just before another county picks up that same ratio again.
The wheel number he calculated was something like 2200, which means that the largest common denominator of the fractional ratios that were applied was that wheel number, with dozens or hundreds of ratios balanced around their end goal result (which in Georgia was I think 14.6%).
I hope that this summary makes some degree of sense. Let me know if it sounds like I'm speaking crazy talk. It's late and this is complicated, lol.
Brilliant summary. But the key point for me is the actual 'proof' that what he's explaining is true and not some fantasy of a crazy maths guy. I hope I understood this right:
He identified the virtual precincts, the ones which were hijacked by the algorithm, and isolated all of the reporting during the hijack period. He found the average vote yotal was 14.65% Trump. As each leaf in the wheel is weighted the same, he found that he could substitute the reported numbers with a random number of 40,870 for every precinct, an it still gave the Trump percentage of 14.65.
He put the chance of that happening naturally at 1 in 60 billion.
I think the 1 in 60 billion calculation was off by a factor of 10 because he said "I'll give this a 2% (0.02) margin of error but then actually entered in 20% (0.2) but yeah, the point still stands, lol.
And yeah, he showed that no matter how many total votes there were in all of the affected precincts, as long as the Trump votes followed the ratios that he observed, the overall total % of Trump votes for the state would ALWAYS be 14.6% plus or minus the rounding error (which was also fixed AGAINST Trump by what he called the Snapping Effect)
I saw that too! That's when I stopped watching. I believe his results, but I don't think he did a good enough job with the delivery. I am considering making a 5 minute version of his results that is polished, to the point, and geared for normies.
This is a huge oversimplification and I think misses the point and the elegance of his idea somewhat.
From what I understood of his video, the democrats would have to have known what the real polling data was suggesting would happen in a natural election. Let's say it was 52:48 for Trump. They wanted to make sure to rig the election to make sure they actually got a 55:45 Biden win (for instance) since this is a safe margin of victory that's not too outrageous but large enough that it defies credulity to challenge it.
Let's say that they expected 1 million voters in the county they're looking at. They would leave about 70% of those votes untouched in their natural state to make things look natural. Then for the remaining 30%, they would simply substitute the real voting results with fake ones to make up the margin that they need to win by the desired amount. In my example, the 70% of votes that are untouched would be 364K T, 336K B. To win with a ratio of 55-45 with 1 million total votes, the remaining 300K votes would have to have NO MORE than 86K votes for Trump or 28% of the votes. The remaining 72% were would mostly be given to Biden, but some of them (calculated by a bounded random number generator would be given to 3rd party candidates.)
They couldn't just throw in 216,000 Biden votes into one precinct because it would be super obvious, and they couldn't just sprinkle them on the top of a bunch of different precincts because it would not guarantee them a victory and would put them at risk of overvoting on some precincts of other problems that could get them caught.
So they used this wheel method to come up with different ratios: 1/10, 4/65, some even favoring Trump. A whole bunch of them and mathematically calculated how many votes were needed in each bucket ratio to balance the votes across all of these ratios to GUARANTEE that the final result would be the ratio (72% Biden, 28% Trump) they want once everything was compiled. The guy in this video calls these "virtual precincts"
Then they needed to spread these votes out amongst the REAL precincts. They couldn't just assign fixed ratios permanently to one precinct because 1) it would be too obvious, and 2) a lot of the precincts didn't contain enough votes to make up the numbers. So they targeted mostly low turnout precincts (because they'd need to stuff real ballots in these actual precincts to make the results auditable. Might explain why there are SO MANY Biden only ballots, as well as the abnormally high turnout in some precincts) and added votes at these different fixed ratios in an algorithmically determined pattern. Once one precinct's maximum vote potential was nearly exhausted, the algorithm would stop in that precinct and pop up in another precinct, and another, and another, until that vote ratio hit it's "quota" to be balanced against all of the other ratios. That's why he can observe this repeatable pattern of a county reporting a certain ratio then falling off that ratio just before another county picks up that same ratio again.
The wheel number he calculated was something like 2200, which means that the largest common denominator of the fractional ratios that were applied was that wheel number, with dozens or hundreds of ratios balanced around their end goal result (which in Georgia was I think 14.6%).
I hope that this summary makes some degree of sense. Let me know if it sounds like I'm speaking crazy talk. It's late and this is complicated, lol.
Brilliant summary. But the key point for me is the actual 'proof' that what he's explaining is true and not some fantasy of a crazy maths guy. I hope I understood this right:
He identified the virtual precincts, the ones which were hijacked by the algorithm, and isolated all of the reporting during the hijack period. He found the average vote yotal was 14.65% Trump. As each leaf in the wheel is weighted the same, he found that he could substitute the reported numbers with a random number of 40,870 for every precinct, an it still gave the Trump percentage of 14.65.
He put the chance of that happening naturally at 1 in 60 billion.
I think the 1 in 60 billion calculation was off by a factor of 10 because he said "I'll give this a 2% (0.02) margin of error but then actually entered in 20% (0.2) but yeah, the point still stands, lol.
And yeah, he showed that no matter how many total votes there were in all of the affected precincts, as long as the Trump votes followed the ratios that he observed, the overall total % of Trump votes for the state would ALWAYS be 14.6% plus or minus the rounding error (which was also fixed AGAINST Trump by what he called the Snapping Effect)
I saw that too! That's when I stopped watching. I believe his results, but I don't think he did a good enough job with the delivery. I am considering making a 5 minute version of his results that is polished, to the point, and geared for normies.