I showed it to a friend, and I got back a link that "debunked" it and was shown a wikipedia page that calls him a con artist and conspiracy theorist. This media blitz is of insane proportions. They were well prepared.
The analysis in this video is actually quite sketchy.
If you make a few assumptions you can model this and it would appear that the downward trend is an expected result.
Let's assume for example that 95% of down ballot republican voters vote for Trump, and 3% of down ballot democrat voters vote for Trump we can take a precinct with 1000 voters and compare two scenarios:
Scenario A - 10% of the voters are Republican and 90% are Democrat. (Left side of the graph)
With the assumptions above you would see
100 down ballot R votes and 900 Down ballot D votes
Take 95% of the 100 R votes leaves you with 95 Trump votes
Take 3% of the 900 D votes leaves you with 27 Trump votes.
Total number of Trump votes is 122 (95 + 27)
Result is that 10% down ballot R but 12.2% are Trump votes
Scenario B - 90% of the voters are Republican and 10% are Democrat. (Rightside of the graph)
900 down ballot R votes and 100 Down ballot D votes
Take 95% of the 900 R votes leaves you with 855 Trump votes
Take 3% of the 100 D votes leaves you with 3 Trump votes.
Total number of Trump votes is 858 (855 + 3)
Result is that 90% down ballot R but 85.8% are Trump votes
This explains the exact downward trend you see in the graph above despite seeming counter intuitive that in republican dominated areas the difference between R votes and Trump votes actually trends downwards.
The important thing here is that you would need to exclude all of the down ballot democrat voters that voted for Trump in order to see the predicted linear pattern since you would then correctly be comparing the percentage between R voters and Trump voting R voters.
That's my take on the data at least, maybe someone else can explain why I'm wrong here, but I think it's important that we have these discussions openly especially if we are looking for evidence that would hold up in court (which should be the end goal here IMO)
Because most people vote the same down ballot. This tendency doesn't magically go away the deeper into Trump country you are. This is the method used to change the trump votes, yet not change the total voters and be harder to detect. As the county will stay red, and the total amount of voters don't change either , yet a sizable amount of votes for the popular get swapped each time.
The left side of the graph shows that trump outperformed just straight republican ticket, meaning he got votes with democrats down ballot in addition to straight republican tickets.
The Y axis is Trump votes percentage MINUS Straight Republicans percentage
What they say before the graphs is the Y-axis is %Trump individual candidate votes - %straight republican ticket. The graphs are labeled "Trump Candidate %" but they seem to be mislabeled.
Say there's a 80% Trump / 20% Biden precinct and the centrist 5% of voters on either side use individual ticket. In 100 people:
75/(75+15) = 83% Trump of straight ticket votes
5/(5+5) = 50% Trump of centrist, individual candidate votes
So what you think the graph shows is 80% overall Trump - 83% straight ticket and it should be close to the center line. But what it really shows is 50% - 83% or very far away from baseline.
Not everybody that gets an individual ballot is a centrist, but many will be because they are conflicted on who to vote for. So what this graph is really showing is %centrists for Trump vs %republicans for Trump. As a general rule centrists are about 50/50 so the more heavily Trump the more they will be away from his percent of single-ticket votes.
So why is there a sharp bend at 20%? Because before that there aren't any straight-ticket Trump votes, they are all Democrats voting for Trump on personality, but Democrat for other offices. So straight ticket Trump is 0 for the first 20% until you start getting precincts with actual Republicans in them. In other words it's measuring how wide centrism is.
I showed it to a friend, and I got back a link that "debunked" it and was shown a wikipedia page that calls him a con artist and conspiracy theorist. This media blitz is of insane proportions. They were well prepared.
Shiva didn’t gather the data. The clemson engineer did.
I wish they would make their data available so the study could be replicated.
Well, the data may or may not be good, but Shiva is a con artist who claims to be the "inventor" of email.
Take anything he says with a grain of salt.
Thank goodness Dinesh tweeted that! It's indeed a great video, worth watching.
The H1B1 spies working for Twitter were told to dispute this.
Dems had to skim votes off trump or else all their fake ballots would be spotted because there would be massive more ballots than people in the state.
Higher turnout for trump just made it easier to skim.
Incredible information to use!
WE HAVE THE SMARTEST PEOPLE
The analysis in this video is actually quite sketchy.
If you make a few assumptions you can model this and it would appear that the downward trend is an expected result.
Let's assume for example that 95% of down ballot republican voters vote for Trump, and 3% of down ballot democrat voters vote for Trump we can take a precinct with 1000 voters and compare two scenarios:
Scenario A - 10% of the voters are Republican and 90% are Democrat. (Left side of the graph) With the assumptions above you would see
Scenario B - 90% of the voters are Republican and 10% are Democrat. (Rightside of the graph)
This explains the exact downward trend you see in the graph above despite seeming counter intuitive that in republican dominated areas the difference between R votes and Trump votes actually trends downwards.
The important thing here is that you would need to exclude all of the down ballot democrat voters that voted for Trump in order to see the predicted linear pattern since you would then correctly be comparing the percentage between R voters and Trump voting R voters.
That's my take on the data at least, maybe someone else can explain why I'm wrong here, but I think it's important that we have these discussions openly especially if we are looking for evidence that would hold up in court (which should be the end goal here IMO)
Because most people vote the same down ballot. This tendency doesn't magically go away the deeper into Trump country you are. This is the method used to change the trump votes, yet not change the total voters and be harder to detect. As the county will stay red, and the total amount of voters don't change either , yet a sizable amount of votes for the popular get swapped each time.
The left side of the graph shows that trump outperformed just straight republican ticket, meaning he got votes with democrats down ballot in addition to straight republican tickets.
The Y axis is Trump votes percentage MINUS Straight Republicans percentage, aainst X which is just Straight Rep ticket percentage
So it plots the difference between Trump and Straight Rep ticket
The higher the Straight Rep ticket, somehow after 20% gradually he doesn't get more than the party, but less almost 30% less
Why would Republicans act differently in the Redder counties?
What they say before the graphs is the Y-axis is %Trump individual candidate votes - %straight republican ticket. The graphs are labeled "Trump Candidate %" but they seem to be mislabeled.
Say there's a 80% Trump / 20% Biden precinct and the centrist 5% of voters on either side use individual ticket. In 100 people:
75/(75+15) = 83% Trump of straight ticket votes
5/(5+5) = 50% Trump of centrist, individual candidate votes
So what you think the graph shows is 80% overall Trump - 83% straight ticket and it should be close to the center line. But what it really shows is 50% - 83% or very far away from baseline.
Not everybody that gets an individual ballot is a centrist, but many will be because they are conflicted on who to vote for. So what this graph is really showing is %centrists for Trump vs %republicans for Trump. As a general rule centrists are about 50/50 so the more heavily Trump the more they will be away from his percent of single-ticket votes.
So why is there a sharp bend at 20%? Because before that there aren't any straight-ticket Trump votes, they are all Democrats voting for Trump on personality, but Democrat for other offices. So straight ticket Trump is 0 for the first 20% until you start getting precincts with actual Republicans in them. In other words it's measuring how wide centrism is.
I have confirmed you are a moron
He is plotting the difference between Republicans voters and Trump voters
Republicans wouldn't change behavior based on how red their county is, not in this way.
The explanation is that it is there are more votes to steal from Trump there, without alter totals