Compare to the House race that wasn't that close and went to the Republicans and you'll see the Dems were really turning up the cheat codes to try and win this one!!!
I did not write any python for this one. I did write some python trying to find issues out in VA but was unsuccessful. For this, I just got the data off the website and combined some of the precinct level information across multiple downloads to get the right data set.
Unfortunately, the method seems to have a flaw. You don't know how many straight ticket voters there are in total.
So if there is a relatively low number of straight ticket voters, even just a difference of 1-2 can have significant impact on the difference in percent for that precinct.
You can figure out straight ticket pretty well the more down ballot & by precinct info they give you which NE has good precinct level info. I took the worst down ballot R %age in the precinct and used that for the straight ticket.
I have also been doing enough of this in other states which weren't getting as much attention and this is the first match I have found. You can also notice the pattern in the data before you even plot it.
Neighboring Sarby county which is only partially in the congressional district also exhibits some funny patterns but not the same exact shape that's clearly a software algorithm. My guess is the algorithm might work differently if there's less precincts but I don't know enough to say it's legit like Douglas. Also, I know Omaha pretty well and wouldn't expect there to be much old-fashioned voter fraud like they do in philly.
Wait. They gave you down ballot percent for the precinct and down ballot vote total, so you could find the total number of down ballot Republican tickets for that precinct? *not down ballot, I mean straight party ticket
Because the actual number of total party down ballots and their percent breakdown by party has been the only real issue I've had with Dr Shiva's method. You never know how many straight ticket votes for either party, so a precinct with relatively few can see a massive percent change from just one person hitting wrong one wrong button.
Elections in Philly are more fraudulent than they are legit these days.
Looks like Douglas county NE follows the Dr. Shiva discovery.
Data is available by precinct: https://electionresults.nebraska.gov/
Compare to the House race that wasn't that close and went to the Republicans and you'll see the Dems were really turning up the cheat codes to try and win this one!!!
Your work?
Yes
Is the x axis normalized?
Nice, let me ask you, Michigan has a two-track voting system as Shiva described. Does Nebraska have the same or did you have to modify the math?
Nebraska does not publish in two tracks. I used available down ballot data to replicate as similar as I could the stated concepts Shiva described.
Very nice! Are you planning on doing any more?
I think it would be awesome to see how this pattern may or may not occur in different areas of the country.
Yes, going to work on it more tonight and as long as it takes.
Fantastic! Thank you for all your work!
I saw the 30k ish Nebraska flip votes. I was wondering if it was all in NE-2. Hopefully they will look into this. Nice work
I will report it.
I did not write any python for this one. I did write some python trying to find issues out in VA but was unsuccessful. For this, I just got the data off the website and combined some of the precinct level information across multiple downloads to get the right data set.
I have reported it. I will look at the VA project & see if what I've been doing will add to what they're working on
What are the axis variables?
I used the same method as here: https://www.pscp.tv/va_shiva/1BdGYYjgkgQGX?t=20m2s
Y-Axis is %age that the President's votes performed different to down ballot republicans. X-Axis is how Republican the Precinct is overall.
Unfortunately, the method seems to have a flaw. You don't know how many straight ticket voters there are in total.
So if there is a relatively low number of straight ticket voters, even just a difference of 1-2 can have significant impact on the difference in percent for that precinct.
Or is that data available?
You can figure out straight ticket pretty well the more down ballot & by precinct info they give you which NE has good precinct level info. I took the worst down ballot R %age in the precinct and used that for the straight ticket.
I have also been doing enough of this in other states which weren't getting as much attention and this is the first match I have found. You can also notice the pattern in the data before you even plot it.
Neighboring Sarby county which is only partially in the congressional district also exhibits some funny patterns but not the same exact shape that's clearly a software algorithm. My guess is the algorithm might work differently if there's less precincts but I don't know enough to say it's legit like Douglas. Also, I know Omaha pretty well and wouldn't expect there to be much old-fashioned voter fraud like they do in philly.
Wait. They gave you down ballot percent for the precinct and down ballot vote total, so you could find the total number of down ballot Republican tickets for that precinct? *not down ballot, I mean straight party ticket
Because the actual number of total party down ballots and their percent breakdown by party has been the only real issue I've had with Dr Shiva's method. You never know how many straight ticket votes for either party, so a precinct with relatively few can see a massive percent change from just one person hitting wrong one wrong button.
Elections in Philly are more fraudulent than they are legit these days.
https://bitcadia.github.io/DownBallot/index.html
All states individually.
https://bitcadia.github.io/DownBallot/Outputs/CountyDownBallotDiffsByStraightCompare.html?First=TX&Second=IL%2CMN%2CWI%2C+FL
Compare states
Not mine so I can't vouch for accuracy.