Election Data Analyzer

More factoids about the 60 homes where couples enjoy trekking down to the Clerks Office every 45 days for about 6 months to change the Party affiliation on their voter record, (after all what could be more fun than that), 83% are REP ranging in age from mid 30s to 80s. They all appear to be married couples judging by the same last name. Next, we will see if these married couples toggle there parties TO THE SAME VALUES...between periods over and over again.

Another factoid about the "Party Togglers" I speak of above...apparently the practice of changing your mind about which party you belong to about every 45 days for 6 months runs in the family. Of the 925 unique address where these "togglers" live, 60 of them contain 2 registrants! Wow! Dinner table conversation in December 2020...."Honey, how about we change our party affiliation from REP to Unaffiliated this month? Great idea Babe. I'll pick up some forms from the Clerks office tomorrow. Honey, while you are at it, grab a few extra forms. You never know if in Feb, Apr or May we may want to change our minds again."

Finally, here is the distribution of full name lengths. I don't see anything unusual in here to investigate. I typically will look at the full names with really low or really high character counts. Sometimes you will find strange results. The most interesting last name I ever found was something like "No More Walking Grey Horses" (scrambled version)....and that was just the last name.

Here is the YOB distribution of the "party togglers". Apparently there are 80-90 year olds who can't make up their minds what party they should belong to so they change it about every 45 days on average for 6 months during a period of time where there is no election going on.🤨

And here is the precinct distribution. Just for information. I am only showing the last digit. The X-Axis is the count. I will use this to help with canvassing. I don't see anything unusual.

To continue the thread above, I now have the 985 "party togglers" extracted from the main EDA analysis and have run a new EDA analysis on just that population subset. Looks like the "party togglers" really like to vote. 932 of the 985 voted. Incredible. Again, please try this with your own data. I am just trying to present different ways of using the EDA tool.

If this does not make your blood boil well your heart is not pumping. The left pie chart is the party affiliation of the entire population from one EDA analysis. On the right is the party affiliation of the subset of 984 records that have thier party affiliation "toggling" between parties. Red is REP. Blue is DEM. 😡

This is the year that awareness of the fraudulent election system we have transitions into action. If you are one who settles into the couch every night to shut off reality through watching the boob tube, it's time to put down the remote and take action. It is clear our elected officials have no plans to confront the fraud. They simply aren't brave enough. It is up to you! If you are in a state that has laws which allow you to create an initiative, or proposition to get on the November 2022 ballot that would declare the 2020 election fraudulent and since our elected officials did not fix it, therefore the 2022 election is also fraudulent. Fraud obviates everything. All of those elected during this time are immediately unelected. They pay us back their paychecks over time and any measure passed they voted on is deleted. A new election will be held within three months using hand counted paper ballots and done in public. Each precinct will post the results in public and you then can take a selfie with it then check your county's data online and all the way up to your state. The time is now. There is no more waiting for someone to do something. YOU are the person you've been waiting for to rescue us from this insanity!

Folks, I am going to take on a very deep case study based on the information above for the 985 voters where the registrations party affiliation are flipping. Here is the basic approach so you can try this at home. That is the whole point, don't believe me, look at your own data this way. I loaded a new EDA file with the entire data set for County and am running it. Then, I will take the 985 Voter IDs in question and using the Filter01 tab extract out the individual records, probaply in several chunks and then merge them. Once I have those records, I will take that subset and run another new EDA analysis on just those 985 address. Now I have as a baseline the entire county and the subset, both in EDA. This will be interesting to compare some of the dashboard charts. Then I will take the 985 and using MapIt tab, classify each one by TYPE. At that point, the data analysis is done and the data interpertation phase starts. After that, I will try and arrange a canvass to ask these 985 people why they keep changing their party affiliations. Hopefully this focused approach is going to quickly find problems rather than spend countless hours knocking on fruitless doors....we will see. This is goingto take some time.

Let me pose a hypothetical. Aren't the number of registrations by Party in a District taken into account when Redistricting occurs? What if it was possible to flip some Party affiliations to drive a District border decision "this way" or "that way" based on the Party affiliations in a Precinct or District? https://redistricting.lls.edu/redistricting-101/who-draws-the-lines/

Party Switching cont'd: Here is another county. I wanted to see if the first was just an abberation. It appears not be. Out of ~220,000 registrants, 985 are toggling between parties. Not only are they toggling between parties, I made a few other observations. They only toggle between 2 parties, never more than that. 99%+ toggling stops between period 4 and 5. Finally, if I sort on those that changed between periods 2 and 3, there are NONE that have stayed the same between 1 and 2. What that means is that there are NONE that are for example REP-REP-DEM-REP-DEM because that type of pattern would have 1 and 2 being the same before a change between 2 and 3, for those 985 that are toggling. Remember, these periods are very close, starting in Dec20 and ending in June 21. Time to reach out for an official explanation, if there is one.

Party Switching: This is a sample set of 2532 records of voters who are switching parties between 5 successive periods of time over 6 months. Is this how real people behave? Each color is a different party. There are "people" here who decide they are a REP one month, then become Unaffiliated, then back to REP, then back to Unaffiliated. Maybe the explanation for that is not having any designation of Party on whatever triggers the update causes them to be set to Unaffiliated? But what about the "people" who are DEM then REP in succesive months then back to DEM etc....? Folks, this is really weird. I don't know what to make of it at this point to be frank. I have many other questions about these patterns. I am going to dig. I suggest you do the same. Let's see what we come up with.

In this example, I just looked at the "volatility" of the party status alone of the voter IDs of the same 5 periods. This comes from a list of 485,000 registrants. 2,169 had greater than 75% changes on party status alone over 5 periods! This means that at least 4 out 5 periods that the person changed the Party status. Here is an example. This REP voter toggled between REP and Unaffiliated for 5 straight periods. How does this happen? How much do we have to prove before we just jettison the entire roll and start over? Do we really need to canvass this? Can't we just prove the data is garbage or highly suspect and start again?

And one final example for this type of case study. Orange shows changes between periods. So is this just sloppy book keeping or something more nefarious? I don't know. If it is sloppy book keeping doesn't that call into question the database management?

Here is another example of a voter with a high score for record "volatility" over 6 months, same county, same rolls as above. This person also moved twice and registered using different address. Notice the toggling between REP and Unaffiliated on this one, unlike the one above which toggled between DEM and Unaffilated.

A few observations about President Trump's Florence, AZ rally....

I am working on a new concept related to the TPDA. But, it looks at a more granular level of changes to individual registration records though time. The algorithim produced this result with what I am calling "high volatility" because it changes alot through time. I have included some actual data here but kept it anonymous. This is real data. The color coding indicates where changes were made between successive periods of time. These are the dates of the registration roll. The pictures are the two address that this "person" is moving between every 1-2 months (yea right) which are 2 miles apart. What a weird first name, right? It keeps changing, betwen a normal name and nonsense. I can assure you, the middle and last look like nonsense to me as well. Then look at the party. Toggling between DEM and UAF every 1-2 months. It seems this way of looking at the roll might have some usefulness. This is a demonstration of the lack of data integrity on our rolls. Why do we tolerate this?

For example: "North Dakota Legislative Council" Estimated Fiscal Impact of a Live Streamed Hand Count: The one time costs for: "$30,000 to stand up websites for the 3 counties that do not have a website on which they could post their counts and livestream" "$500,000 to for the technology necessary in the 53 counties to enable the live streaming" "The ongoing costs for: $16,000 per election - 1,600 poll workers x $10 (average wage per hour for a poll worker) for the extra hour necessary for the hand-counting of ballots" ------------------------------------------------ Take aways: $546,000 for the first year $16,000 for the following years $1.51 per ballot for hand counting the first year. $0.04 per ballot for hand counting each year afterwards Hand counting takes one hour. ------------------------------------------------ From the ES&S contract with the state of ND: $9,531,887 the first year or $26.36 per ballot Ongoing Hardware and Software Maintenance, Support $559,390 PER YEAR or $1.55 per ballot

Folks, Take a look at these numbers. Incredible.