2000 Mules Ballot Harvesting Video — A Reality Check
By John R Crary
As a computer scientist, I spent a career working with pharmaceutical research scientists to manage their data and information operations. This involved chemical & biological registration, systems analysis, data validation, data migration, and discovery analytics. These systems allowed scientists to search and report on novel interactions and associations electronically. This required the ability to access multiple scientific domains to identify critical relationships for preclinical studies. I am using those skills to dive deeply into the 2000 Mules video and its conclusions.
Scope of 2000 Mules Election Claims
The video’s primary accusation by the True the Vote organization is the use of drop boxes in large cities and that they are susceptible to large-scale ballot harvesting. They intend to prove that this activity favored Biden for president and why he was elected. Using Atlanta, Georgia, as an example, they then extrapolate these conclusions to the rest of Georgia and the other battleground states Trump lost.
After watching the video, it makes a point that definitely will register with people who think that the 2020 presidential election was stolen. The video indicates that they identified 242 people as mules, and that dozens of them were later found to be part of the local BLM riots. I can only presume that these are accurate GPS data, but I need to look closer at how the video analyzed the data and made their mule count conclusions.
The use of cell phone GPS data is a big industry and is best used as a marketing tool to contact people and entice them to visit their businesses and make purchases. It gets even more effective if they can connect GPS data to a person’s identity, allowing organizations to profile customers for targeted advertisement. More generally, it is used by businesses for “Customer Enhancement, Network and Supply Chain Optimization, Workforce Coverage and Support Force Optimization, GeoMarketing and POI Enhancement, Route Planning, and Optimization.”
The 2000 Mules investigation did not have GPS business locations but instead had the GPS locations of election drop boxes, which seems straightforward enough. There is also a list of undefined “Stash Houses or Nonprofit Organizations,” which are GPS locations where they claim the mules collected their ballots for delivery to drop boxes in the same area.
The 2000 Mules video logic for identifying mules is as follows:
Collect petabytes of cell phone GPS data from October 1, 2020, until the 2020 election for 309 drop boxes and an unknown number of Stash Houses in the Atlanta metro area. Then examine those records to identify election mules by the following rules:
Mule Rule: Any mobile device “near” an election drop box 10 or more times AND a stash house more than 5 times in two weeks is considered aberrant election behavior. They then regard this unusual behavior as proof of ballot harvesting and those people then classified as ballot harvesting mules.
Examination of 2000 Mules Election Claims
What True the Vote means by “near” or how long the mule must remain at each location is not defined. The fact that cell phone GPS tracking is only accurate to about 16ft does not seem to be considered in their analysis. Also, by going “near” 10 drop boxes in 2 weeks, does it mean the same drop box or different boxes on the same or different days, a really important point?
Why or how the stash houses were chosen was never provided, and what evidence they have to support those determinations remains unknown. They only claim these are nonprofit organizations and fail to indicate if these are churches, charitable groups, or political organizations but nothing specific. Also, there are claims of “Ballot Collectors” providing the stash houses with ballots, where the mules get paid $10/absentee ballot to take them to area drop boxes.
The “Collectors” who gather ballots for the stash houses are also not defined or the conditions by which they operate. The Collectors would need an extensive reservoir of registered voters with matching signatures; otherwise, the ballots would be rejected and not counted. That sounds like a lot of challenging and complex maneuvering with significant risk.
So why go through all the work?
If this is the scenario the video proposes, something is terribly wrong. Why not mail absentee ballots into the voting precincts rather than hand carry them to drop boxes? That would be much easier, cost much less, and would not expose those committing the crime. Organizing ballot harvesting on a grand scale using so many people would be rather stupid… just mail the damn things!
So what does science say about these types of studies?
Remember that these metro areas are living neighborhoods for many people that go through their lives visiting friends, doing business, walking dogs, and so much more. So, I should think that these metro areas where the drop boxes are located would see a vast amount of random foot traffic. It would be a real effort to separate the mules from the sheep with little clarity for determining their intent.
You must understand that the Atlanta metro area has just under 6.1 million people. That is a lot of people wandering about living their lives and going about their daily business. Much of where they go and how many drop boxes or stash houses they pass have a statistical nature.
Get John R Crary’s stories in your inbox
Join Medium for free to get updates from this writer.
Typically in this kind of study, you would see a probability curve like the one shown below. The dark green bars represent the many people who pass near or visit 0, 1, 2, or 3 drop box/stash house “Mule” locations over two weeks. These low frequency citizens would dominate the population of people in the metro area. These bars are truncated at the top to better illustrate the lower graph of proposed mules.
The light green bars show random mule location encounters due to ordinary citizens moving around the metro area doing their daily business. These people likely have no idea they are passing near these locations unless actually depositing ballots. The 2000 Mules video does not question the intent of these people because they do not meet the Mule Rule defined above.
Lastly, we have those citizens represented by the red bars, and those that the 2000 Mules video implies are responsible for ballot harvesting. This group would be a small set of individuals, so we need to consider why these individuals are visiting so many mule locations. We need to understand the paths of people like postmen, delivery drivers, and police on neighborhood patrols. We also need to include average citizens doing their daily thing, such as going out for walks, running, shopping, and not-so-average people like drug dealers fishing for customers. Did the video investigations eliminate any of these patterns?
Going “near” mule locations can be explained by life activities other than those harvesting ballots. It not only depends on the size of the local population, the density of vibrant businesses, walk-through traffic, the locations of drop boxes in the community, and the location of implied stash houses. Changes in any of those variables would change the results in dramatic ways. I am not much of a legal expert, but scientifically the ballot harvesting theory described in the video is the actual fraud and, I suspect, the reason why these claims have not garnered any law enforcement investigations.
A single person passing a low number of drop boxes has a high probability. Other people passing a high number of drop boxes has a low probability, but depending on their lifestyle and work habits, it is certainly not zero. It is not unreasonable to assume that out of 6.1 million people in a city, there would be 242 individuals that meet the mule rule requirement.
But the Mules were at the BLM riots!
Yes, I would have to agree with that general point! It would be reasonable to think that many people wandering the community would also show up at a local BLM riot. In either case, there is no way to determine that they were committing crimes unless their GPS locations put them inside businesses vandalized during the riots. This would be the same as the January 6th insurrection in Washington, DC. Just the fact that they were trespassing should be enough for criminal investigations. Anyone destroying property, and committing violent crimes must be punished.
What this means from a scientific perspective
Most likely, the current mule numbers are just a measure of normal human flow and are unrelated to any specific criminal activity. The 2000 Mules video is just an attempt to use regular citizen activity to prove an illusional political point that hopes to reflect criminal intent.
The problem I see here is that accusing people of crimes based on GPS location data is a can of worms that can be used against anyone. One time I was running with a friend in San Francisco, and we ran into Golden Gate Park, then realized we were standing in a group of gay men. Not that I want to criticize other people’s lifestyles, I really do not care except that it is not who I am. The point is that if someone monitored our GPS signals, they might have reason to claim we were also gay. There are millions of such scenarios, so be careful what you call evidence!
Watching the 2000 Mules video, I counted one man depositing six ballots into a drop box and, cannot for the life of me, explain how that means he was a mule. This is what the narrators are implying, but if you check the Georgia voter guide, you will see that this is not an unreasonable deposit for a family.
My biggest concerns about the 2000 Mules video are:
· The numbers they report could easily be due to regular community activity.
· What determines a person being near a drop box is unclear.
· Why does a person’s aberrant behavior a cause for claiming ballot fraud?
· The characterization of Stash Houses is vague as only being nonprofit organizations.
· The group of individuals called Collectors that furnish the stash houses with illegal ballots does not seem plausible beyond just family and friends and is not at all viable to the extent described in the film.
· Absentee ballots undergo validation processes before being mailed out and after receiving them. So any anomalies such as unacceptable ballots and signatures, duplicate submissions, and unlisted voters will be discovered and eliminated from the vote counts.
· Finally, instead of ballot harvesting using a complex crime scheme, why don’t they just mail the absentee ballots using regular mail? It would be much cheaper and more secure than organizing hundreds of people to run ballots to drop boxes!
The 2020 Mules video paints a large picture of theoretical ballot fraud in our country, but much of that picture remains a blank canvas. Could so much election fraud activity operate in our cities without someone telling the authorities resulting in criminal indictments and prosecutions… I think not!
It is not that I think that public oversight of elections is not appropriate and valuable to our democracy, but it is a threat to our democracy if it attempts to degrade the state’s authority in favor of party interests. I believe in the US Constitution, law enforcement, the separation of church and state, and the value of a balance between the states and the federal government. It is not perfect, but it has served us well for almost 250 years, so let us keep it going for the next generation of Americans.
