Hi again!
I'm Justin Mealey, and I testified at the Georgia Senate hearing today. Our team provided hard evidence of voter fraud, using the same data the Georgia certified the state with.
Here's a copy of the testimony: https://rumble.com/vcay7j-data-scientists-shocking-election-testimony.html
I wanted to do an AMA so that people can ask more questions related to our data methodology, clarify items about the voting process which we painstakingly investigated across multiple states, and hear your ideas about we could better get the word out about the fact that we seem to be one of the only groups operating off of hard, irrefutable conclusions based off of data.
EDIT: Thanks so much for the questions (heading to bed) -- hope I was able to clarify a few things for you guys. We'll ask Dave (the head data scientist who also testified from my group) to come do any AMA tomorrow as well.
EDIT 2: I'm sort of back right now (9AM EST) so will be periodically checking for new questions as I refresh tdw looking for spicy memes to repost on facebook.
EDIT 3: (10:32AM EST) I'm going to post a reply to a MrCaveman (which, thank you for the question) that I really want everyone to read:
https://thedonald.win/p/11RO7PRc9Q/x/c/4Drwoe2gIJ7?d=50
When doing work that you deem is important, the most vital thing you can have is focus. A lot of the times that means putting to the side all of the noise that surrounds a certain path. The poll pads are the noise when it comes to the actual ability to commit fraud during this election.
If you were creating a system to enact a fraud, how many points of contact would you design for that system to interface with in the voting process? How many confederates would you need to enable in that system? One way we've discovered only requires one true confederate to enact in a county, and we've actually identified some of these actual confederates. Depending on how things go, we might have to just release that in a video in the future.
My point being, that while your intentions are good (as most everyone's on this site's are), they distract from the actual fraud. By distracting from the actual fraud, parts of which we've proved through hard data analysis, it actually detracts from the ability for us to bring that fraud to light and abolish it.
Please, for the love of God, stop talking about poll pads.
While it's perfectly fine that people are approaching things from that perspective, like I mentioned in the beginning of the testimony, it's just not our approach. We only care about things which can prove where, for instance, votes move from Trump to Biden. Those are the hard facts that are irrefutable as long as your dataset is correct, and our dataset is literally the same data that the secretary of state uses.
Thank you for your participation on this site and willingness to interact.
The weakest part of your team’s testimony: there was NO demonstration of a change in the ground situation.
Historically, were there ever negative votes or “switches” in the Edison data feed or similar sources? Was this particularly evident in 2020 in areas of notorious fraud? Or were there comparable events in 2018, 2016, etc? I am looking for a delta.
Not sure if you think this observation is on point, but some mention of this in your presentation would make the conclusion considerably stronger, in my opinion. Thanks again!
The Edison dataset was only a starting point. The dataset we're using mainly is the SCYTL data, which is 1-for-1 the secretary of state's data.
What level of negative votes would be okay? Would it matter that votes go negative previously, even if a single instance of it points towards a systems issue?
Thanks so much for your response. I appreciate your level of detail. Your correction is noted.
Edison, by my corrected lights, is the precursor to network reporting. Scytl/SOS is where the negative anomalies would arise - undetectable in the Edison feed unless the refresh catches it at just the right interval.
My skeptical mind is not won over. Unless a change in the ground situation is demonstrated, then ANY level of negative vote tabulation is acceptable.
I’m on your side, but I would prefer your presentation be unassailable.
My imagination tells me that negative votes could result from any number of ground situations. For example:
Precinct A reports 1000 new votes. They send a region-wide spreadsheet with +1000 votes straight to the national reporting service. But it fails to include 1000 precinct B votes, because precinct B changes were simply not saved to the shared spreadsheet. So it appears that 1000 votes were “subtracted” from precinct B.
After multiple updates, auto-saves, and cloud communications, the results from precinct A and Precinct B sync up again. The spreadsheet data are reconciled a few minutes later.
If I were the opposition, I would simply argue it’s is an artifact of cloud computing. Therefore, I wouldn’t care how MANY negative votes were detected... my imaginary model explains it.
An easy way to overcome this objection would be to demonstrate a significant change from previous election feeds. That would produce a strong inference of intentional interference and clear fraud.
By contrast, while the “exact number” switches you showed were impressive and compelling to me, they are still only an argument from improbability.
I want to see an airtight DEMONSTRATION that something unusual occurred in the districts notorious for fraud. That would require some reference to or acknowledgement of 2018, 16, 14 etc data, in my opinion.
Again, thanks so much for your work.
I am not an expert in anything. My “imaginary” model is likely flawed, but it is logically plausible to a layperson.
Overcome that objection please, and shut down that logical loop succinctly in your presentation if possible. Then there will be winning.... the likes of which you have never seen;)