May 17, 2022Liked by Mathew Crawford

I was reading Benford breakdowns the day after. They stole that fuckin election.

Expand full comment

Excellent job widening the use of the tool! Let’s call it Extended (or Generalized) BL.

One thing that drives me crazy is people equivocating voter fraud and election fraud. I understand that all the talking heads do, which doesn’t help. And even historical uses appear to admit this thing I consider a mistake. Perhaps it’s intentional. As in, to cover the more pernicious issue.

I think it’s important to distinguish between front end fraud and back end fraud. The back end fraud being machine- and hand-counting for example, and the front end fraud being illegal voting and fake ballots for example. The illegal voting seems to be what the powers that be want us to argue about. But the data suggest it’s the lesser problem, and in most if not all cases doesn’t make a difference in outcomes. I’d like to see serious research conducted into each facet, starting with the most pernicious. Extended BL seems to be a great tool to illustrate geographic areas where the research is worthwhile.

Another thing that drives me crazy is the tribal nature of this issue. Red team didn’t seem to care too much when Trump seemed to benefit. Blue team seemed downright giddy when Biden benefitted. In fact, a blue team friend of mine has been railing against this kind of fraud since 2004. When I brought it up for 2020, his answer was something like, “well, our team finally got one, so I’ll become active again next time”. A red team friend of mine won’t entertain the idea of issues in 2016. Until and unless we approach this apolitically, we’re all doomed. I personally haven’t voted since 2000, after I became convinced that fraud has decided every election since. With more research, possibly all the way back to 1960. Real apolitical research will horrify the public and undermine the US on the global stage. That’s why I think it’ll never happen. Instead, we’ll be given the pressure release valve of voter ID and illegal voting as a distraction.

Lastly, 2000 mules is extremely distressing to me. Most people watching it that think there was fraud will be too wrapped up in the “proof” that they will miss the methodology. Pay close attention to the methodology (first 35 minutes ish). It shows that we’re a breath away from digital slavery/imprisonment. The group of people most likely to resist will be the same people paying attention to the “proof” not the methodology. They are then primed into thinking the methodology is ok. It isn’t, and we should all be horrified that people with enough money can buy this. We’ve created an industry that will help law enforcement bypass critical safeguards to keep them from abusing citizens. It’s not ok. Not even when it helps your cause. If it’s not ok for your enemy to use it against you, then it’s not ok to use it against your enemy.

And anyone using the, “well it’s gonna be used/done by them/someone, so why not me?” argument should know that’s how George Soros sleeps at night. Go find the video interview where he explains his role helping the nazis. I mean, I’m not perfect either - I might even have made the same decision as him at that age (13) - but I’m disgusted nonetheless that he doesn’t show regret or remorse. Even if I made the same decision, I doubt I’d say anything more than, “I’m sorry”. Reporter: “How do you sleep at night after helping the nazis?” Ideal Soros answer: “I don’t.”

Expand full comment
May 17, 2022Liked by Mathew Crawford

On some comment thread a week after the election, I came across the same Chicago data and I suggested that someone practiced in the art should look at the precinct voting data in base 3, or look at the data as logs of various bases. I was going to try it myself, but I forgot all about it.

Expand full comment
May 17, 2022·edited May 17, 2022Liked by Mathew Crawford

I am writing a series leading up to this on my own substack. It's a really interesting a surprisingly complex issue. The short version of my take is that Zipf's law is a little more generally useful for things like this. It's very closely related.

The worst take on it was by Stand up maths: https://www.youtube.com/watch?v=etx0k1nLn78

Essentially he was saying that whether or not a district follows Benford's law can't tell you anything, it's just an expression of the underlying data structure. Well, sure, but with that attitude you may as well chuck the entire edifice of scientific statistical mechanics out the window, because that critique applies to everything.

Also, my preprint paper discusses some of the issues around using these principles in analysis:


Expand full comment

I read this whole article, but my understanding of it is ignorant...however, I intuit that it has done nothing to change two suppositions I have been holding for awhile:

1. Trump couldn't win in 2024 even if he won.

2. If Dems do well in 2022 and hold on to power, then we can know that the Dem party, big media, big tech, big banks, big corps, the military, the CIA, FBI, CDC, FDA, NIH and Pharma have the tech to decide who will win, no matter what Americans voted for.

Expand full comment

Somehow, part of the article vanished upon publication, but I think it's all back now.

Expand full comment

“If they ignore you, you still win”

Mathew Crawford and company,

I understand you and your associates are having difficulty getting some sort of official response of any kind to your concerns of the safety signals of the Covid-19 vaccines and other issues surrounding the disease and the response to it.

Now I don’t mean to give this as legal advice, so don’t take it as such, but I’ve stumbled onto a method that you all might be able to make use of to your advantage. I’ve attached several sources below that argue and claim to explain how to make use of notices and affidavits to elicit a response to your objections and concerns.

The method might be similar to what worked with Safeway in Hawaii and DMED. They claim success with the United Auto Workers and even being instrumental in stopping the OSHA mandate.

It is particularly attractive as it would give everyone, every individual concerned with a given issue with a word processor, something to do from their house.

Their argument is that with notices sent as they outline, and if they don’t desist then sending affidavits counting as sworn testimony, that the recipient has to respond. They would have to reply point by point explaining where they have constitutional authority, or to rebut your affidavit on matters of fact, within a time limit or the notice and affidavit can be used as evidence before a court of record, and that they ‘acquiesce’ to everything in your affidavit as true if they ignore it. If they lie in their response it counts as lying under oath.

In short, the argument is that if they ignore you completely it is as good as a confession that you are right or they are breaking the law. The catch is you have to have a solid argument with the truth. They claim to have success getting institutions to relent on a number of issues by sending these; even if they don’t openly respond or acknowledge them, the problem stops. If not they are trying to gain access to the grand juries and the courts. It is a bit more involved than just the usual petition or letter however, and you all would have to pick the addressees of these notices and affidavits.

I know that these sources might be on one end of the political spectrum, but perhaps their particular method has merit despite your differences. Please share this with anyone who may be interested or you think would benefit from it. Including The Unity Project, Steve Kirsch, Robert Malone, Alex Berenson, Jessica Rose, Toby Rogers, Aaron Siri, and, El Gato Malo.

YouTube / Odysee / Rumble Channels

- ' Affidavit Mommas 2021 ' / anonymous on Odysse / Affidavit Mommas 2021- (This should have the most concise explanations) The original channel I was going to show you is down.

“Affidavit Mommas on the Process of how you do Notices and Affidavits”

“What you might want to know about Notices and Affidavits”

Other Website: https://affidavitmommas.com, they have prepared examples here they also have a telegram account.

- ' Dave Cares for You ' YouTube is still up, also on Odysee and Rumble

He has a lot of videos and is tied to Josh Barnett below. He seems to have been the first to figure this method out, even though he doesn’t have a formal education.

“The Hidden Secret Revealed that stops Gov Corruption [Share RAPIDLY TO FREE THE PEOPLE]Dave Jose”


“Big Victories versus UAW …”


“Federal Mandate Owned Josh Barnett and David Jose method drops affidavit in Fed case (Praise God)!!!”


Other Website: www.RestoreMYRepublic.com

- ' Josh Barnett for US Congress ' - (Yes I know, this is not an endorsement)

“How to stop mandatory V’s” now seems to only have a Odysee channel

Other Website: https://www.barnettforaz.com

Expand full comment

I don't grok this, but it makes me want to get back into mathematics. Thank you.

Expand full comment

I first heard about Befords law during the 2020 election. I learned more about wikipedia than anything else from that - watching it being updated in real-time to try to discredit it for election results.

Expand full comment

I liked Crawford's filter, since you lack a filter. Good one!

Expand full comment

Mathew, I can see only the first paragraph of the paper as an image. Can you provide a link to the whole paper?

Expand full comment

For us old folks, the most interesting factoid about Benford's Law (and maybe the only interesting one) is not mentioned here. BL was first noticed by an astronomer observing the pattern of wear on the pages of a book of logarithms is nonuniform. Some of us are old enough to have had such a book, and even know why such used to have a purpose--so much so that the pages would wear. Also, not mentioned here is that there is no connection between 2000 Mules and BL analysis except that both have something to do with the 2020 election (one far more than the other, in my opinion.) Regarding the possible extension of BL to the election data problem (precincts are intentionally constructed to have similar populations--pretty much the opposite numeric distribution that is amenable to BL) by using BL in non-integer bases. I remain skeptical without reading all the references. The underlying data points are all in the same order of magnitude regardless of how you play with the units and I have trouble getting past that.

Expand full comment

Great idea regarding use of smaller bases. I had looked at the BL issue right after the 2020 election but had moved on quickly after the objection of the district size limitations reducing the orders of magnitude and (supposed) applicability. Using bases less than ten easily removes this objection.

Expand full comment

Election fraud very well may be allowed to make people think that there's a good guy and a bad guy.

I recently found out via James roguski that trump was the one who put the authoritarian amendments proposals to the WHO...

After all, each candidate has to be vetted by their party in a primary.

So all this number play might prove something, but it just reinforces the idea that we even have a choice.

And a funny from George Carlin on voting https://youtu.be/xIraCchPDhk

Expand full comment

for the mathematically illiterate the article seemed to be leading up to something it never reached....

Expand full comment
Comment deleted
Expand full comment