The Wikipedia article has curiously been edited by a throwaway Wikipedia contributor around the time the GitHub repo got published:
https://en.wikipedia.org/w/index.php?title=Benford%27s_law&t...
Ask HN: Benford's Law and 2020 Presidential Election
11–16 of 16 posts
Re: Ask HN: Benford's Law and 2020 Presidential Election
#12I used to be a data analyst in a fraud investigation team and we would often use Benford’s Law to figure out whether a nominal account in a general ledger contained dodgy transactions. It’s a useful tool but needs to be used in the context of a wider investigation. If the graphs in the image are accurate then that’s a pretty big red flag that warrants investigation but I don’t know exactly the context of the graphs or how they were created - that’s pretty important to ascertain. But yes... definitely worth further further investigation!
Re: Ask HN: Benford's Law and 2020 Presidential Election
#13Perhaps the fraud can never be proven, but there's no doubt for me that it happened.
Maybe the Benford anomaly is the confirmation of something strange that can be observed in these images: https://twitter.com/daphnechen_/status/1324014079061745674
For four years, despite all the lies, the attacks on Trump's dignity, the massive and ferocious media brain wash, Trump got eight millions more votes. Trump got between 15 to 30 thousand people at each rally during the campaign, Biden could not even join 30. Let's see if the people seeking truth can prove what really happened.
Re: Ask HN: Benford's Law and 2020 Presidential Election
#14Here's a new paper from Walter Mebane (University of Michigan Political Science and Statistics) on inappropriate applications of Benford's Law to the 2020 election: http://www-personal.umich.edu/~wmebane/inapB.pdf
Re: Ask HN: Benford's Law and 2020 Presidential Election
#15There's some real questions that should be answered, no matter how you lean. The following link was shared to me giving a quick summary of this: imgur.com/a/gxuVAxc I mean if everything's fine then somebody should be able to explain why this isn't an issue, right? And there's also the infamous and mostly censored "4am vertical spikes": imgur.com/a/TwZqBQ7 I just want an explanation without being censored. These are r…
You are not the only one. I'm curious as well. I always expect a few statistical oddities per election year, but there was a whole lot more than I usually recall hearing about this year.
Here's a new paper from Walter Mebane (University of Michigan Political Science and Statistics) on inappropriate applications of Benford's Law to the 2020 election: http://www-personal.umich.edu/~wmebane/inapB.pdf
Re: Ask HN: Benford's Law and 2020 Presidential Election
#16Couldn't logistical circumstances of vote counting explain the oddities we see just as well? I don't think this has to mean the election was manipulated.
I think if there is doubt, a recount should be possible. The same scrutiny was allowed when the allegation of Russian influence was made. But until that is done, maybe conceding to the preliminary winner would be a good cause of action. Because I think the accuser has to prove manipulation.