Benford's Law and 2020 Presidential Election

I don't know enough about Benford's law to draw conclusions, but is N=O(1000) a large enough sample size to expect it to apply? My intuition suggests that if county size does not span enough orders of magnitude then Benford's wouldn't apply because the distribution might lie in one of the "spikes" of the Benford curves, rather than reflecting the average. For example, if many counties had a population size starting with a "4", you'd expect more vote counts to start with "2" (presuming each candidate received close to 1/2 of the votes).

Hopefully someone with a stronger math background can expand in a more fluent way :)

There'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 reasonable questions to have answered, and until they are, I definitely have some doubt in what's being reported.

Couldn'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.

I 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!

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

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...

From a comment:

> This repo is attempting to apply Benford's Law to vote count distribution, so that's what actually needs to span multiple orders of magnitude. Precinct distribution is a factor in vote count distribution but it doesn't tell the whole story. I don't see the Milwaukee data in this repo, but take a look at the Chicago data: https://github.com/cjph8914/2020_benfords/blob/main/data/chi... Biden's vote totals are solidly contained within one order of magnitude, the 100-999 range. Trump's vote totals range from single digits into the hundreds, across three orders of magnitude. Jo Jorgensen is mostly in the 0-20 range, across two orders of magnitude.

Perhaps 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.