YOU CANNOT USE BENFORD’S LAW TO DETERMINE ELECTION FRAUD. Though it might highlight or give small indicators to fraud, it cannot be used as direct evidence. Shit light this is going to make us look stupid.
Edit:
A short 17 min vid and a paper that addresses this very issue.
Its used commonly to help detect fraud, of course its not 100% proof, its a statistical analysis, but combined with evidence it helps paint the story. Its smoke, and leads us to the fire. It does not make us look stupid if you have an understanding of what it is...
YOU CANNOT USE BENFORD’S LAW TO DETERMINE ELECTION FRAUD. Though it might highlight or give small indicators to fraud, it cannot be used as direct evidence. Shit light this is going to make us look stupid.
Edit: A short 17 min vid and a paper that addresses this very issue.
https://www.youtube.com/watch?v=etx0k1nLn78
https://www.cambridge.org/core/journals/political-analysis/article/benfords-law-and-the-detection-of-election-fraud/3B1D64E822371C461AF3C61CE91AAF6D
Its used commonly to help detect fraud, of course its not 100% proof, its a statistical analysis, but combined with evidence it helps paint the story. Its smoke, and leads us to the fire. It does not make us look stupid if you have an understanding of what it is...
Not it doesn't paint a story or smoke. It makes us look stupid and undercuts all the valid vote fraud evidence.
Benford's law is not valid for datasets with structure. Benford's law is not valid for datasets that don't span orders of magnitude.
Voter Precincts have a set range of sizes based on logistics and geography, it is a complete misapplication to use Benford's law on vote counts.
When it is done for areas where no vote fraud is alleged Benford's law is also violated (because it is invalid for this situation)