257
posted ago by dahdahdah_dahditdah ago by dahdahdah_dahditdah +257 / -0

Pedes, I have created a GitHub with reproducible Benfords Law analysis of Milwaukee vote anomalies. This is designed to be rapidly extensible. Please please please, suggest more suspicious datasets and I can rapidly add them to this analysis.

Here's the webpage:

https://voteanalysis.github.io/benfordslaw_election2020/

... and here's the github page so anybody can reproduce my results:

https://github.com/VoteAnalysis/benfordslaw_election2020

Comments (19)
sorted by:
You're viewing a single comment thread. View all comments, or full comment thread.
7
Diet_Loam 7 points ago +7 / -0

I think the page could use and explanation of what benfords law is and how it useful/relevant

4
Ghost0_ 4 points ago +4 / -0

https://www.youtube.com/watch?v=XXjlR2OK1kM

Here's a relatively entry level lesson on it.

TLDW: there's a counter-intuitive property of random number distributions. When you analyze the first digit of each number in the data set, you should find the number 1 most frequently at ~30%. Each subsequent number is less frequent. When analyzing the trump and biden votes from each precint, trumps vote totals follow this distribution. Biden vote counts do not, indicating human tampering.

3
Diet_Loam 3 points ago +3 / -0

I understand this, however, if I want to share this page broadly it would better if the explanation was included on the page. It’s a fondled idea and you only have so long to get peoples attention. So multiple links not ideal.

Thank you very much for writing out the great description!