8722
Comments (685)
sorted by:
You're viewing a single comment thread. View all comments, or full comment thread.
1
Magister_Caeli 1 point ago +1 / -0

Would be interesting to see that comparison

2
2terms3scoops 2 points ago +2 / -0

Here is the resultant chart, and the source code if you're interested.

1
Magister_Caeli 1 point ago +1 / -0

Can you explain why Benford's law was used to determine other elections as fraudulent, i.e, India in 2009 I think? Did they use a different number that wasn't votes per precint and was spread across more magnitudes?

2
2terms3scoops 2 points ago +2 / -0

I’m not familiar with the exact circumstances but I would assume you’re essentially correct, they could have used votes per pricinct however if the precincts varied significantly in size and there were a large number of them, I’m by no means a statistician but by manually scrolling through some of the milwuakee wards they all seem roughly the same size, in the 500-1000 votes per ward range. That’s why Biden/Hillary never had many leading ones, few wards are large enough to give them 1000 or more votes, and they almost always got more than 100. On the other hand trump had a lot of 100 vote wards and much fewer closer to 5-600. Trumps also likely fits benfords better because his totals do scale more orders of magnitudes since there were actually a significant amount of wards he got less than 100 votes in.