Share a link to his benford analysis excel sheet, then I will tear it apart. I did the analysis myself 1 month ago. I fell into the exact trap that he did.
He would've realized his mistake if he did the analysis on all counties in the US.
100 years ago, a native american meant what you today call a white American.
Leftists are good at shifting words, right?
Why was a native American what we call a white American? Because the US was a white ethnostate and so-called whites were the natives.
The benford analysis from 1 month ago was not correct. I did an extensive benford analysis on every single state using MIT time series election data.
I'm sorry but benfords law only applies to certain data sets, not every data set.
Just do it yourself right now.
Do 2 normal distributions.
1 around 100.
1 around 500.
Then do benfords analysis on those 2 distributions of numbers.
It is not hard to test. The one around 100 will look like it fits but if you actually look closer, it doesn't fit at all. The one around 500 obviously doesn't fit and will have way more digits with 6, 5 and 4 than expected from benfords law for OBVIOUS reasons.
It is probably just bots.
The way to fix this is through machine learning.
I can help them if they want it but it would probably take a month of data gathering at least before seeing results + heavy moderation effort from both tdwin users and mods.
Then it is easy for the AI to filter 90% of the bots. It would massively improve over time too. The more data, the better it would work.
Peterson helped me, as an atheist, to realize the good of religion.