2020 Election Statistical Anomaly?

[necrovore]

Quote

As you said, the fact you are hungry is not an ingredient for cooking. Yeah, hungry is the analogy with intuition, but you seem to miss that cooking is the analogy with reasoning about facts. Intuition is not an ingredient for reasoning about facts.

I don't see why you think I missed this. At the end, I did say, "if you reason correctly, you should have an argument that doesn't depend on the emotion at all, but on evidence and logic."

43 minutes ago, Eiuol said:

Even then, I don't see any confidence intervals or standard error on the github graphs...

Election officials don't report confidence intervals when they report election results.

Edited November 10, 2020 by necrovore

[Eiuol]

23 minutes ago, necrovore said:

Election officials don't report confidence intervals when they report election results.

I made a silly mistake when interpreting, so nevermind about that sentence. 

 

 

[Eiuol]

I wanted to add for clarification before:

I get why there are no error bars for the distribution of first digits (the numbers aren't approximations, but exact numbers). The problem is that those graphs don't really violate your expectations, not even the Chicago one. Chicago is strange even Trump. The 9 is so high that I could even suspect that Trump is more likely to have committed fraud than Biden. 

[necrovore]

5 hours ago, Eiuol said:

The problem is that those graphs don't really violate your expectations, not even the Chicago one. Chicago is strange even Trump. The 9 is so high that I could even suspect that Trump is more likely to have committed fraud than Biden. 

I do think it would be better if they had a "weirdness" score.