Edit - Skeptics Stack Exchange

You are not logged in. Your edit will be placed in a queue until it is peer reviewed.

We welcome edits that make the post easier to understand and more valuable for readers. Because community members review edits, please try to make the post substantially better than how you found it, for example, by fixing grammar or adding additional resources and hyperlinks.

10

@AndrewGrimm I have no idea, wasn't part of the question. Either edit the question adding references to claims that it should be used, or ask a new question.
– BKE
Commented Nov 7, 2020 at 22:34
3

"For example, if all the precincts have around 800 voters and one candidate consistently takes 40-50% of votes, then it is expected, that the most frequent first digits will be 3 and 4." To follow up on this: if one candidate trails behind in a lot of precincts but dominates in a few others - say, takes from a range of 25-95% of votes - then that will produce a wider spread of first digits, but still tending to avoid 1 (for these numbers). Whereas the other candidate would have a spread of 5-75%, thus having results that span more than an order of magnitude, making the law hold better.
– Karl Knechtel
Commented Nov 8, 2020 at 5:42
3

You can reduce the scale of the data needed to apply Benford's law by using a smaller logarithm (i.e. instead of base 10, what if we did base 5?). I'm curious to see the same analyses applied in these cases too.
– Anon
Commented Nov 8, 2020 at 9:25
7

Chicago also provides the same data for previous elections. we see the same pattern in 2016. Moreover, I did the same analysis for the German federal elections 2017 and there Benford's law (1st digit) is simply not applicable at least for the major parties. I guess, this also relates to precincts having roughly the same size.
– Quickbeam2k1
Commented Nov 8, 2020 at 19:01
3

This answer gets to the heart of the question, but it may also be worth adding that it is pretty common for observed digits to not follow the Benford's Law distribution exactly. But if that's the standard that you want to use then comparing levels of conformity doesn't make sense. The Biden-Harris Benford graph isn't particularly Benford-conforming, but neither is the Trump-Pence graph. If neither conforms, then neither is really "better" than the other. And that's assuming that Benford applies well to vote counts, which as in the answer is unclear.
– Upper_Case
Commented Nov 9, 2020 at 18:07

| Show 10 more comments

Correct minor typos or mistakes
Clarify meaning without changing it
Add related resources or links
Always respect the author’s intent
Don’t use edits to reply to the author

create code fences with backticks ` or tildes ~
```
like so
```
add language identifier to highlight code
```python
def function(foo):
print(foo)
```
put returns between paragraphs
for linebreak add 2 spaces at end
_italic_ or **bold**
quote by placing > at start of line
to make links (use https whenever possible)
<https://example.com>
[example](https://example.com)
<a href="https://example.com">example</a>

formatting help »
answering help »