1. February is the most frequent birth month for Best Director winners, the second least frequent for Best Actor winners, and the least frequent for Best Actress winners.
2. June is the most frequent birth month for Best Actress winners, the second least frequent for Best Director winners, and the least frequent for Best Actor winners.
3. April is the most frequent birth month for Best Actor winners.
Does this reflect the general population for birth months though?
Excellent example of bad data science: no meaningful/significant differences due to low frequencies, failing to account for the same person occurring multiple times..
8 comments
Some patterns emerge:
1. February is the most frequent birth month for Best Director winners, the second least frequent for Best Actor winners, and the least frequent for Best Actress winners.
2. June is the most frequent birth month for Best Actress winners, the second least frequent for Best Director winners, and the least frequent for Best Actor winners.
3. April is the most frequent birth month for Best Actor winners.
My[ previous post ](https://www.reddit.com/r/dataisbeautiful/comments/1j1s80p/oc_movies_released_in_december_are_way_more/?utm_source=share&utm_medium=web3x&utm_name=web3xcss&utm_term=1&utm_content=share_button)was about how Best Picture Oscar winner movies are most frequently released in December. There was a good explanation for this.
But why anomalous pattern exist for birth month is not clear to me.
PS: From my Medium article : [https://medium.com/@RajLnk/the-oscar-anomalies-f163b94b1c5d](https://medium.com/@RajLnk/the-oscar-anomalies-f163b94b1c5d)
Unfortunately this data isn’t significant enough to prove a relationship. Still cool though
It’s probably just random chance, and “likely” only applies to the past, not the future. I don’t think there’s anything deep here
So there’s equal distribution accounting for small sample size and has no impact. Got it!
Is there some astrological shit in here?
This screams 2-3 favorites in a bucket are skewing results from winning multiple times.
[John Ford](https://en.wikipedia.org/wiki/John_Ford) won 4 times for Best Director and was born in February.
Does this reflect the general population for birth months though?
Excellent example of bad data science: no meaningful/significant differences due to low frequencies, failing to account for the same person occurring multiple times..
Comments are closed.