Earlier today, it was claimed that “to a statistically significant degree” that James Harden’s performance declines in cities with high-rated strip clubs.
A reddit user found a correlation between James Harden’s lack of performance on the road and the quality of strips clubs in that city…. pic.twitter.com/xWg0awhp6K
— Darren Rovell (@darrenrovell) November 7, 2019
There are so many things wrong with this study I don’t know where to begin. But here’s a few:
- The scatter plot looks like a poorly decorated Christmas tree.
- The correlation is said to be “moderate to strong” when it should really be classified as nonexistent.
- Strip club ratings were based on Google reviews.
To illustrate how poor of a correlation this is (R2 = 0.2094), I have shown that I can derive a better correlation (R2 = 0.25536) between James Harden’s 30-point games over the last four seasons and the number of goats in the state where the game was played.
NBA statistics were taken from basketball reference.
Goat statistics were taken from here.
There will be no slandering of James Harden on my watch. #RunAsOne