“I don't trust linear regressions when it's harder to guess the direction of the correlation from the scatter plot than to find new constellations on it.” https://m.xkcd.com/1725/ pic.twitter.com/A6VphYvQxu
You can add location information to your Tweets, such as your city or precise location, from the web and via third-party applications. You always have the option to delete your Tweet location history. Learn more
“I don't trust linear regressions when it's harder to guess the direction of the correlation from the scatter plot than to find new constellations on it.” https://m.xkcd.com/1725/ pic.twitter.com/A6VphYvQxu
Of COURSE xkcd said it better
xkcd's law
I'll add Klinger's law: "if it's an average value it's most likely used to hide the problem"
If you're really interested, there is an entire literature around the biases in 'ocular regression'.
Good sniff test. How would that translate to data in more than 2 dimensions?
Fixed the line of best fit based on naked eye expectations - surprisingly i got quite a different answerpic.twitter.com/6xnnKDb7FB
I just had exactly the same insight 3 or 4 days ago. In my case it came from Taleb's ideas.
As my economist friend said: “Technical Analysis—AKA drawing random lines in paper”
Any regression graph without the R2 or p-value listed is worthless, even if it visually “looks okay”
The obvious regression line, to me, is straight up
The mathematical model they build is surprisingly deep and interesting, but this scatter plot is... not well chosen.
Omg that regression line 
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.