A compilation of comics explaining statistics, data science, and machine learning

Many (academic) talks or lectures I attend nowadays motivate the central question with a (sometimes humorous) comic strip, perhaps influenced by the fact that there’s always a relevant xkcd; unfortunately, everyone seems to have converged to using the same (small) set of comics, and I’m no exception.

Here, I try to rectify the issue by compiling a larger set of comics that you can use instead. I limit it to comics that explain some relevant concept. There are many more comic strips that mention, use, or relate to these topics.

Note: I didn’t get this list by myself; I used both existing compilations and crowd-sourced more from friends.

Note 2: Of course, this’ll become out-of-date immediately. Feel free to send me comics or link to them through the comments.

Note 3: The categorization below is rough. Of course many of the categories/comics overlap.

Note 4: Medium’s hot-linking of images doesn’t seem to work very well unfortunately. Links to original source included in caption.

Prediction, summarization, and interpretation

https://xkcd.com/1725/ — Note: I’m not a huge fan of this one; regression with low R² has value in certain cases.


https://xkcd.com/1478/. Note — if it isn’t obvious: this is not a guide.

Conditional Expectation