The Drunkard's Walk

How Randomness Rules Our Lives

[Amazon Link]
(paid link)

I previously enjoyed Leonard Mlodinow's memoir of his postdoctoral time at Caltech, Feynman's Rainbow, quite a bit. And I'm always up for a breezy popular science book every few months, so…

It's a very entertaining look at probability and statistics, branching into statistical mechanics and chaos theory near the end. Like most books of this genre, the math is kept to an absolute minimum. Given that limitation, the breadth of topics covered is admirable. (The famed Monty Hall Problem? Check, it's here.) Mlodinow also does the history of the appropriate bits of math, with many yarns about the characters involved long the way: Pascal, Bayes, Gauss, Cardano,… And Bernoulli! Or, actually, numerous Bernoullis. I'm ashamed to admit that I've always thought there was one guy named Bernoulli.

Adding to the fun: Mlodinow is not averse to making cheap-but-funny jokes every so often. Not laugh-out-loud, but deserving of the occasional snort or moan.

If I had to pick a nit: nobody who's looked at the issue doubts that math, especially probabilistic math, can be counter-intuitive. Casinos rely on this, for example. And (ahem) see that previously-mentioned Monty Hall problem. But (arguably) Mlodinow overdoes this; I'm not sure he's seen a psychological study he didn't like. In at least one case, that leads him astray, on page 161:

In fact, in recent years psychologists have found that the ability to persist in the face of obstacles is at least as important a factor in success as talent.

… with a footnote to a 2005 Psychology Today article, "The Winning Edge." This is that "grit" thing that was popular back then. But: only a few weeks back, I'd read The Quick Fix by Jesse Singal. And that book had a chapter that pretty convincingly depicted "grit" as an overhyped psychological fad. (If you don't want to get Singal's book, here is an article he wrote on the topic.)

So this kind of wrecked the bits of the book where Mlodinow reports psychological "studies" uncritically. I kept wondering: has this research been replicated?

Never mind. It's a fine book, and ("odds are") you'll learn a lot.


Last Modified 2024-02-15 9:34 AM EDT