Signal and Noise

The Bad: there's a lot of noise in Nate Silver's new book The Signal and the Noise: excessive name-dropping and celebrity sound-clipping, ugly charts with mislabeled axes or ill-chosen scales (linear when they should be logarithmic), and an author who too often stands between reader and idea. "I grew up in ...", "I visited ...", "I interviewed ...", etc. And there are typographical errors, some inadvertently funny, e.g.: "... the definition of rocket science is using relatively simple psychics to solve complex problems ...".

The Ugly: Silver's lack of subject-matter technical depth is evident in his discussion of computer chess and climate modeling. He too often relies on catchy anecdotes in place of statistics. Perhaps it's part of a conscious effort to popularize a technical topic, but it clashes with a core principle of the book itself. And there are some big blind spots, as in a discussion of online poker that totally ignores the possibility of deliberate cheating by the invisible House. What touching faith in human honesty!

But the Good: overall The Signal and the Noise does an excellent job of explaining Bayesian ideas, how to weigh evidence and update forecasts, and what are the common pitfalls that so-called "experts" fall into. The book's subtitle, "Why So Many Predictions Fail — But Some Don't", captures one of its strengths. Silver is no Phil Tetlock (though he quotes Tetlock, appropriately). But he does bring balance and critical thinking to a complex subject. And he offers sharp insights, for instance in citing Bill James on mature thinking and respect for others:

There are a lot of things I wrote in the eighties that weren't right," he told me. "The big change was my having children. I know it's a cliché, but once you have children you start to understand that everyone is somebody's baby. It is an insiders-outsiders thing. You grow up and these people are characters on TV or video games or baseball cards — you don't really think about the fact that these guys are humans and doing the best they can."

As are we all. While immersed in The Signal and the Noise I missed my stop on the subway — a signal that correlates with a well-written book. There are few surprises if you already know Bayes Theorem, if you've read Tetlock's Expert Political Judgment, and if you remember some rules of elementary probability. But whether you do or not, Silver is a fast, enjoyable, worthwhile read.

(cf. Introduction to Bayesian Statistics, Statistics - A Bayesian Perspective, ...) - ^z - 2012-10-25