Book of the week: The Signal and the Noise: Why So Many Predictions Fail—But Some Don’t by Nate Silver

(Penguin, $28)

“We live in an era of ‘Big Data,’” said Burton G. Malkiel in The Wall Street Journal. As statistics whiz Nate Silver notes, our digital age produces 2.5 quintillion bytes of new information each day. It’s thus amazing that in certain fields, professional prognosticators remain no better than a coin flip in predicting future outcomes. Enter Silver, who shot to fame when his polling website FiveThirtyEight.com proved startlingly accurate in its predictions about the 2008 presidential election. His new book offers a whirlwind tour of the art and science of forecasting everything from hurricanes to poker hands, and offers tips on how to separate meaningful patterns—the “signal”—from mere distractions. You might not agree with all his conclusions, but Silver’s “breezy style” makes statistics analysis engaging.

Silver is “aiming for something bigger here” than a handbook for handicapping future elections, said Alex Koppelman in the Los Angeles Times. Apparently, “he wants to change how we think about predictions in every aspect of our lives,” to consider the imperfections in our thinking so we can be better at everything from planning around the weather to noticing a global housing bubble. Owing to a lack of strict focus, the book “never quite reaches its potential.” Yet it “sparkles” often enough that it could become the kind of culture changer Silver hopes it might be. It’s “easily” the best book about statistical forecasting yet written for the average reader, said Walter Hickey in BusinessInsider.com. Among the “cool things” it reveals: The computer models used by the National Weather Service generate remarkably accurate predictions, but local TV meteorologists still can’t be trusted. The main problem? They consistently inflate the chances of rain so they won’t be blamed by their viewers for any surprise downpours.

Subscribe to The Week

Escape your echo chamber. Get the facts behind the news, plus analysis from multiple perspectives.

SUBSCRIBE & SAVE

Sign up for The Week's Free Newsletters

From our morning news briefing to a weekly Good News Newsletter, get the best of The Week delivered directly to your inbox.

From our morning news briefing to a weekly Good News Newsletter, get the best of The Week delivered directly to your inbox.

Sign up

Yet the local weatherman looks like Nostra-damus compared with some political pundits, said Jordan Ellenberg in The Boston Globe. Studying predictions made on TV’s The McLaughlin Group, Silver found that the show’s panelists are right less than 50 percent of the time. And “boy, does Silver do a number on economists” for failing to predict the 2008 financial crisis. But he never suggests that their failure, or similar mistakes, can be avoided through “a rigorous adherence to cold, objective math.” Surprisingly, he believes that human intuition has an indispensable role to play, providing a check on the math just as the math provides a check on our hidden biases.