The Signal and the Noise (Nate Silver, 2012)

I've been keeping tabs on Nate Silver since discovering his blog FiveThirtyEight.com back in 2008. As he rightly states in one of the footnotes, it was a source of comfort for an Obama loyalist like myself, knowing that no matter how bad a day my candidate was having, he was winning the race. In fact, in 2012 Silver accurately predicted how every state would go, including Florida, besting is own impressive 2008 forecast. How he was able to do it is at the heart of this book. For the first time in, well, forever, having too much data is more problematic than having too little. The effect of this is that we are losing the ability to see the signal through the noise.

Chapter by chapter, Silver explores a wide variety of topics, ranging from baseball and Texas Hold 'Em to climate change and presidential elections, applying his experience and wisdom to each area. Frequently, the problem is that we are exposed to "hedgehogs" who make a living off making bold predictions, most of them completely wrong, because it gives them tremendous exposure and media credibility regardless of their track record. Meanwhile "foxes" have far less media appeal, but, thanks to their acknowledgement of a great number of variables, tend to make more accurate forecasts. Unfortunately, all of those variables makes them acknowledge a number of possibilities, rather than one glamorous certainty, so they aren't the profound pundits on TV. Silver tells us: "Be foxy."

I don't visit FiveThirtyEight.com (now a proud part of the New York Times conglomerate) much anymore, but has expanded into an impressive operation, with a full media team working on everything from sports to foreign affairs, sifting through the data and applying a very mathematical approach to their analysis. After reading this book, I wholeheartedly agree that more probability and statistics need to be in math curricula, on an equal footing with algebra and geometry. They certain help us gain a more profound sense of the data-soaked world around us, finding the signal in an increasingly noisy environment.