The Signal and the Noise: Why So Many Predictions Fail but Some Don't
Amazon Best Books of the Month, September 2012:People love statistics. Statistics, however, do not always love them back.The Signal and the Noise, Nate Silver's brilliant and elegant tour of the modern science-slash-art of forecasting, shows what happens … see full wiki
Confident predictions are not the same as accurate predictions...
May 20, 2013
Nate Silver is the current rock star of statistics and predictions based on the overwhelming accuracy of his forecasts in the 2012 election results. His book The Signal and the Noise: Why So Many Predictions Fail -- but Some Don't does a good job in explaining the fundamentals of statistics, probability, and predictions. If society understood these concepts, we'd spend far less time arguing from polarized positions "supported by facts"...
Contents: A Catastrophic Failure of Prediction; Are You Smarter Than a Television Pundit?; All I Care About Is W's and L's; For Yours You've Been Telling Us That Rain Is Green; Desperately Seeking Signal; How To Drown In Three Feet of Water; Role Models; Less and Less and Less Wrong; Rage Against The Machines; The Poker Bubble; If You Can't Beat'em...; A Climate of Healthy Skepticism; What You Don't Know Can Hurt You; Conclusion; Acknowledgments; Notes; Index
Silver seeks to explain how we often mistake noise, an overwhelming flow of information, for signal, the pieces of information that are most important in terms of forming decisions. Because of the high visibility of pundits and "experts" on various media channels, we get led astray by confident predictions as opposed to accurate predictions. You only have to go back to the 2012 election season to watch how experts espoused sure bets, only to be completely wrong when the results were in. You can see the same thing on shows that talk about the stock market or sporting events. The talking heads go from prediction to prediction, and no time is ever spent going back to see how accurate they were. So long as they're confidently wrong, people will listen. Instead, Silver works towards a view of determining the probability of something happening, and then acting on those probabilities.
The Signal and the Noise is a book that you can read at various levels. The first part of his chapter examples can be followed by readers at almost any level. He does go into more depth and detail related to the science of each example, and it's easy to get a bit lost if you don't have the same background as he does. Still, the overall message (and signal) is easy to grasp, and it's one that can change the way you view the world around you. Now, if we can only have someone analyze all the network experts on sports and show them up for the talking heads that they really are...