Chances are if you are reading this blog, you know who Nate Silver is. Yeah, yeah, the guy who correctly predicted the last two presidential elections. The cover model for Nerd Weekly. (I mean that as a compliment.) Silver’s prediction methods on his New York Times Blog, Five-Thirty-Eight, gave liberals great comfort during the post-debate days of 2012, and apparently, they were all-out ignored by conservatives who believed a win was within grasp when it apparently never was.
So just before the election, Nate Silver released his book, “The Signal and the Noise” which is subtitled “Why So Many Predictions Fail–But Some Don’t.” The book was buzzed about and covered nicely, but then the election happened, Silver was nearly perfectly correct in his prediction, and now the book is in full bloom as a bestseller.
I don’t pretend to understand all of what Silver writes about in “The Signal and the Noise,” but I found it to be an enlightening look at predictions, data, history and politics. Silver opines that we aren’t very good at predictions, but places his stock in systems that help improve our ability to predict. He cites the Bayes Theorem, which is a formula that aids in prediction-making. There are tons of fascinating examples in the book–about baseball, politics, earthquakes, climate change, and my favorite chapter, gambling. Silver points out what is signal (the truth) and what is likely noise (what distracts us from the truth).
I read this a few months ago and the piece that stuck with me most was related to business, which is unusual because there’s so much politics in the book. If you work someplace that uses big data to make decisions, then you should read this book. It’s a cautionary tale for how to use that data effectively, how much to rely on it, and how skeptical to be of its ability to predict the future. As Silver points out early in the book, “We think we want information when we really want knowledge.” That’s truth. What’s also true is that we often mistake information for knowledge. We convince ourselves that because we possess a ton of data, we must be smarter in our ability to predict what it is coming. But that data inherently includes our own biases (another key theme in Silver’s writing) and is often incomplete or the wrong data set.
The baseball and politics chapters are strong too, and left an impression on me. Much of what Silver writes about “Moneyball” makes this book a natural companion to Michael Lewis’s work.
I feel like I’m phoning this one in. I’m reading Meacham’s “Thomas Jefferson” right now, and it’s capturing nearly all of my reading attention. I’m writing this now because I wanted to clear out the queue and now that’s done. On to 2013.
Thanks for reading.