Review: The Signal and the Noise
The Signal and the Noise: Why So Many Predictions Fail-but Some Don’t
by Nate Silver
published by Penguin
original publication: 2012
size: 5576KB (545 pages)
categories: nonfiction; statistics; politics & current events
simultaneous device licenses: six
part of the Kindle Owners’ Lending Library: no
suitability for text-to-speech: low, due to the number of illustrations
real page numbers: yes
Whispersync for Voice: yes (read by Mike Chamberlain)
“My hope is that we might gain a little more insight into planning our futures and become a little less likely to repeat our mistakes.”
Is it possible to predict the future?
Sure…predict, after all, means to “pre-speak”. Anybody can say what the future is going to be…the question is, how accurate are they?
No question, Nate Silver is remarkably accurate…for the second Presidential election in a row, Nate has called all fifty states. In this last one, Silver’s http://fivethirtyeight.blogs.nytimes.com/ also called almost every Congressional race.
However, this not a book about “how I did it”. It’s a book about prediction generally…yes, with advice, but there is as much caution about the limitations of the science, and the other inherent flaws (such as overconfidence) in the process, as there is endorsement of what Silver does. That’s refreshing, honestly.
There is little surprise here that pundits act like they are mad at Nate Silver. If the most probably answers can be derived objectively, it reduces the value of the supposedly superior individual. I think one of my favorites parts of the book was the analysis of pundits…and how getting things right didn’t tend to make one a more successful TV talker.
The book is also not just about politics, but about sports (especially baseball, an early success for Nate Silver), online poker, weather, earthquakes, and even terrorism.
Are you afraid of math? Well, the book is a bit mathy, although it’s written largely in a narrative style. There are a lot of graphs, with little explanation about how to interpret them. If you don’t know your Y axis from your X axis, that could be a problem.
Overall, though, this can be read by anybody, and there is sage advice here.
I don’t know that Silver is looking for any more challenges, but I’d love to see the statistician predict the movie box office for the summer. Like baseball, that’s a data rich environment. We have lots of stats about the people involved in movies (what their previous movies have done, what awards they’ve won, and so on), and there are many websites out there that both make predictions and solicit them from readers (giving some crowd sourcing for predictions). A lot of money is also riding on that, and it’s quite how profile. Just a suggestion…
Bottom line is that I recommend this one. We all make predictions every day: is that car going to change lanes? Do I need an umbrella? Will the boss come over to see if I’m getting my job done? Understanding the process by which we come to those estimates, and the alternative approaches out there, can be a real benefit…even if you don’t understand what every bubble chart means. ;)
This post by Bufo Calvin originally appeared in the I Love My Kindle blog.