Imagine a guy with glasses who used to model baseball stats and play online poker nailing the outcome of the 2012 elections. And when I say “nailing”, I mean that he correctly predicted the U.S. Presidential contest in every one of the 50 states (and nearly every U.S. Senate race, too). He even performed better than some of the most widely-used polling firms. Now imagine that he gives his thoughts on making these types of predictions. That’s exactly what Nate Silver does in his new book The Signal and the Noise.
I’ve worked in what’s now being called “data science
” for nearly twenty years. The title of Silver’s book – The Signal and the Noise – presents an important and sometimes overlooked part of this science. The “signal” is what we’re looking for in the data, and the “noise” is all the stuff in the data that gets in the way of what we’re looking for.
With companies like Facebook, Twitter, LinkedIn and Netflix delivering new products based on data, more attention is now being focused on what we can learn from all this new information. Political polling has been around for a while, but Silver managed to take these scientific principles and apply them in a new way, leading to results that are astonishing to most people. Silver ended up beating some of the oldest and most storied polling firms, such as Gallup, highlighting real biases in their polling (for example, Gallup performed poorly for the third straight national election
, and Silver noted that Gallup polls were biased toward Republicans by as many as 7 percentage points
In his book, Silver focuses on how these techniques can be applied in nearly every area of forecasting from baseball to poker to weather forecasting to earthquake predictions. He does focus on some technical things (such as Bayesian reasoning), but does a good job of not letting that get in the way of his story. More broadly, here are four points that I thought come out of Silver’s book:
- You can make better decisions if you get more information. Silver pooled together the predictions from over 20 polls into one larger and more accurate prediction for the presidential election. He also points out in his book how new information can be used to update our own predictions.
- There’s a lot we don’t know, but don’t let that stop you. When we get information and then make decisions, there’s always a chance we’re going to be wrong because we don’t know everything. Many people, including Silver, call this uncertainty. We have to learn to live with uncertainty, and make the best decision possible.
- Be aware of your own bias. Gallup didn’t recognize that their polling techniques led to errors in their own predictions. Now they have to regroup in the wake of Silver’s successes. We need to be open to the information that’s in front of us and be aware of what information we may not be getting.
- Humility leads to better decisions. If we are humble, we will be aware of any unintended biases and we will recognize the uncertainty before us. As it turns out, this is the best anyone can do in making decisions.
So if we’re honest with ourselves and the information we are gathering, we can make better decisions and learn from any missed predictions. In life, we have to be willing to learn, try, and learn again.
If you’re interested in learning more about Silver’s take on statistical reasoning, I would highly recommend reading his new book. I received the book as a Christmas gift from my wife, and I’m glad I got the chance to read it.
Question: Have you read The Signal and the Noise? If so, what did you learn from Silver’s book? You can leave a comment below.