As with many disciplines, polling is both an art and a science. Nate Silver, one of the most respected pollsters, and founder and editor-in-chief of FiveThirtyEight (an internet news site dedicated to using number analysis to tell stories), considers himself both a statistician and journalist. His goal is to build statistical models that tell a real story about what is happening or could happen in the world. He focuses on politics, but also covers sports, economics, science and culture.

One of Silver’s biggest principles is that making predictions or solving problems requires both numbers knowledge and domain knowledge – the perfect blend of hard facts, but also the training, education and ability to digest and analyze the facts to make a prediction or offer a solution.

The more I do this, the more focused I get. One misconception people have about the big-data-slash-analytics world is, if you know how to do analytics, you can solve any problem. I mean, not really, right? You still need a lot of domain knowledge about the field you’re trying to study.
~ Silver, in an interview with KK Ottesen for The Washington Post

I apply his point to today’s habits of “Google M.Ds.” We search something, read a few quick overviews, and suddenly are self-diagnosing everything from vitamin deficiencies to cancer. Silver reminds us that just because we have accurate data or information in front us, our understanding is likely incomplete if it isn’t our domain.

When people are smart but don’t necessarily have expertise, they tend to fill in the blanks with stuff that’s speculative… fill it in with a kind of B.S. that sounds good at a cocktail party, but which doesn’t necessarily hold up that well to scrutiny.

Regarding polling and elections, Silver correctly predicted the outcome of the 2008 presidential election in 49 of 50 states, in all 50 states in 2012, and he was closer than nearly everyone in his prediction for the outcome of 2016, giving Trump a 30% chance where most polls only gave him a 1-2% chance.

We were quite emphatic that the election was competitive, and that Trump had a chance. To us, the fact that Trump won this kind of narrow electoral college victory was exactly the scenario that our model identified as the reason he was more likely to win than people assumed. Because we’d done the historical work and the data work and the reporting work to actually kind of think through these things a little bit more deeply.

Silver has mastered how to merge the truth of data and the value of institutional knowledge – a clarity that allows him unparalleled accuracy in the news he shares.