Interview with Nate Silver about 538

Here is an excellent interview with Nate Silver about his new project, interesting throughout.  Here is one bit:

People also think it’s going to be a sports site with a little politics thrown in, or it’s going to be a politics site with sports thrown in. I understand why people say that — what we’ve been known for, plus ESPN, plus ABC News. But we take our science and economics and lifestyle coverage very seriously.

Some of the interview made me a little nervous.  He inveighs against New York Times Op-Ed columnists (juicy passages, click on the link if you wish), but their knowledge is more synthetic and also more novel than I think Silver recognizes.  I am not sure why “predictable” points of view are necessarily less likely to be true, or less likely to be important, even though they are (to me as well) less interesting to read.

Here are some more words from Silver:

We’re not sociopaths, which means that we look at the world and have opinions. But we’re not trying to do advocacy here. We’re trying to just do analysis. We’re not trying to sway public opinion on anything except trying to make them more numerate. I would say we’re not going to do a ton of public-policy coverage. We think that space is pretty rich now with competition. I also think with something like the health-care bill, it’s going to take years to get a good sense of how that’s working and how it’s affecting the market.

That too makes me a little nervous.  For instance there is the risk of assuming that the most important issues always or usually involve measurement.   Technocrats who rail against the ideologies of others are often the most ideological people around, even if their biases do not line up with the political spectrum in the usual manner.  Is there really such a thing as “just do analysis”?  Is it not better to make the underlying value presuppositions more explicit?  And why the knock at people who don’t have opinions about public affairs?  They’re not sociopaths, and frankly I’m not even fully comfortable with a blanket condemnation of sociopaths.

Earlier today I was reading John Hauer’s excellent The Natural Superiority of Mules.  It is a deliberately species-ist book, without a shred of objectivity, and the title reveals the blatant biases of the author.  The book has data, but is not data-driven.  It is “advocacy of mules driven.”  Get the subtitle: “A Celebration of One of the Most Intelligent, Sure-Footed, and Misunderstood Animals in the World” (eyes roll).  Yet I learned a great deal from it, and I will read any web site that can do as well.