Nate, Maybe The Midwest Is More Livable, No?

So, Nate Silver, current Grand Poobah of All Statistical Knowledge, has moved to New York, and in this week's issue of New York magazine he takes his data scalpel to the old New York argument of what neighborhood is the best place to live. The traditional New York answer is "Mine!" and everybody finds some way to work backwards and justify that. Clearly we could do better with real, objective data, right?

Well, no. I am a huge, huge fan of data driven everything in general and a big fan of Nate Silver and his website, FiveThirtyEight.com, in particular. But when it comes to reducing all the subjective factors in this to one objective measure, Nate's results are (objectively speaking!) insane. Take the no. 5 neighborhood, Greenpoint. I live just a few blocks from there. Look, Nate, a friend of mine recently was suckered into subletting a place in Greenpoint. Her block was one of the most depressing places I'd ever seen in New York, and there are many dreary places in New York.There is "access to transportation," but there's only one subway line, the G. The one line that doesn't go into Manhattan.

On the other hand, Jackson Heights, where I grew up, is rated at number 15. I'm not advocating anyone move to Jackson Heights. Really. I don't live there myself now. But Jackson Heights coming in below average for restaurants? Huh? Again: huh? Jackson Heights is where India and Colombia share a border. Trust me, Nate, you can find a place to eat there. I promise.

Obviously everyone in New York will have their opinion on this and about a million other things (For instance, the schools ranking: it doesn't matter the same way as in other places, because New York high schools use a hugely complicated citywide application and matching process). But there is one more general point worth making here. In general, the "surprise" of this story was that neighborhood like Astoria (number 3) and Greenpoint did so well. But this is almost entirely a consequence of including the cost of housing as the single biggest factor--by far--in the ratings. The problem here is that when you combine scores for desirability and price into one number, you will always get "surprising" results near the top because they are cheap. So essentially what you wind up with not the advertised list of "most livable" neighborhoods, but a list of "the best values."

Invariably, lists of this sort trumpet their surprises, which mostly aren't really big surprises at all. When you consider the price, yes, guess what? The Camry will do better than the Lexus or the Mercedes. But that doesn't tell you anything about what's the better car. It just tells you that you get diminishing returns as you move up the price ladder. We knew that already.

The ultimate irony here, though, is that every year similar lists of "best places to live" come out in outlets such as Money magazine, and--unsuprisingly--New York city never seems to do very well. Just like Nate's list of neighborhoods, those lists put a premium on "affordability" and invariably are filled with perfectly decent and unexciting places with nicely priced homes. Many of those places are wonderful, and I have no doubt that they are great values. Most sane folks would choose to live there, not in a place like New York, which (a) is a terrible value, (b) encourages all sorts of status anxieties, and (c) is basically full of crazy people. And nonetheless there are people who choose to live here despite New York's obvious objective deficiencies. Including Nate Silver himself. Which, when you think about it, pretty much disqualifies Silver from doing this sort of article in the first place, because nobody who chooses to live in New York can make any reasonable claim to caring all that much about what places in the world are "objectively" livable.