Thoughtbasket

I’m a little late in getting to this, but a semi-recent suite of letters to the editor regarding Malcom Gladwell’s New Yorker piece on basketball’s full court press provided me with a reason to write something I’ve been stewing over for years: Malcom Gladwell is massively overrated. I have been disagreeing with him for years, since before he published The Tipping Point, back when he was just a New Yorker staff writer.

The letters about the full court press, which sadly are not on the New Yorker website, generally press [heh heh] on the theme that Gladwell’s conclusion was superficial if not downright specious. This makes sense to me, since when I read his article, it clearly seemed to be wrong. And that has been my problem with Gladwell all along: too often I feel like his conclusion doesn’t make sense.

I should state here that I don’t want to fully attack Gladwell. Sometimes his conclusions are correct. And in all cases I think he is a truly talented writer with an amazing skill at explaining complex ideas in clear and concise language. He just sometimes leaps to unwarranted conclusions.

Here is how the typical Gladwell article works: he presents a few facts, which he then links into some sort of “surprising” conclusion (e.g. underdogs should always press or a few well-connected hipsters started the Hush Puppies trend) and then does a series of riffs on the implications of this conclusion. But his conclusion is based on some analysis he has done of those first few facts. My problem is that he never gives you the details of this analysis or how he did it. That means that you can’t tell if he’s right or not. He might have misread the data, or skewed it to fit his thesis, or just screwed the analytical pooch. You have to trust his analysis, and if you don’t, his whole article is meaningless.

The real problem here is that Gladwell’s analyses are sometimes wrong. As the letters indicate, his view of the press is flawed. Of course any new strategy can prove effective for a while, but fundamentally all a press does is move the locus of competition from the basket to the backcourt. Once teams get used the press, the good teams will break it the same way they can outscore the bad teams once under the basket. In addition, as the letters pointed out, it is often the favorite that presses, leveraging its physical advantage.

Similarly, Gladwell’s first — and still most famous — analysis, in The Tipping Point, is equally problematic. Recent research in trends has shown that that randomness and social dynamics provide a full explanation for why some trends explode and some don’t. Gladwell’s whole structure of Mavens and Connectors is irrelevant. Check out Duncan Watts’ work on music popularity to see how wrong Gladwell really was. See links here, here, here, here and here.

Here is Watts from the NY Times magazine: “This means that if one object happens to be slightly more popular than another at just the right point, it will tend to become more popular still. As a result, even tiny, random fluctuations can blow up, generating potentially enormous long-run differences among even indistinguishable competitors.”

As statisticians always say, just because the data looks like a trend doesn’t mean the trend exists. Gladwell’s problem is that he likes to make a thesis out of a few data points, without doing the work to truly understand whether his thesis actually has causative properties.

Or, as MIT professor Ezra Zuckerman put it in a letter to the New Yorker regarding Gladwell’s piece on the Wall Street meltdown, “This is an interpretive leap drawn from two facts….But Gladwell’s logic is faulty.”

I want to emphasize again, however, that I am not fully anti-Gladwell. His writing is great, even if his analysis is sometimes flawed. And his review of Free, the idiotic book by Chris Anderson, is right on the money. So to speak.