March 30, 2014

#FiveThirtyEight — Nate Silver as Biggus Dickus? (updated)

Nate Silver, posing for a cheap knockoff of Rodin's "The Thinker."
Photo from The Guardian.

I've already written one blog post about how I'm not a total fan of Nate Silver's new incarnation of FiveThirtyEight as part of the ESPN "stable." My critique, like many others, and per the theme of this blog post, was about "data," namely, in this case, his hiring of a known climate change skeptic.

Update, March 28: Silver claims to be listening to some of his critics, namely on that climate change piece.

But, is this real, or is it a head fake by Nate Silver on his response to criticism of the Roger Pielke Jr. story? Per his claim that Pielke is not a "skeptic" (not even his knowledgeable critics have called him a "denier," Nate), and Silver not saying who is the the person he has commissioned, I'm not holding my breath. Given Pielke's reaction, Silver's "commissioned response" had better be damned good. (end update)

And, hey, this side of Neil deGrasse Tyson and brilliantly hyped overmarketing, where would you go but Bristol, Conn., to buff your buttocks?

Turns out I'm not alone. Far from alone.

A leading pooh-pooher? Paul Krugman, taking a break from economics, but not taking a break from data, in accusing Silver of, in part — taking a break from data.

That said, I said I'm not alone, and far from alone. Krugman himself links to this piece from Noah Smith in his first sentence, and quotes, as shall I: 
In sum, this so-called “data-driven” website is significantly less data-driven (and less sophisticated) than Business Insider or Bloomberg View or The Atlantic. It consists nearly entirely of hedgehoggy posts supporting simplistic theories with sparse data and zero statistical analysis, making no quantitative predictions whatsoever. It has no relationship whatsoever to the sophisticated analysis of rich data sets for which Nate Silver himself has become famous.

The problem with the new FiveThirtyEight is not one of data vs. theory. It is one of “data” the buzzword vs. data the actual thing. 
As I said? Bristol, Conn. and marketing. (Go to Urban Dictionary under "Simmonsfication," for example. Or read some of my blog posts about him.)

Don't stop there, though. Smith does a piece-by-piece analysis of several different FiveThirtyEight pieces, and compares them with outside writing from places like Atlantic. His thoughts, right before those two graphs above?
Looking at a bunch of other posts, you can see that this is par for the course. And not one of the posts attempts a single quantitative prediction, which is what Silver has famously thrilled the world by doing in the past.
In short, playing with numbers isn't the same as analysis. (This said, why does Noah say he's still a fan of Silver's, other than him drinking too much of Silver's marketing Kool-Aid?)

So what does Silver do? Shoot the messenger. And, other than his cute graph, he does so without any analysis. As for Krugman's latest, the link up top? He, like Noah Smith, critiques something specific — Silver's Obamacare coverage.

Update, March 30: That said, there's other people weighing in on how "data" isn't "analysis." Latest? And simply crushing Silver, here's Massimo Pigliucci.

Piglucci starts here:
Like Wieseltier, I am aware that the state of public intellectualism and opinion making isn’t exactly without problems. But, with Wieseltier, I find it oddly naive of Silver to talk as if “ideological priors” (otherwise known as beliefs about the world) weren’t inevitable in anyone (including Silver), and — within limits — were not actually a good thing.

Moreover, as a fellow Bayesian, Silver ought to know that his own analogy is ironically flawed: in Bayesian analysis you always begin with priors, and the whole point is to revise those priors as new data comes in. That is, embedded in the very fabric of the Bayesian approach [4] is that you start with beliefs, you add data (collected on the basis of your beliefs!), and end up with (likely modified) beliefs. You just can’t take the belief components out of the analysis, it’s integral to it, and it’s both affected by the data one gathers and determines which bits of information “out there” actually get to count as data.
Sounds like Massimo is politely saying that Silver doesn't like that his ox is being gored.

And, in future reference to a piece by Leon Wieseltier, Pigliucci also seems to hint that Silver is missing the forest for the trees:
Wieseltier ends his piece by asking whether numeracy is truly the American public’s most pressing problem. Seems to me that a vibrant democratic discourse could use more numeracy among its participants, and Nate Silver has certainly contributed his share in acquainting people with the power of data crunching. But that’s peanuts compared to the hurdle of fostering critical thinking abilities without which no amount of data crunching will help move society forward.
Totally agreed. Unless an understanding of proper use of statistics is grounded in a broader proper critical thinking methodology, it's of relatively little value.

In his own piece, Pigliucci shows that by going next from Nate Silver to naive misuse of Ngram. He then asks, in a variant on a timeless classic, "Who's Assessing the Assessors?" (end second update)

Jonathan Chait has another interesting angle on the original Silver-Krugman brouhaha.
The real cause of Krugman’s disdain is the sheer ambition of Silver’s new venture. Silver’s great added value was to bring basic statistical literacy to the fields of political forecasting and sports commentary, which are dominated by old-line hacks who rely on horse sense and either disdain data in any form or use data very badly.

The new FiveThirtyEight tries to expand this revelatory contribution to other fields. The trouble is that many of those fields, like economics and climate science, already have real experts. Silver’s role, at least in its crudest form, represents the kind of autodidactism that Krugman rose to fame decrying. His war against Silver is nothing terribly new, but merely the return of an old love, or, more accurately, an old hate.
That said, Chait and Ed Kilgore, where I saw the Chait link, are both neolibs. If Silver has any political leanings, it's of the same. Krugman, who at one time was a full-bore free trader, isn't necessarily such a paleolib on all issues as some might thing.

So, my take? Some of this is not Silver thinking Krugman is jealous, or Krugman actually being jealous, of success, or platform locations. It's elbow-throwing in the neolib public square. Or elbow-throwing in the public marketing square. Alert ... alert ... Wodewick!

And, per my header and the GIF, we'll see at some point, in some way, just how thick or thin Silver's skin is, especially if his "brand" is slow out of the gate. Per my first blog post, beyond the Roger Pielke issue, I'm not that impressed so far. And, beyond neolib elbow-throwing, I'm sure this is an issue itself, per Chait. Silver figured that by the power of his "brand" alone, he could woo the masses in new fields, and he's simply ....


Nate, it's a word to learn to add to your vocabulary.

And, until you learn to admit you're wrong at times, get thicker skin, and grow a pair, the likes of Lisa Needham are going to bitch-slap you even harder than I did. Meanwhile, I remembered to put a "no follow" on Nate's link; I don't need to aid his marketing empire.


And, beyond the possibility of wrong, what if you're overrated, or at least if what you do is overrated?

Good one here. Note that, on things like Google Flu, the bulls-eye fallacy, the file-drawer fallacy, and other related issues are in play.

No comments: