Atlas Shrugged: The Mocking

Wednesday, May 4, 2011

Strawman vs. Krugman

We really don't need another post stating that Megan McArdle set up a strawman and knocked it down again because she couldn't stand to see praise of Paul Krugman. (A study of pundit predictions demonstrated that almost all of Krugman's predictions during a one year time period were correct.) While I admire single-minded persecution of a target, she's just embarrassing herself.

But there was a very interesting passage in the paper McArdle purported to refute.

While initially investigating whether higher levels of education and experience correspond to higher predictive accuracy, [Philip] Tetlock ultimately concluded that cognitive style was the most important influence on prediction accuracy. Using the framework derived from Isaiah Berlin’s essay The Hedgehog and the Fox that “hedgehogs know one big thing; foxes know many things, (Berlin, 3)” Tetlock separated experts into two groups with competing cognitive approaches to prediction and found “the hedgehog-fox dimension did what none of the other traits did: distinguish more accurate forecasters from less accurate ones in both economics and politics” (Begley, 45).

According to Tetlock, there are clear differences between hedgehogs and foxes. Hedgehogs “know one big thing” and “apply that one thing everywhere,” express “supreme confidence in their forecasts, dismiss opposing views and are drawn to top-down arguments deduced from that Big Idea”; they “seek certainty and closure, dismiss information that undercuts their preconceptions and embrace evidence that reinforces them” (Begley, 45). Foxes “consider competing views, make bottom-up inductive arguments from an array of facts, doubt the power of Big Ideas” and “are cognitively flexible, modest and open to self-criticism” (Begley, 45). Ultimately, “what experts think matters far less than how they think: their cognitive style” (Begley, 45). Tetlock found that foxes outperform hedgehogs in prediction accuracy in virtually all fields, across all time periods, and across the various levels of expertise.

Expert Political Judgment also considers two types of general skepticism found in theoretical literature about prognostication. Tetlock mentions both radical skepticism, which is the belief that nobody knows anything, and ontological skepticism, the idea that the nature of the world is unpredictable. Both are ideas well illustrated by Rick Perlstein, a contributor to The Nation. Perlstein’s disbelief in and distaste for prognosticators stems from a blend of radical and ontological skepticism. Perlstein’s article “Pundits Who Predict the Future are Always Wrong” goes so far as to “call punditry a sin” (Perlstein, 12). Perlstein dismisses forecasting because of ontological skepticism, alleging “history does not repeat itself, nor does it unfold in cycles” (Perlstein, 13). Not
only does Perlstein claim “there’s nothing you can really know about the future at all,” he warns that “to pretend therwise is an insult to democracy” (Perlstein). Appealing to radical skepticism and criticizing conventional wisdom, Perlstein concludes that political prognostication “blinds us to the only actual, ineluctable reality--that no one knows what the future holds” (Perlstein, 11).

As I outlined here, McArdle analyzes based on ideology, not facts, and therefore is often wrong since her ideology is often wrong. McArdle frequently states that everything is too hard and nobody can know anything ever, which is not a surprising attitude in someone who accepts facts that fit her preconceived notions and dismisses facts that do not. How can you insist you are always right when the lying facts show you are not? How can anyone trust anything?? If the facts are right than the ideology is wrong, and the ideology cannot be wrong. For most people, their ideology is based on their emotional needs and to deny their ideology is to deny them--their value and values, their feelings and thoughts.

McArdle's commenters smell blood in the water and are exceptionally rude to the young study author who responds with civility in the comments. Neither they nor McArdle care as much about accuracy as they care about destroying anything that mars their perfect ideology, their perfect fantasy world.


Anonymous said...

I hope Evan got an A on the project.


Susan of Texas said...

And a purple heart. The condescension and insults in comments are unbelievable.

Anonymous said...

Interesting review, Susan of Texas. I can't bear to go over there and watch McMegan and crew tear up an actual scholar. There's one other thing I think needs to be mentioned before I go off and re-read the Isaiah Berlin--radical humility. The true scientist, in any field, is radically humble in the face of the uknown while still pursuing the goal of pushing back against the darkness and shedding light even into the penumbra around that which we don't know.

I disagree with Perlstein's radical scepticism because although its true that the future is unpredictable in a generic sense the future of any system is not unpredictable. Think of it as the difference between forecasting the absolute future for an immense and unpredictable and chaotic system that includes the universe as a whole to predicting the outcome of republican and democratic negotiations over the budget? We can't confidently state that the earth will be hit/or not hit by an asteroid propelled by the demon zorg tomorrow. But we can confidently predict that Democratic Senators and Republican Senators will probably do this or that tomorrow during session X based on their previous behavior.


charles pierce said...

Her commenters are absolutely adorable. Can you put an IPad into a Trapper Keeper so you can scribble "Mr. Megan McArdle" on the cover?

Kathy said...

"No one can really know anything" is a sophomoric cop-out of dullards.

ifthethunderdontgetya™³²®© said...

As I outlined here, McArdle analyzes based on ideology, not facts...

I call that "making up sh!te" so you can get paid by our corporate overlords.

AKA b.s.

Anonymous said...

McCardle and Douthat both reason backwards from the conclusion they want to reach--and its always a conclusion based on serving the interests of the powerful. In Megan's case its corporations and the rich, two groups for whom she has an almost sickeningly slavish devotion. In Douthat's case its to the Catholic Church and a bodiless notion of morality which floats free of everything except for concern about how to control women's sexuality and fertility.

This is what makes reading McCardle or Douthat (or, for that matter, David Brooks) both incredibly fun and tedious. Its like doing some kind of intellectual sudoku (I'm not a player so forgive me if this comparison is inapt) or perhaps I mean like a non intellectual "where's waldo?" or accrostic. Once you know what the key word or concept is going to be you can rapidly navigate through the maze--perhaps a better example would be moral paint by the numbers?

I gave up reading Brooks when I was able to determine exactly what lies he would adduce to produce what propaganda point within the first two sentences of any given essay. Even though by that time Brooks had mastered the art of beginning with a piece of pseudo intellectual boilerplate that would appear to be heading in some other direction. He'd modeled himself on the incomperable Stephen Jay Gould and so he'd begin with baseball, or some recent piece of misunderstood science or philosophy and then work his way rapidly to the point: rich people good/working class people bad or whatever meme he was being paid to propagate.

I should add that saying that he'd modeled himself on Steven Jay Gould is like saying that a lump of unworked clay is modeling itself on Michaelangelo's David.


Kathy said...

I wonder if ArgleBargle has ever taken Krugman on in person, on TV? THAT would be something to see. And after she polishes Him off, lets watch her debate Chomsky.

UncertaintyVicePrincipal said...


That's exactly what reading those people has always felt like to me, and I had exactly the same experience with Brooks in particular, it became so predictable that even wondering how he would turn the rambling pseudo-philosophical musing into supporting some ultra right wing point was just not fun anymore.

After a while it's like watching a dog roam around on his twice daily walk; you don't know when exactly he'll stop sniffing around seemingly aimlessly and just piss on something, but you know he will.

Anonymous said...

Uncertainity vice principle--

That was a thing of beauty. The only thing I don't like about your analogy is that the dog doesn't think its doing anything very brave or thoughtful when it pees while Brooks always has that smug air of one who thinks he's enlightening his inferiors. All the more annoying as he is endarkening them.


Lurking Canadian said...

In this corner, we have a Nobel laureate.

And in this corner, we have someone who took a few Econ classes in b-school and believes nobody can know anything ever.

At the risk of committing the fallacy of appeal to authority, I know where my money is going.

Syz said...

Appeal to authority is not a logical fallacy when the person in question really is an expert. Sometimes, the fallacy is called "Appeal to Misleading Authority" to make this distinction. McMegan is a classic example of a "misleading authority".

Anonymous said...

speaking of David Brooks, has anyone seen this blog post at the London Review of Books? The 10 Worst Sentences in David Brooks' book.

A taste: Like most women, she got lubricated even while looking at nature shows of animals copulating, even though consciously the thought of being aroused by animals was repellent.

I honestly cannot believe he got paid to write that shite.


Rugosa said...

"People like Lindsay Graham cannot go on Meet the Press and say "Yup, we're going to lose on November 2nd" even when it is completely obvious that this is what will happen; they need to present an optimistic bias for their base."

So McArdle basically admits that pundits/politicians (at least the ones like Lindsay Graham) prognosticate based on ideology rather than facts, but does not see that she is supporting rather than undermining the study's conclusions.

David in NYC said...

@Lurking Canadian --

Don't forget that among Megan's many other qualities is the utter inability to use a calculator.

@Battochio --

Yes. Although that probably is an insult to real hedgehogs, who are assuredly smarter than McBlargle and her fans. I am particularly amused by the many who insist that Megan is the fox and Krugman the hedgehog (argument by assertion, I guess).