Tuesday, April 24, 2012

Commenting on Drum

No one reads "Robert Comments" so I will post a comment here. Many more over there.

Drum writes that Nobody Really Cares About Political Predictions

most pundits don't really care if they're right. It's not like they have any money riding on their predictions, after all. But predictions stir the pot, and unusual predictions stir the pot even more.

Wonderful post. I absolutely agree with you and Waldman. There is a very low penalty for confidently made grossly wrong predictions. I have some thoughts.

First evaluating pundits has become much easier than it was. Now a google search will get you part of the way there, then (when you worked in high tech) one had to get ink on one's fingers leafing through old newspapers. So there has been a ranking of pundit prognostication. The results included the unsurprising facts that Krugman was much better than the others and that many pundits did worse than a flipped coin.


The fact is that someone has an incentive to meta-analyse pundits (people have incentives to find something new to write about -- they aren't all like you with ideas about everything). Watch out.

A) I will never make another forecasts (after the one I just made).

But no one (except, of course, for Krugman who linked to the study) cares.

In fact, I think pundits are rewarded for making firm predictions. I also fear that predictable predictions are favored by cable gab fest bookers. This doesn't just lead to bad predictions but to boring TV.

I think the really alarming thing is that your argument applies to matters where good predictions matter more than the horse race -- predicting that Clinton tax increases would cause a slump and Bush tax cuts a boom has not been punished at all. Yes elections are important, but forecasting them isn't. Forecasting the effects of policy is very important. But the situation is even worse as there isn't a moment of truth.

It's not just the future. Getting facts wrong is allowed. But before despairing I return to my first point. The web has made things better. The incorrect predictions are on the web. Hords of bloggers remind each other of them. Bloggers don't have tenure as I do and columnists almost do. I don't predict that the current pathologies will last, or that things will improve, since (see A above -- I can't repeat it -- that would be a prediction).


chrismealy said...

I read your comments. You should keep them on the regular blog though. Nobody should have more than one blog.

Robert said...

Oh Christ I have five blogs (one of which has two posts ever and the other which has none for years).

I just put two comments over there, before reading your comment here. OK comments here.