I Predict That No One Will Remember the Accuracy of Your Predictions.

The Daily Beast's Benjy Sarlin gives us an interesting flashback to 1994, when the prognosticators -- Charlie Cook, Stu Rothenberg, Larry Sabato -- all gave their confident predictions that Republicans would pick up around 25 seats in the House. In the end, they actually picked up 54. My favorite part is when Sabato says, "My slogan has always been: 'He who lives by the Crystal Ball ends up eating ground glass,'" by which I guess he means to say that if your predictions turn out wrong, you'll be held to account.

But you won't. Cook, Rothenberg, and Sabato are still successfully plying their trade, and 1994 was hardly the only wrong prediction they ever made. Here's the thing about this game: There is no accountability. You might get rewarded for making a surprisingly accurate prediction, but you will never, never, never get punished for making an inaccurate prediction. Let's say that it turns out that Republicans pick up only 30 seats, or 90, instead of the 50-60 everyone is now saying they'll get. Will a single person in the media say, "You know, maybe we shouldn't bother listening to Charlie Cook anymore"? Of course not. What keeps the press calls and the speaking invitations coming is the appearance of inside knowledge, the idea that they have systematically applied their unique understanding of politics and revealed truths that others can merely guess at.

You would think that the changing media environment -- in which newcomers with excellent records like Nate Silver can break in, and we can easily access the projections of political scientists working from a variety of models, and all of the information that the "gurus" like Cook use can be accessed by anybody -- would take these guys down a notch. But it doesn't, partly because their views are still sought by the old-school journalists who remember when they were the only predictors in town, and partly because after the election, no one remembers whether they were right or wrong.

So here's something I'd like to see. What if after the election, we look at all the predictions made by professional (and maybe not so professional) prognosticators, then hold the ones who got it wrong up for ridicule and contempt? If you're going to make a prediction, you at least ought to be mocked when it turns out wrong.

-- Paul Waldman

You may also like