False Prophet Criticizes Nate Silver’s Predictions


Josh Kraushaar, editor of the National Journal, was wrong about the 2012 election. Really, really wrong. As in so wrong that he predicted a Romney victory. But he still feels qualified to criticize Nate Silver, who has absolutely nailed the last two presidential elections.

I’m a numbers guy. As a baseball fan, I pore over box scores, regularly second-guess managers who use old-school tactics, and was probably one of Nate Silver’s first readers and an early subscriber to the sabermetric reference book Baseball Prospectus, where he made a name for himself projecting player outcomes. In reporting on and analyzing politics, I rely greatly on fundraising reports and polling data to inform the trajectory of key races.

But count me underwhelmed by the new wave of Senate prediction models assessing the probability of Republicans winning the upper chamber by one-tenth of a percentage point. It’s not that the models aren’t effective at what they’re designed to do. It’s that the methodology behind them is flawed. Unlike baseball, where the sample size runs in the thousands of at-bats or innings pitched, these models overemphasize a handful of early polls at the expense of on-the-ground intelligence on candidate quality. As Silver might put it, there’s a lot of noise to the signal.

Well yes, that’s why they’ll be continually updated as new polls come out over the next few months. That’s how these things work. You’d think he’d understand that, being a “numbers guy” and all. And let’s look at how this “numbers guy” did at predicting the numbers in 2012:

There was Kraushaar, predicting GOP presidential nominee Mitt Romney would triumph in Florida, Virginia, North Carolina, Colorado, Iowa and “one of the Rust Belt firewall states” en route to the White House.

Romney, of course, lost all of those states save for North Carolina.

And there was Kraushaar, writing that Minnesota could be the state that surprises. Silver gave President Obama a 99.7 percent chance of carrying the state. The incumbent ultimately triumphed there comfortably.

I’m sure Dean Chambers will be along soon to “unskew” all these polls for us and the holy trinity of bad election predictors will be complete, along with Kraushaar and Dick Morris.

Comments

  1. says

    @1

    In his world, “candidate quality” really means “ideological purity.” The fact that that’s is an overall loser in a general election is lost on him.

  2. scienceavenger says

    OK “numbers guy”, please give me your numbers with regard to “on-the-ground intelligence on candidate quality.”

  3. Kevin Kehres says

    Well…naturally, those states where he was wrong was where the massive voter fraud occurred, doncha know…

    Better defund ACORN before it happens again.

  4. dingojack says

    Interesting the article I read (not by Nate Silver BTW) indicated that:
    a) Initial poll results tended to correlate moderately with the final result, and are within 6 points of that result (ie. the Republicans getting the senate is a toss-up, somewhere between 44.1-56.1 favouring the GoP to 55.9-43.9 favouring the Democratic Party)
    b) the President’s favourable numbers are weakly correlated to the chances of winning an election with 43% approve being the minimum to win (although the spread is a lot larger so that level of approval is not a guarantee of success).

    Dingo

  5. StevoR : Free West Papua, free Tibet, let the Chagossians return! says

    @4. Kevin Kehres : “Better defund ACORN before it happens again.”

    From what I was told on another thread I gather ACORN was pretty much totally defunded already.

Leave a Reply