The Final Bell – Nate Silver Destroys Karl Rove

But I beat Rove too.

I’m calling Florida for Obama.  The state will have to have a recount but I don’t think it matter; Obama wins the state.  That makes Silver 51 out of 51 races correct.  Rove came in at 45 of 51.  Which makes it sound closer than it was.  The two differed on only 6 races and Silver won all 6 of them.

I, however, got 3 of them right; I lost Virginia, Florida and Colorado.  I feel pretty good, I was only VERY wrong on Colorado.  Florida and Virginia were very close.  I did pick the Badger State to flip thinking that Walker would save the day, so I get that one wrong.

Clearly we are going to have to watch 538 going forward.

 

7 responses to “The Final Bell – Nate Silver Destroys Karl Rove

  1. I only got NC wrong — and I expected that. I made a long shot prediction for NC, and it was close. I also got the Senate right except for one long shot – I thought Carmona might win a surprise (usually there’s one Senate surprise so I figured he’d be a good bet). The election really did follow what the polls predicted, and Nate Silver’s model worked very well. I like it when data and social science methods are validated. I’m not sure where Rove, Barone, Will and the others got their predictions (Dick Morris, I know, pulls them out from between his toes).

    • I only got NC wrong — and I expected that.

      I did the same thing with Wisconsin.

      Dick Morris, I know, pulls them out from between his toes

      He is SO wrong that I think he is still working for the Dems.

      By the way, what did the democrats do to piss him off so much that he became a republican?

  2. Nate did a great job in 2008, too. This time that year, I saw Shep Smith on Fox News put up Silver’s prediction compared to the actual result, and explain, “You know how he does that? He has a time machine.” He then said something about how Silver had used his time machine to bend reality to get the Red Sox to win the World Series.

    I’d actually come across Silver originally from his work as a baseball analyst. He’s every bit as good as using statistics to make predictions and cut through BS in that work. You know, he originally started his polling work under the pseudonym “poblano” at Daily Kos. So he’s kinda like the UnSkewed Polls guy!

    • Nate did a great job in 2008, too.

      Yup. Mostly because he works for the Times I’ve always held him with a degree of doubt. After all, you could take 1000 predictors and at least 1 of ’em has to be right. But that he did it twice is certainly cause to pay attention.

      I’d actually come across Silver originally from his work as a baseball analyst. He’s every bit as good as using statistics to make predictions and cut through BS in that work.

      When I read that he started out in baseball I immediately took notice. Little politics in baseball. Though I was resoundingly castigated as a communist for suggesting that Johan Santana go on a 4 day rotation with his “bullpen work” be actual real game relief.

      • Ephemeral reminded me of his start as a diarist at the Great Orange Satan.. I do remember Five Thirty Eight from before the Times picked him up a few years ago, based I assume on his excellent work. He was a must read in ’08.

  3. Credit to y you pino! Well done. Nate has been getting it right for a while – he’s a geek statistician, in love with numbers, so good on him too.

    May I say I’m glad that I got one thing right anyway – I predicated a very close popular vote and an electoral college blowout. Yeah!

    We iz so smart you and me.

Leave a Reply to Pino Cancel reply