Mudbrain says: "Only a demoshill like you considers the NY Times to be anything but a prog propaganda piece, wsmith." You seem to judge the veracity of EVERYTHING based on the source, not on the actual data. You lack the basic ability to distinguish reality from fantasy. Let me be as diplomatic as possible - YOU-ARE-AN-IDIOT!

Right, he bombed them using his drones. This from someone who believes Obama hasn't met the requirements to be president. You're like a child - an angry child.

Better than a pedophile, right wilford? I find it hysterical that ANYONE thinks I'm a big Obama supporter. I guess if you possess simple binary brain function in a black and white Manichean world, everything must be either 'a' or 'b'. Real people don't think that way - that appears to be unique to Town Hall nutters.

Nice vapid word salad there, maryann. I bet you believe that you're clever...

PS - Grifter, "projection" has a meaning. Just repeating a word you saw in the wrong context makes you "look" stupid.

Right, Grifter. Let's just call off the election and call Romney the winner then. What a dolt!

"Your hero is fading fast now" Grifter, peruse this and then try to reconcile your inane comment with what it says: http://fivethirtyeight.blogs.nytimes.com/2012/10/29/oct-28-in-swing-states-a-predictable-election/#more-36832 Wishful thinking indeed...

Silver is NOT a pollster. And if he was, NOBODY has turned their "decision making" over to him. Polls measure public sentiment, they don't shape it.

Get back to me when you move from middle school to high school. Maybe you can take an elementary statistics course there.

Kevin, you did a pretty decent job of explaining. You even came to the correct conclusion: "The answer to Nate Silver isn't to go back to traditional political punditry... but it's to develop ever more statistical models in the hopes of finding a better way... to forecast." But you HAD TO throw this in: "Circling back to Silver, if Romney pulls out a victory... it'll be a data point for Silver to add into his model to try to 'get it right' next time." WRONG, WRONG, WRONG, WRONG, WRONG, WRONG! It will do NOTHING to change statistical modeling, it will simply show that a 22% probability (actually 27% at the time of my post) is INFINITELY greater than 0% probability. IT'S A MODEL!