What can we learn from the mess of the 2012 election? Could someone who isn't Nate Silver use past polling errors to make predictions for this next election? It's time to start looking at the old polling data to see what might have gone wrong.
There is a very real chance that the Republican voters in Kentucky and South Carolina will vote for the Tea Party challengers because they view their senators as too moderate -- and that is problematic.
Nate Silver's numbers tell us where the campaign is at a given moment. And good punditry can tell us how the campaign got there. Even though the statistics maven is riding high now, Nate Silver needs to realize he is only half of the equation.
Again, not quite right. What Silver does is not an "art"; it's mathematical modeling. And Silver doesn't make judgments the way pundits do, trusting their gut, their instincts, or their experience. He assesses polls based on how they performed historically, among other things.
In polling, a year is an eternity. A little less than a year before the 2008 election, Hillary Clinton led now-President Barack Obama by 27 points, and Obama just barely tied with candidates Giuliani and McCain.