Here’s the best explanation I’ve read of why pollsters discount mid-term results, from Barrons, http://online.barrons.com/article/SB50001424053111903463204578048870829726606.html#articleTabs_article%3D1
“Political scientists generally discount the predictive power of midterms—and with good reason—but this time may be different. Turnout for these contests generally is short of its potential by 30% to 40%, which is too low to be reflective of the nation’s mood, the argument goes. And a party that is out of power typically scores gains in a midterm because its adherents are frustrated, angry, and consequently more energized than the incumbent party. This was true in 2006 when Democrats took back Congress from the GOP. In 2010, a backlash against Obamacare by Republicans certainly was the biggest factor in their rout of the Democrats.
But Morgan reasons that the 2010 election was not your run-of-the-mill midterm. “It was historic,” he says, comparing it to epic GOP victories in 1894 and 1924. “This was as good as any of them. In the Senate we won 25 seats and lost 13. This is astounding.”
Can someone tell me what I’m missing in the graphic below? It looks like Indy’s are breaking for Obama by a point.
I think you are correct. Tide turning to Obama perhaps??
Few things that jump out at me in the internals.
Notice that Obama has a huge 20 point advantage in the “East”
However, Romney has a four point advantage with support at 51-47 in the Midwest. Considering that Illinois is in the Midwest, it would seem Romney is well positioned then in the other Midwestern states.
Also Obama’s only real advantage is with the under 29 age group. So how many of this age group thatare deemed as likely voters because they voted in the Obama youth wave in 2008, will actually bother this time?