Electoral / popular poll split?

I mentioned the possibility 10 days ago, that a split could occur between the electoral college and popular vote polls, mainly because Obama's got an enormous advantage in electoral college numbers if you examine the battleground polls.  Today, it's become an avalanche of media now taking up this issue.

• Ezra Klein follows the Gallup poll's 7 point lead for Romney, noting that there is some fuzziness in the numbers and that a split could occur.
• Michael Falcone and Amy Walter take up this split and the importance of Ohio.
• Charlie Cook talks about the real possibility of a split.
• USA Today's David Jackson mentions that if you tally up RCP's numbers, Obama would win big in the electoral college, even while losing the popular election.

While none of them mention why the split is important (but because you've read my blog, you already know that) the problem with a split is, that there is no national referendum on a candidate's platform.

With regard to the Gallup poll's swing, I think it's worth remembering that you always toss out the outliers, and clearly a +7 for Romney is an outlier, when Rasmussen (again, a conservative pollster) shows the race tied.

And if you follow the 538 blog or predictwise, odds are still a 2:1 or greater split for Obama to win.

In other words, it's slightly harder for Obama to win, and Romney and Republicans now think that they're headed for a win, but I called it months ago, and it's still true today: Obama's going to win.