Ezra Klein quotes, and then affirms, Dan Balz and Jon Cohen's representative and seemingly unshakable "narrative" of the presidential race:
"[T]he lack of [polling] movement underscores intense polarization — about nine in 10 Republicans back Romney, and a similar proportion of Democrats support Obama — and a relatively small percentage of voters say there is a 'good chance' that they could change their minds before November."
This race could go either way. But it’s not likely to break dramatically in one direction or another.
On Klein's second point, agreed. However the second point violently conflicts with the first point--"This race could go either way"--if state-by-state polling has been correct. In fact if that polling is correct, then the election won't be anywhere near close.
Again, like it or not it all comes down to the decisive battleground states, in which Obama has held a small but consistent and statistically significant lead. It's interesting but immaterial if national polling reflects immovable polarization and a tiny percentage of persuadable voters; indeed these facts only help to confirm the thesis that the election's outcome is essentially already knowable.
If voters are polarized and unpersuadable--good. That should lock in the swing states. Even if Obama's battleground leads shrink to a mere 50.1 to Romney's 49.9, Obama would carry 100 percent of those states' electoral count, and thus the day.
And that, rather obviously, is the contradiction contained in political journalism's jackhammering "narrative" of a close, very close, ever-so-close race. The various statistical realities from the battleground states, which journalists also cite on occasion, literally do not add up to their edgy narrative.