Not sure how to account for this one. Take a look at these numbers on Iraq buried in the new Quinnipiac polls of Colorado, Michigan, Minnesota and Wisconsin...

As you can see, in every one of these key battleground states, majorities favor staying in Iraq until the situation is "stable," and then withdrawing troops, which in some ways is closer to the GOP's current stated position, though not really identical with it. In all four, decided minorities favor starting withdrawal immediately, with the goal of completing withdrawal in 18 months.

The funny thing about this is that there are two ways of reading it. Either these numbers favor McCain, for obvious reasons. Or, alternatively, they may suggest that McCain is screwed -- in states where majorities favor staying in Iraq until stability is achieved, he's still losing to Obama, in some cases by sizable margins.

Late Update: As a commenter rightly notes below, one key takeaway here is that the debate over Iraq that unfolds in the coming months will be heavily influenced by who successfully defines the terms of this debate -- terms like "stability" and "winning."

Late Late Update: I agree with those of you who say the hed wasn't quite right, and have changed it accordingly. And yes, the question in this poll was badly phrased, which probably explains the findings. That said, it isn't enough to merely note this and conclude that Obama has already won the debate over Iraq. While majorities nationally favor withdrawal, there still is a debate over Iraq looming, and Obama has to win it.