Today I’m offering my first crack at a new feature on Electionate, where I offer a daily polling round-up and quick takes on matters that I wouldn’t write about otherwise. So what happened today?
While two polls showed Obama ahead and above 48 percent of the vote in critical Virginia, the day’s big newsmaker is Romney’s 5-point advantage in Colorado. As far as I’m aware, this is the first poll to show Romney with a lead of any kind in a state that voted for Obama by 9 points in 2008.
Given that this is the first poll showing Romney with an advantage, it would be wise to wait for more evidence before concluding that there’s been a shift in the race. That's especially true since 38 percent of the sample was 65 or older, compared to 13 percent in the 2008 exit polls. Given that Quinnipiac weights for age, that’s a very odd outcome—and also distinguishes this issue from party-ID, which isn't used to weight most public polls.
Now this doesn’t suddenly make the poll great for Obama, since he only had a slight lead among voters younger than age 65, but the race would have been a dead-heat with a more typical age breakdown. The bottom line, as always, is that it’s best not to focus on a single poll, and the broader set of recent polls suggest that Obama probably still has a slight edge in Colorado.
Otherwise, there’s not much movement in today’s polls—which is generally good news for Obama, at least if you share my view that he's relatively well positioned. That's especially true in Wisconsin, where two polls putting Obama over 50 percent suggest that the state isn't quite within Romney's reach. Obama also had a decent day in Virginia, where Obama held a lead and exceeded 48 percent in two polls. The Reuters/Ipsos poll probably raised a lot of eyebrows by showing Obama up by 7 points, but Obama had a big edge in their prior polling, so their numbers don’t indicate much of a change.
Odds and Ends
—The best case for a Romney victory is that a majority of voters have clear reservations about Obama’s performance, and that they’re just not going to vote to reelect the President. It's a solid case and Jay Cost made it this morning, and while I’ve implicitly offered my thoughts on this question in previous posts, I have one additional thought on Cost's particular iteration: a majority of voters do not necessarily disapprove of Obama’s performance, as Cost seems to imply. Now, a majority doesn’t approve either, but it’s not as though a majority has resolved that Obama’s doing a terrible job. And that’s potentially an important distinction: it would be hard to argue that Romney wouldn't be much better positioned if Obama’s disapproval were 50 or 51 percent, rather than 47 or 48 percent.
—How much will Romney gained from the switch to likely voters? Hard to say. But we can say that in June, that Reuters would have found Obama’s 6 point lead among registered voters falling to 3 points, as the electorate became older, whiter, and better educated than the pool of registered voters.
—There were also two odd sub-state polls: AARP found Romney up two among Floridians over age 50 and a WTOP poll found Obama up by just 2 points in northern Virignia. I've never heard of the latter poll and it's wildly inconsistent with Obama’s narrow statewide lead, since Obama won northern Virginia by twenty points in 2008 (although that depends on the definition of northern Virginia). On the other hand, a close race among Floridians over 50 is quite consistent with the close statewide race shown by most polling firms.