That’s true enough, but reducing the race to a “3-2-1” sound bite makes Romney’s task appear misleadingly easy. Take the “3” part. It looks like Indiana will revert to the G.O.P. column. However, even after Obama’s shift on gay marriage, which was widely expected to hurt him in the South, he is still leading in Virginia by three or four percentage points, according to recent polls. Romney has a narrow lead in North Carolina, which I’ve suspected all along that he’d end up winning, but it’s too close to call. Then there’s Ohio, where Obama is still just ahead, and Florida, which appears to be tied. For the Rove strategy to succeed, Romney has to run the tables in these four states.

That’s possible. At this stage, though, it seems unlikely. Ultimately, everything depends on what happens at the national level: swing states almost always follow the overall trend. And the national polling data still offers some encouragement for Obama. Despite the poor jobs figures, which, together with Santorum dropping out, are what gave new life to the Romney campaign, Obama’s approval rating has held remarkably steady during the past couple of months. At the end of March, when the economy still seemed to be on the up, the Real Clear Politics poll average, which combines all the major surveys, put the President’s approval rating at 47.3 per cent and his disapproval rating at 47.6 per cent. Today, Obama’s approval rating is 48.3, and the disapproval rating is 47.8. Allowing for a bit of statistical noise, there hasn’t been any change in either figure since March…

Romney is still Romney—a compromise candidate with a lot of baggage whom few Americans have great enthusiasm for. While his approval ratings have risen since March, they are still below Obama’s in most polls. Even on the economy, which is his trump card, he has yet to establish a consistent lead over the President.