Those are awful numbers for Obama. It's long been a truism that an incumbent at under 50% is considered in danger.

Consider that he won in November with 53%. Even in his strongest potential match-up, he can't match that number. And that's against a woman whom he and his allies have spent the better part of a year trying to actively destroy. And he still underpolls last November's results.

Each of those match-ups includes about 10% undecideds: and those usually break 2-to-1 against the incumbent. When you break out the undecideds: Obama actually loses to Huckabee in a squeaker, bests Romney only by a couple of points. The BEST he does is slightly improve over his McCain matchup. And again, that's only against Palin (without any additional benefit she might gain from naming a strong #2 on her ticket).

If you take the numbers at face value, they seem that Obama has an advantage. But I'll bet you dollars to doughnuts that Axelrod is absolutely sweating these numbers because they actually betray a great deal of weakness for Obama. And unemployment is still rising with fewer and fewer people are willing to "blame Bush" for the state of the economy as every month passes.

On thing to consider also is the RCP average.

When I found RCP in 2004, they had mentioned that their analysis concluded that a POTUS JA over 50% means an odds on favorite to re-elect, whith the chances dwindling as the average fell. I forget what Bush's JA was on election day 2004, but whatever it was probably reflected the close election results against Kerry.

Also, RCP includes polls of "Adults" in its sample, believing that the solid polls will ultimately counterbalance any shitty one.

According to their own poll, it's of voters, no identification of whether they are registered or likely. Rasmussen, is the standard because of their consistency, and a very narrow margin of error.