Here's a recent statement on Bush's approval ratings and what they portend for his re-election from the conservative website RealClearPolitics.com.

As a crude measuring stick for the state of the presidential race, an over 50% job approval for the President should translate into a Bush victory. A 45% - 49% job approval will mean a close race, but I would give President Bush the advantage. A 40% - 44% job approval for the President would translate into a dead heat race, and below 40% and you would have to give the advantage to Kerry.

Oh really? And just what do they base this cheerful assessment on? They don't say, but it is worth noting that this represents a considerable lowering of the bar for Bush (just keep it at 45 or above and even 40-44 isn't so bad!) compared to earlier Republican claims about his approval ratings.

Here's what I had to say back in February of 2003 when Matthew Dowd, pollster for the RNC, was touting 50 percent or above as the magic number for Bush:

....we don't really know an incumbent president at 50 percent can't be defeated. After all, we only have approval rating data from 1948 onwards, so there are a very limited amount of cases to consider -- to be precise, eight, if we restrict our attention to incumbent presidents. Of these eight, none had a rating of exactly 50 percent in July of their election year and the closest was 53 percent. So maybe 53 percent is the magic level (if there is one) -- Dowd's use of the 50 percent level is purely arbitrary and slants the rather thin historical record in Bush's favor. In addition, the president who had this 53 percent rating and was, in fact, reelected (Ronald Reagan in 1984) also had the benefit of a strong economy. So what happens at 53 percent and a weak economy, like Bush currently has? We don't know. And we certainly don't know about 50 percent and a weak economy, the scenario Dowd seems to be trying to cover with his confident historical assertions. Indeed, the closer one looks at Dowd's "50 percent and you're golden" rule for re-electing incumbent presidents, the shakier it looks. What he's trying to pass off as an iron law of history is, in fact, a tendentious reading of a very modest amount of real data.

What goes for Dowd goes double for RealClearPolitics and their attempt to make 45 percent the magic number.

Especially since there's really only one case that falls into the 45-49 percent approval category--where Bush is now--and that's Gerald Ford. And he lost. And there's really never been a 40-44 case (Carter and Bush I were sub-40 guys) so how can they say that such a rating translates into a "dead heat"? It seems more reasonable, if you're going to play this game, to say that a rating in between Ford's and Carter/Bush I's translates into a likely loss, not a tossup.

These guys are just whistling past the graveyard. Their attempts to make up fake historical laws are just designed to (1) paper over the fact that his current approval ratings are more bad than good news for an incumbent president; and (2) give them bogus talking points so that, if Bush falls any farther, they can still claim he's bound to win.

Don't be taken in. They're worried. Real worried.