A recent poll by Rasmussen has Obama's approval numbers continuing their march south. A clear majority disapprove of the job he is doing--53%, while only 46% approve. Moreover, those that strongly disapprove have climbed to a whopping 40%. Only 30% strongly approve of the job President Obama is doing.

Of more immediate interest is the surprisingly quick reversal of fortune for Congressional Republicans, who were well ahead last year at this time and pretty much even at mid-summer this year. Since then, the Republican lead has grown to six percentage points.

Other polls have different results, which shows how much the wording of the question matters, but the trend is identical in all of them---Democrats in steep decline, Republicans ascending.

The Generic Congressional Ballot has an inherent bias towards Democrats in most cases, largely because of the problem of determining who is a likely voter or not. To use the recent New Jersey governor's race as an example, various constituencies that voted heavily for Obama simply did not show up at the polls this time around (blacks, youth), so polls using the demographic model of last year's election had Corzine and Christie in a dead heat, while the actual election result had Christie winning by a comfortable 4%.

Historically, the bias has been anywhere from 5-10%, so the actual preference for Republicans is likely understated by some margin--perhaps a considerable margin.

You and I will know what next year's election will really look like when challengers declare for Congressional races. When prospects to turn a seat are slim, both parties adopt the sacrifical lamb strategy, by finding some party loyalist with no real prospects for higher office, and run them as a candidate. The high quality candidates always wait for the sure thing--races they are virtually guaranteed to win. If we see a lot of good quality Republican candidates challenging the Democrat incumbent, then its a pretty good sign that its going to be a blow-out.

ADDENDUM: Gallup concurs