Senate predictions generally were in agreement that the outcome would be close. The evidence that the House was likely to flip was powerful, yet some prominent doubters held out. I am not talking about Karl Rove or Ken Mehlman, who had to put on a game face until the final buzzer, but about people who weren't being paid to keep a stiff upper lip.
Despite overwhelming evidence to the contrary, some conservative writers, lobbyists, and activists denied what seemed obvious to many of us: that something very big was on the verge of happening. In the major national polls in October, an average of just 29 percent of likely voters said the country was headed in the right direction, while 64 percent said it was on the wrong track.
Midterm elections tend to be referendums on those in power on Capitol Hill and in the White House. In past midterms, when Congress's job-approval rating was above 40 percent, the majority party lost an average of only five House seats. When the job-approval rating was below 40 percent, the controlling party lost an average of 29 seats.
This October, congressional approval averaged 27 percent, and disapproval averaged 66 percent. At the other end of Pennsylvania Avenue, President Bush's approval rating averaged 38 percent, 1 point below the lowest job-approval rating that President Clinton had in any Gallup Poll during 1994.
Finally, the much-maligned generic congressional ballot test -- which doesn't predict House seat changes but does signal which way the political wind is blowing and how strong it is -- showed Democrats running 10 to 13 points ahead in polls from June to September. The Democrats' edge ballooned to 15 points during October and narrowed to 11 points in the week before the election. (Over the years, this gauge has tended to tilt about 5 points too much in the Democrats' favor.)
So when the national popular vote, according to figures compiled by Rhodes Cook for the Pew Research Center, went 52 percent for Democrats, 46 percent for Republicans, and 2 percent for others, no one should have been shocked.
Do the math: An 11-point Democratic lead on the generic ballot test, minus 5 points for the gauge's Democratic skew, translated into a 6-point Democratic victory. When the 6-point Democratic popular vote win is measured against the GOP's 5-point win in 2002 and its 3-point win in 2004, it clearly constituted a wave. And the wave didn't stop on Capitol Hill: The Democratic advantage over Republicans in state legislatures went from 15 seats (3,650 versus 3,635) to 662 seats (3,985 versus 3,323), with gains in every region.
And what about House seats? Republicans appear to have lost 26 of the 43 GOP-held seats that The Cook Political Report had rated either as "lean Democratic" or as toss-ups. Furthermore, Republicans lost three of the 26 GOP-held seats that The Cook Report had rated "lean Republican" or "likely Republican."
But before the Democratic wave actually hit, some Republicans insisted that "Karl is a genius -- he's not going to let it happen," that the GOP financial advantage (the smallest in 20 years) would pull them through, or that their party's vaunted 72-Hour Program would ensure enough votes for victory. Now we hear things like, "Well this was the historical average," or the loss doesn't mean much because "the close races just broke Democratic."
Those rationales ignore the fact that Republicans won or are running ahead in seven of the 11 House races being determined by 1 percentage point or less and that they won 34 of the 66 House races in which the winner got less than 55 percent of the vote. Another excuse offered is that scandals created the Democratic success. But GOP misbehavior handed only a few races to the Democrats.