One point about the "history-defying" character of Republicans holding steady or picking seats up at the midterm election: this is less interesting than it appears, and less interesting than people are making it out to be. In particular, if the House GOP holds steady that doesn't mean that the Republicans were tactical geniuses or have some curse-defying level of popularity. Those big midterm losses are heavily concentrated among weak freshmen who got swept in in presidential coattails. But the House GOP lost seats in 2000 (and in 1998 and 1996). There's no big freshman class of people who wouldn't have been in the House but for W's win in 2000. (An obvious point: while the popular vote has no particular constitutional meaning, losing the popular vote does not make for strong coattails.) Indeed, weakness has been weeded out of the House on both sides pretty ruthlessly in the last eight years. The major predictor of the size of the House swing against a President's party in the midterm election is the size of the swing to the President's party during the Presidential election, measured in House seats gained. The history-defying accomplishment was losing seats while gaining the Presidency in 2000. Defying the curse of the midterm election just follows from that.