by Nicholas Beaudrot of Electoral Math
Since the Paul Hackett's run at the OH-2 seat last August, we haven't had any elections that can usefully gauge whether or not dissatisfaction with Bush runs deep enough to cause dissatisfaction with Washington Republicans as a whole. We also don't have much information on the shape of public dissatisfaction with Republicans. Is Bush losing support in Orange County & Waco Texas, bringing the Republican margin from 80% down to 60%, while keeping enough support in Connecticut and upstate New York to let Reps like Chris Shays (R-CT) and Peter King (R-NY) skate by? Or are there enough districts that have swing from Red to Blue to take back the House?
The best we can do is look at who's bothering to turn out in primary elections, and the few pieces of district-level polling that are publicly available. In the primaries, the news has been encouraging. For instance, in IN-2, IN-8, and IN-9, Democratic challengers all scored victories over Republican incumbents, based on the number of ballots drawn. Likewise, Bob Casey , Jr. seems to be dragging his party to victory in Pennsylvania, scoring "wins" over two incumbents and making over a half-dozen races more competitive. In most elections, Republican turnout does seem to be down from historical norms, and Democratic turnout is up. But not all the news is good news. Democracy Corps' polling, which focuses on the far end of what one might consider a "competitive" race, shows only the tiniest bit of movement towards Democrats.
So at the moment, we can't forecast a tremendous takeover in November. The odds look good for significant Democratic gains, but perhaps only enough to take back the House by the slimmest of margins, if at all. Francine Busby's (D) June 6th special election in CA-50 will give everyone a bit more hard data.