"Crossover" Senators and 2006

In today's Washington Post, Dana Milbank and Brian Faler discuss the decline of so-called "crossover" senators - Republicans and Democrats who represent states which voted for the opposing party in 2004. After that election, as they note (I've discussed this somewhere before too but I can't find it just now) the number of crossovers dipped to its lowest point in at least 25 years: today, just 16 Democrats and 9 Republicans in the Senate represent states which voted for "the other guy."

Milbank and Faler note that this number might diminish even more after 2006, with Republicans targeting several of the crossover Democrats and Democrats targeting at least a couple of the Republicans. The trend of fewer and fewer crossovers, says Jennifer Duffy of the Cook Political Report, is "probably the largest contributor to the shrinking center in the Senate. It's something that we've seen happen in the House over the past decade. The same sort of thing is happening in the Senate."

I wanted to take a look and see what we're really dealing with here. Are crossover senators likely to be more centrist than not? Is "crossoverism" a healthy way to gauge the health of the center? And is it in danger?

The sixteen Democrats representing "red states" are: Pryor (AR), Lincoln (AR), Salazar (CO), Nelson (FL), Harkin (IA), Bayh (IN), Landrieu (LA), Baucus (MT), Conrad (ND), Dorgan (ND), Nelson (NE), Bingaman (NM), Reid (NV), Johnson (SD), Byrd (WV), and Rockefeller (WV). The nine Republicans representing "blue states" are Snowe (ME), Collins (ME), Coleman (MN), Gregg (NH), Sununu (NH), Smith (OR), Santorum (PA), Specter (PA), and Chafee (RI).

Without an easy definition of 'centrist' (and I'm not even going to attempt the construction of one), it's impossible to say "Of these, x are centrists." Just going on my gut and what I know of these senators, I'd consider approximately three-quarters of them to generally exhibit centrist tendencies to varying degrees. Not a perfectly clear conclusion, but a general trend toward centrism from the crossovers. You can probably pick out the aberrations fairly easily, but for the most part, crossover senators do seem to be more centrist than their non-crossover counterparts (of the other 75 members of the Senate, I would consider only about a quarter or less to be centrists).

Now, is there reason to believe that 2006 will further diminish crossovers in the Senate?

Of the 25 crossovers, 10 (Baucus, Collins, Coleman, Harkin, Johnson, Landrieu, Pryor, Rockefeller, Smith Sununu) were elected or reelected in 2002; 7 (Bayh, Dorgan, Gregg, Lincoln, Salazar, Specter, Reid) were either just reelected or elected in 2004; , and 8 (Bingaman, Byrd, Collins, Conrad, Nelson, Nelson, Santorum and Snowe) are up in 2006.

Of those 8, Charlie Cook (and generally I agree with his analysis at this point) lists [PDF] two (Bingaman and Snowe) as "solid," two (Byrd and Conrad) as "likely," two (Nelson and Nelson) as "lean," and two (Chafee and Santorum) as "toss-up." I would probably quibble a bit and move Ben Nelson over to "likely" and Chafee to "lean" for the moment, but close enough. As of right now, I think only Santorum is in serious danger of losing his seat ... and I think in this specific instance replacing Santorum with Bob Casey would do far more to enhance centrism in the Senate than otherwise - he is one of the relatively few cases where centrism and crossoverism clearly diverge.

We're still very far out - things could change quickly in West Virginia and North Dakota to make them very interesting, and Rhode Island and Florida could end up being more competitive than I expect they'll be. I have not considered the other Senate races in 2006 and the potential for gains in crossoverism and/or centrism from those (I'll take a look at them later and see if there's anything worth noting there). Just from this element of it though, it seems pretty safe to use crossoverism as a general benchmark for centrist tendencies, and it also seems safe to suggest that neither will suffer serious blows in 2006. Time will tell, and I'll try to update on this situation as we move closer to the primaries and general election.