Last week I did a post suggesting that candidates like Rand Paul and Pat Toomey would offer an interesting test of the frequently-heard argument among conservatives that a more rigorous ideological stance would help Republicans win general elections. My skepticism about the "move-right-and-win" hypothesis earned me some abuse in the comments thread, partly on the legitimate grounds that I didn't systematically demolish it with data and instead relied on the long-standing views of political scientists and election practitioners (a "conservative" assessment of the burden of proof on this subject, but not one that persuades enthusiasts).
As it happens, Emory University political scientist Alan Abramowitz was intrigued by the post, and did a quick study of Senate general elections over the last four cycles to weigh the impact of conservative ideology. His findings, published over at The Democratic Strategist, supply some strong empirical support for the "median voter theorem" that has long influenced political efforts to "seize the center" in general elections.
You can read Abramowitz' write-up yourself, but when he charted every Senate race since 2000, there was a clear negative correlation between conservative ideology as measured on the DW-NOMINATE scale and general election vote percentage. He then went further and controlled the study for factors like challenger strength and the existence of scandals, and again found the same negative correlation. In fact: "For every additional one point increase in conservatism, Republican incumbents lost an additional three percentage points in support relative to their party's presidential candidate."
Abramowitz' study was necessarily confined to incumbents, as the only candidates with a reliable objective indicator of ideology. But the apples-to-apples comparisons of conservative with more "moderate" incumbents indicates that a rightward tilt does not in any way produce some general election advantage.
Now to be extremely clear on one critical point: There are abundant reasons other than candidate electibility for conservatives to urge the GOP to become more conservative, with the most obvious being the desire to achieve conservative policy goals. Most activists, after all, are not involved in politics just to win elections for one "team" or the other, with no consideration of what that means in the real world of consequences for the country. Indeed, I'd say that the prevailing opinion on both the Left and Right is to encourage as much ideology as the political marketplace will bear.
But a clear-eyed view of the political marketplace remains essential, and in a representative democracy, it's also a matter of principle to care about what a majority of actual voters want, if not each minute, then over time. And let's face it, there's really no recent precedent for a major political party reacting to two consecutive bad election cycles by becoming more ideologically rigid.
Perhaps the accumulated economic and political traumas of the last couple of years have changed everything, and made Abramowitz' study, the political science consensus, and recent precedents irrelevant. And maybe the pro-GOP tilt of the political landscape is so pronounced at the moment that the ideology of this or that candidate will not even matter to voters (in which case Republicans can "move right and win" not as a result of ideology, but in spite of it). But we'll have to see some election results before concluding that the conventional wisdom of "seizing the center" is wrong.