At FiveThirtyEight today Harry Enton has a piece bearing the title: “The Senate Races With the Biggest Ideological Stakes.” He uses a defensible if not completely compelling methodology to attach an ideological number to the candidates in ten competitive Senate races and then discusses the gap between candidates in any given race as indicating the “ideological stakes” for its outcome.
I suppose this characterization of the ten races is persuasive, though obviously the ideological “stakes” are highest in any race that tips control of the Senate from one party to the other. But what I found most interesting in these numbers (aside from the fact that it documents the already well-established fact of a major gap between the most conservative Democrat and the most liberal Republican) is that there’s no predictable relationship between the competitiveness of the seat and the size of the ideological “gap” between Ds and Rs. The two states with the greatest ideological polarization, Colorado and Iowa, are arguably the most “purple.” This, of course, completely violates the “median voter theory” whereby you expect the two parties in the most competitive areas to oscillate towards the “center.” The states with the third and fourth highest levels of polarization, meanwhile, are deep-red Alaska and Arkansas.
The only clear trend I see here (in an admittedly small sample) is that it’s increasingly difficult to say there’s not a dime’s worth of difference between the two parties.