About this Blog
About The Author
Email Me

RealClearPolitics HorseRaceBlog

By Jay Cost

« A Note on the "New Progressive America" | HorseRaceBlog Home Page | Obama's Liberal Moment »

Ross Douthat Weighs In

Ross Douthat blogged on my recent discussion with Ruy Teixeira, falling roughly between the two of us. He writes:

I'm on Teixeira's side insofar as it's possible to make predictions about the political future; I'm on Cost's insofar as it isn't...I think probabilities matter a little bit more than Cost allows. Even allowing for his caveats, if you were asked to pick which coalitions you'd rather have at the moment, based on demographic strength alone, you'd choose the Democratic coalition in a heartbeat. Not because we know what's going to happen, but because we don't - and a bet based on probabilities is better than a shot in the dark.

This is not precisely what my point is. I actually do have a prediction based on probabilities: continued two-party competition and roughly divided government over time. That's not a specific prediction for the next cycle - but it is not a "shot in the dark." That phrase implies that I don't have a sense of what to expect, which I do. It just happens to fall in the middle.

My argument is narrowly negative in that I have most recently been arguing against Teixeira's thesis. However, it is broadly positive in that I have written quite a bit about what to expect in the long run and have offered alternative ideas in contrast to "realignment" or "enduring majorities" (or whatever term one might prefer). This negative argument fits into the positive one. See here and here for my most recent assertions.

Sean Trende notes to me the difficulty in using demographic results to predict subsequent elections. "Demographics always look good when you are winning. What could possibly have been the demographic bright spot for Democrats in 1988? Yet they were just four years from renaissance." Sean also asks, what if one were to follow Douthat and predict the next winner based on the current winner? Since 1948, you'd go 7 for 15. Since 1968, you'd go 5 for 10. That's about how well you could expect to do via random, "shot in the dark" guessing.

-Jay Cost