So you’ve got your Old School race status ratings from Cook Political Report and Roll Call’s Stu Rothenberg. Then there’s FiveThirtyEight, which made statistical prediction models famous, and is now operating out of its own ESPN-backed empire. Filling the vacuum left by Nate Silver at the New York Times is The Upshot, which uses similar methods. And now at WaPo the Monkey Cage band of political scientists are launching Election Lab, its own forecast.
Cook and Rothenberg use the traditional categorization of races as more or less competitive, with truly competitive races being labeled toss-ups unless there’s convincing evidence of a “break” in one direction or another. The newbies, following Nate Silver’s direction, all use percentage probabilities. The only other major distinction I see at this early point is that Election Lab, reflecting the proclivities of its “fundamentals matter most” proprietors, will place more emphasis on economic and approval ratings factors than on current polling, certainly until late in the cycle.
Unsurprisingly, Election Lab enters the fray with a more robust prediction of Republican Senate gains than the competition, showing not only Montana, South Dakota, Arkansas, Alaska, West Virginia and Louisiana flipping from blue to red, but also Iowa and Michigan, which most observers show as leaning Dem. That’s particularly interesting since they also have Kay Hagan is pretty good shape in North Carolina.
Whenever I look at Senate races going forward, I will try to give a balanced view from the various prognosticators, instead of cherry-picking one or another. There will, of course, be a tendency for them to converge in the Fall, when current polling becomes abundant and the distribution of resources by national party committees and “outside” funders often narrows the playing field. But at that point the ways in which the different wizards weigh polls could become especially interesting.
The competition between projections should be good for the quality of political journalism not just this year but even more in 2016, by which time the charlatans who just show their own team winning every vaguely competitive contest will, please Jehovah, just go away.
Feed the Political AnimalDonate
Washington Monthly depends on donations from readers like you.