Nate Silver has an article in today’s New York Times Magazine and a nifty interactive web site where you can pick the growth that you expect in the economy along a slider, and your expectations of President Obama’s approval rating, and it will show you the chance that any given Republican candidate can expect in the general election.
He believes that he can quantify the race. Referring to the imperfect predictor of economic growth, however, he makes the point that, “Some political scientists have tried to explain these exceptions by resorting to an alphabet soup of economic indicators, conjuring obscure variable like R.D.P.I.P.C. (real disposable-personal-income per capita), which they claim can predict elections with remarkable accuracy.” He goes on to say that with only 17 elections to test them on, and so many different indicators to refer to, simply by chance some of them will be correct all the time so far.
The same argument applies, though, to the two indicators he has chosen, GDP growth and the President’s approval rating. Moreover, he explains how imperfect each of these has actually been in practice, and on this house of cards he builds a prediction mechanism which provides with significant numeric precision, what the candidates chances of election are.