Ezra Klein asked three political scientists to create a model for forecasting presidential elections:
The final model uses just three pieces of information that have been found to be particularly predictive: economic growth in the year of the election, as measured by the change in gross domestic product during the first three quarters; the president’s approval rating in June; and whether one of the candidates is the incumbent.
That may seem a bit thin. But it calls 12 of the past 16 elections right. The average error in its prediction of the two-party vote share is less than three percentage points.
That sure sounds like a bad model to me. Here's a simpler model that gets 13 of the past 16 elections right: the incumbent party wins if it's been in office for four years, and loses if it's been in office for eight or more years. Even if you insist that Al Gore "won" in 2000 because he won the popular vote, it gets 12 of the past 16 elections right.
So what's the point of adding two more variables if they don't provide any additional accuracy? I don't get it. However, Ezra says the value of the model is that it forces you to think about basic factors without getting distracted by ephemera like Etch A Sketch or Mommygate. So if you want to think about it, you can play with the model here. Have fun!