Why the Redskins—and the Red Sox—matter

International Economy, The, Fall, 2004 by Edward M. Graham

Since my article "Why the Redskins Matter" was published in the spring 2004 issue of The International Economy I have been asked many questions about the Redskins standard for predicting the outcome of the U.S. presidential election. The standard, to review, is that if the Washington Redskins lose their last regularly scheduled home game prior to the election, the incumbent President loses the election. Otherwise, both win! This standard successfully predicted the outcome of every election from 1948 through 2000.

But, in 2004, the standard apparently failed: The 'Skins lost their at-home game against the Green Bay Packers on October 31, 2004, but U.S. President George W. Bush won reelection just two days later. So, the question I have been asked most often and most recently is: What happened? This question has been posed to me by persons high, e.g., a senior advisor to the Japanese Prime Minister, and low, e.g., a Harvard Business School professor who clearly wants to knock me down.

There are two explanations. The first, a rather technical one, is that any statistical predictor has associated with it a standard error of forecast. Otherwise put, no predictor is perfect. Let's put it this way: Even if the Redskins standard gave the wrong result in 2004, it has given the fight result for fourteen out of fifteen elections. Can any other predictor claim a record this good?

But a second explanation is that I seem to have incompletely specified the Redskins standard. A friend of mine--a quite good economist whom I will nonetheless keep nameless--has pointed out where I went wrong. It seems that a Redskins loss in the last home game prior to the U.S. presidential election predicts an election loss by the incumbent unless the Boston Red Sox win the World Series in that year, in which case a Redskins loss predicts that the incumbent will win. I must admit that I had not been aware of this conditionality attached to the Redskins standard when I wrote the article last spring. And, of course, now that we know this conditionality, we know that the Redskins standard remains inerrant.

--EDWARD M. GRAHAM

Senior Fellow,

Institute for International Economics