Can the Super Bowl Winner Predict Market Performance?
The “Super Bowl indicator” is a theory that predicts stock market performance based on which football conference (the AFC or the NFC) wins the Super Bowl. The prediction is simple: If the NFC wins, the market will end the year positive; if the AFC wins, the market will end the year negative. The indicator has been correct an impressive 40 / 49 years – or 82% of the time (source: wsj.com). Go Falcons! Right?
Wrong. Rest easy, Pats fans. The temptation in these fun correlations is to infer that correlation equals causation. In this example, it would mean believing that an NFC win causes the stock market to end the year positive. This is wrong. In reality, the two factors are simply correlated by random chance. Not convinced? Google “Bangladesh Butter indicator” and you’ll have a laugh about how changes in butter and sheep production in Bangladesh have explained – with over 95% accuracy – stock market trends in the 80s and 90s. Obviously, these factors had nothing whatsoever to do with the stock market. Just because the correlation is strong, does not mean there is a causal relationship.
The inappropriate use of correlation to imply causation bleeds into financial articles as well. I’ve seen many articles predict stock market declines in large part based on the historical correlation of the number of years in between previous market declines. But as we know, correlation alone is a poor basis upon which to make predictions. So next time you see such a prediction, decide instead to watch football. Just make sure that you’re rooting for the Falcons. After all, an NFC win couldn’t hurt, right?