Skip to main content

Worried that the famous "Sell in May and Go Away" seasonal pattern spells the imminent doom of the stock market's powerful recent rally? You're worried for nothing.

Before I tell you why, I need to first give a nod to the statistical foundation of this seasonal pattern, which also goes by the name Halloween Indicator. That foundation appears to be so strong that it might seem surprising that I would disavow the pattern. Since the Dow Jones Industrial Average was created in the late 1800s, its average gain during the so-called "winter" months (Halloween through the subsequent May Day) was 5.3%, in contrast to an average gain of just 1.6% during the summer months (May Day through Halloween).

This difference is significant at the 95% confidence level that statisticians often use when determining whether a pattern is genuine. And this pattern has received academic recognition, beginning with a 2002 article in the prestigious American Economic Review by Sven Bouman, CEO of Saemor Capital, a Netherlands-based investment firm, and Ben Jacobsen, of the TIAS Business School, also in the Netherlands.

Scroll to Continue

TheStreet Recommends

The reason you nevertheless should not bet on this pattern over the upcoming summer months: The recent discovery that it traces almost exclusively to just one six-month period every four years. This particular period is the one that begins on the Halloween just prior to the mid-term Congressional elections and lasts until the subsequent May Day. Upon removing that one-of-every-four winter periods from the sample, there is no statistically significant difference between the stock market's winter and summer returns.

Credit for this discovery goes to Kam Fong Chan, a senior lecturer in finance at the University of Queensland in Australia, and Terry Marsh, an emeritus finance professor at the University of California, Berkeley, and CEO of Quantal International, a risk-management firm for institutional investors. (They present their findings in a study circulating on the Social Science Research Network.)

What this research means: The "Sell in May and Go Away" pattern would more accurately be called "Post Mid-Term Election Strength." That pattern tells us nothing about the market's tendencies over the rest of the four-year Presidential cycle.

So, strong as the "Sell In May and Go Away" pattern appears on the surface, it in fact is not a compelling reason to go to cash at the end of April.

That doesn't mean the bull market will necessarily continue over the next six months, of course. But if the market does stall or fall, it won't be because of this seasonal pattern that also goes by the name "Halloween Indicator."