The ‘Sell in May’ Strategy: What the Data Actually Shows About Presidential Cycles and Market Timing

One of Wall Street’s most enduring seasonal adages — “Sell in May and go away” — is once again facing scrutiny as investors weigh whether to reduce equity exposure heading into the historically weaker summer months. A deeper statistical analysis reveals that the so-called “Halloween Indicator” shows meaningful significance during only one specific year of the four-year presidential election cycle, raising important questions about whether 2025 qualifies as the year to pay attention.

The “Sell in May” phenomenon — sometimes called the Halloween Effect or Halloween Indicator — has deep roots in financial folklore, with the full British version of the saying being “Sell in May and go away, and come back on St. Leger Day” (a reference to a famous horse race held in September). Academic researchers Sven Bouman and Ben Jacobsen published a landmark 2002 study in the American Economic Review documenting the pattern across 37 countries over decades of data. Their findings confirmed that stocks in most major markets delivered meaningfully lower returns during the May–October window compared to the November–April stretch. Subsequent research, including work by finance professors who cross-referenced the seasonal effect with the well-known presidential election cycle theory pioneered by Yale Hirsch in the 1960s, found that the pattern’s statistical robustness is not uniform across all four years of the cycle. The data suggest that the May–October underperformance is most pronounced and statistically significant during the first year after a presidential election — the post-election year — when the incoming administration often front-loads controversial or market-unfriendly policy initiatives before the next midterm campaign begins.

This distinction matters significantly for 2025. As the first year of a new presidential term, this year fits precisely into the window where historical data most strongly supports the seasonal sell signal. The presidential cycle theory holds that equity markets tend to be weakest in the first two years of a presidential term, as new administrations pursue ambitious — and sometimes disruptive — legislative agendas. The third year (the pre-election year) has historically been the strongest, as incumbents and their parties seek to stimulate the economy ahead of the next vote. According to data compiled by the Stock Trader’s Almanac and independent researchers, the S&P 500’s average return during the May–October stretch of post-election years has been near zero or slightly negative going back to the mid-20th century, compared to modest positive returns during the same stretch in midterm, pre-election, and election years. This year’s environment — marked by significant tariff policy uncertainty, Federal Reserve interest rate deliberations, and geopolitical tensions — adds additional fundamental reasons some strategists argue the seasonal pattern could hold.

However, it is essential to note the significant counterarguments. Many academic finance experts and portfolio managers argue that even well-documented anomalies tend to weaken or disappear once they become widely known — a phenomenon called “anomaly decay.” Transaction costs, capital gains taxes triggered by selling, and the psychological difficulty of re-entering the market at the right time can erode or eliminate any theoretical advantage. Research from Vanguard and various academic institutions has consistently shown that time in the market tends to outperform attempts to time the market over long horizons. In 2020, for example, markets crashed in March but staged a ferocious recovery through the summer months, meaning investors who sold in May missed substantial gains. Similarly, 2009’s post-election summer saw enormous recovery gains after the Global Financial Crisis bottom in March. Each cycle carries its own unique macroeconomic backdrop that can easily overwhelm seasonal tendencies.

Looking ahead, investors weighing the “Sell in May” approach in 2025 will need to balance this statistical tendency against the current macro environment. With the Federal Reserve navigating a delicate path on interest rates, ongoing trade policy developments, corporate earnings growth projections, and a stock market that has already experienced significant volatility in early 2025, the summer months could prove turbulent. Market participants will be watching closely whether the post-election year pattern holds once again — or whether 2025 becomes another exception that reminds investors why rigid calendar-based strategies remain controversial among professional money managers.