Stats › Seasonality and calendar › Seasonality

S&P 500 seasonality

Returns/monthly · calendar effects · 1950 to 2025 · price return · 6 sources · data as of 30 Jul 2026

Some months of the year have, on average, gone better for the stock market than others. That's a true statement, it's been true for seventy five years, and it's also the most over read statistic in the whole business. Both of those things can sit together, and this page is my attempt to hold them at the same time.

What's on the page: every calendar month since 1950, with two different measures side by side, the average return and how often the month actually finished positive. Those two disagree more than you'd expect, and the disagreement is the most useful thing here. Then the four named calendar effects that get built on top of the monthly data, sell in May, the Santa Claus rally, the January barometer and the presidential election cycle, each with the numbers its own advocates publish and the objections that come with them. I wrote it because seasonality is where a lot of people first meet the idea that markets have patterns, and it's very easy to walk away with the wrong lesson. The gaps between months are real. They are also small, much smaller than the range of outcomes inside any single month, and a decent number of the famous effects lean on arithmetic rather than on prediction.

Six sources, all listed at the end. Where a claimed effect has a promoter, I've quoted the promoter's own figures rather than a critic's, so nobody can say the numbers were shaded against them. The objections are mine, the numbers are theirs, and you can check both.