Market Seasonality: "Sell in May," the Santa Claus Rally, and Other Patterns

Markets love stories, and one of the most enduring is that the year has a "rhythm." That certain months bring more, and others — less. And indeed: when you gather decades of historical data and arrange it by calendar, patterns surface. November and April shine, September limps, and around Christmas a little rally often appears. These patterns have names, history and even their own folklore.

What matters is how you read them. Seasonality is a historical tendency, computed backward in time — not a forecast for the specific year in front of you. In this article you will see where the most famous seasonal patterns come from, what the data actually shows and — most essentially — why this is a curious educational fact, not a timetable to trade by.

What seasonality is — and what it is not

Seasonality is a simple idea: you take a long history of a stock-market index, group the returns by calendar period (month, week, even a single day) and look at the average value across the years. If you gather every September from the last three decades and average them, you get the "typical" September.

Here is the first trap. The average is a summary of many different years, not a description of any one of them. An "average September of −0.5%" does not mean September will fall — it means that, looked at backward as a group, Septembers are a little weaker than the other months. One September was sharply down, another confidently up; the average between them says surprisingly little about the next.

That is why the right way to read seasonality is as a curious historical pattern, not as a train timetable. Selurio shows exactly this data in the monthly heatmap on the Market page — the 12 months, colored by the average historical return of the S&P 500 index for the period 1994–2026. Here they are in a table too:

Three things stand out immediately: November is the strongest month (+2.6%, positive in 78% of the years), April is close behind, and September is the only one with an average decline. It is precisely these observations that give rise to the named patterns.

"Sell in May and go away"

The most famous seasonal pattern is British in origin and at least as old as the 1800s. The full old saying goes: "Sell in May and go away, and come back on St. Leger's Day" — sell in May, disappear for the summer, and come back for St. Leger's Day (an annual horse race in mid-September). The logic then was social: the London financial elite left the City for the summer and activity died down. The modern American version simply says "come back around November" — which is why the pattern is also known as the Halloween indicator.

Behind the slogan stands a real historical tendency: the period November–April on average outperforms May–October. In Selurio's data the difference is exactly that — the winter half brings about +1.2% average monthly return, and the summer half about +0.7%. Winter wins.

But note the nuance the slogan hides. "Sell in May" sounds as if May is bad — yet in this data May itself is perfectly decent (+1.2%, positive in 70% of the years). The weakness of the summer half comes almost entirely from one month — September. In other words, "Sell in May" is a catchy way to describe a true average, but it is a blunt shovel where reality calls for tweezers.

The Santa Claus rally

The second famous pattern has a strictly defined window: the last 5 trading days of the year plus the first 2 of January. The term was introduced by Yale Hirsch in his "Stock Trader's Almanac" in 1972, and the historical average return for this seven-day window is about +1.3% — noticeably more than a typical random seven-day stretch.

Many explanations are offered — festive mood, the investing of annual bonuses, fewer active institutional sellers during the holiday, end-of-year portfolio adjustments — but none is proven. Hirsch himself left a well-known warning too: "If Santa Claus should fail to call, bears may come to Broad and Wall" (if the year-end rally fails to appear, a falling market may arrive at the corner of the exchange). A pretty rhyming omen — but it remains folklore, not a mechanism. The absence of one rally in one December is not a signal of anything specific for the following year.

The January effect — and why it fades

A related pattern is the so-called January effect: a historical tendency for stocks (especially those of smaller companies) to perform more strongly in January. The classic explanation is tax- and calendar-related — in December investors sell losing positions to realize losses before year-end, and in January they buy back, which lifts prices. Add the reinvested annual bonuses and the "cleaning up" of portfolios for the annual reports.

And here is the instructive part: the January effect weakened drastically over recent decades. The reason is its very fame. When enough people know about a pattern and start to front-run it, they eat it up — the buying moves earlier, the effect dissolves. In Selurio's data January today is solid but not exactly exceptional (+0.8%, positive in 61% of the years). This is the first big hint of why seasonality is hard to turn into money: well-known patterns self-destruct.

Why September is the historical weakling

If there is one true "character" in the calendar, it is September — the only month with an average decline in the lineup (−0.5%). The theories are again several: investors return from the summer and rearrange portfolios; many mutual funds close their fiscal year at the end of October and sell off losing positions in September–October; the historically greater volatility of autumn adds nerves. But there is no single proven cause — the September weakness is one of the honest market anomalies, more speculated about than known.

And here the numbers call for attention: although the average is negative, September still ends positive in 56% of the years. That is, in most Septembers the market rises — the average is dragged down by a few very heavy months. If you sell every September "just in case," in more than half the years you will simply miss out on gains.

Here is a summary of the four patterns side by side:

Why this is NOT a timing strategy

Up to here the patterns sound almost like a free lunch: exit in May, come back in November, buy before Christmas. On paper — tempting. In practice — a way to systematically harm yourself. Here is why.

• Transaction costs. Every entry and exit carries a spread and fees. A single "Sell in May" means at least two trades a year; the more aggressive versions — many more. The historical difference between winter and summer is small (about 0.5 percentage points per month) and melts quickly under the weight of costs.
• Taxes. Every sale is a potential tax event and an administrative burden. Frequent "churning" of positions can turn a paper gain into a realized, taxed gain — with a buy-and-hold investor it would simply have kept growing untaxed. Tax rules depend on the instrument and the jurisdiction, but the direction is clear: timing adds friction.
• The missed best days. This is the killer. The best and the worst trading days come clustered together — right in the middle of exactly the turbulence that would tempt you to leave. Frequently cited calculations for the last two decades show that if you were invested the whole time but missed just the ten or so best days, your final result could be halved. And a large part of those best days come only a few days after the worst ones. If you step out "for the bad ones," you risk missing exactly the rebound.
• The average is not a promise. A share of positive years of 56% for September means the pattern is unreliable for any specific year. Thirty-odd observations are a small sample; much of the "signal" is plain noise.
• Patterns self-destruct. The fate of the January effect is the warning. If a pattern wins too many followers, they themselves eat it up.

In other words: seasonality explains the past average well, but is a weak instrument for betting on the next year. It is a thermometer of history, not a crystal ball.

Time in the market, not timing

The takeaway is not "seasonality is a fiction" — the tendencies are real and thoroughly deserve your curiosity. The takeaway is that the distance between "an interesting historical pattern" and "a reliable strategy for profit" is enormous, and most people cross it only to harm themselves.

The more boring approach almost always wins: stay invested and contribute regularly, instead of guessing months. That is exactly the idea behind regular investing (DCA) — you remove the question "is now the moment," because the moment is every month. If the jolts along the way worry you, the answer is not timing, but the right balance between risk and return and diversification, which smooth out seasonal and all other swings.

Look at the monthly heatmap in the Market the way you would look at an old chart of sea currents: it tells you where the water historically flowed a certain way. Useful for understanding, beautiful to look at — but not a promise of which way the wind will blow tomorrow.

Sources

• Britannica Money — "Sell in May and go away": how the saying arose and what the data shows for the winter versus summer halves: britannica.com/money/sell-in-may-and-go-away (opens in new tab)
• Britannica Money — "Santa Claus rally": a definition of the window, the origin with Yale Hirsch and the historical average return: britannica.com/money/santa-claus-rally (opens in new tab)
• Corporate Finance Institute (CFI) — "January Effect": the mechanism of tax-motivated selling and why the effect fades: corporatefinanceinstitute.com/resources/economics/january-effect (opens in new tab)

The figures in the tables are historical average values for the S&P 500 index (period 1994–2026), rounded to one decimal — a historical tendency, not a forecast for any specific year.

Continue reading

• DCA: why regular investing beats timing — the more boring but more profitable answer to the question "is now the moment."
• Risk and return — why swings are the price of returns, not an enemy to avoid.
• Diversification — how to smooth out seasonal and all other tremors.
• Volatility — what really measures the "nervousness" of the market that makes timing so treacherous.

Subscribe

Subscribe to the newsletter and get new calculators and articles straight to your inbox — no signals and no spam.