Stock market seasonality comes with the catchiest slogans in finance. Sell in May and go away. The Santa Claus rally. The September effect. They rhyme, they stick, and they get repeated so often that people start treating them as laws. Some of them do show up in the long-run averages. None of them are reliable enough to trade blind. The interesting question is not whether they exist, it is which ones survive a proper look and how much weight any of them can bear.
Sell in May and go away
The best known of the lot. The claim is that equities do most of their work in the winter half of the year, roughly November to April, and drift or stall through the summer half, May to October. Look at long histories of major indices and there is something to it. The average return in the winter months has tended to beat the summer months by a meaningful margin over many decades and across several markets, which is more than you can say for most calendar folklore.
But sit with what that actually means before you act on it. It is an average across many years, not a rule that fires each year. Plenty of individual summers have been perfectly good, and some have been excellent. Selling in May would have parked you in cash through some strong rallies. The effect is real in aggregate and unreliable in the specific, which is the recurring theme of all seasonality and the reason it belongs in the context column rather than the signal column.
The Santa Claus rally
The narrower version of this one has held up reasonably well: the last five trading days of December and the first two of January have shown a positive tilt more often than not across a long history. There are plausible reasons, from thin holiday liquidity and year-end fund inflows to tax-driven buying and a general seasonal lift in sentiment. It is one of the more consistent seasonal windows in equities.
The trap is the size of it. Even when the tilt shows up, the move is usually modest, and the window is only a handful of days. A pattern that is directionally reliable but small is easy to overtrade. The transaction costs and the risk of being caught on the wrong side of a headline can swallow the entire seasonal edge. Reliable and worth trading on its own are not the same thing.
The September effect
September has the weakest reputation of the calendar, and it is the one place where the negative seasonal story has some staying power. Across long histories September has tended to be the softest month for major indices on average. Nobody has a fully satisfying explanation, which is worth noting, because a pattern without a clean mechanism should always get more scepticism, not less. It may be partly a real behavioural effect and partly the kind of artefact that survives simply because it has been repeated into everyone's expectations.
What survives scrutiny, and what that is worth
Strip away the slogans and a few things do survive. The winter half tends to beat the summer half. The turn of the year has a positive lean. September tends to be soft. These are genuine tendencies in the averages. Here is the honest part that the slogans leave out.
The samples are small. One September per year means a few dozen observations even over a long history. That is thin evidence for a pattern you would risk real money on, and it means a handful of dramatic years can shape the whole picture.
The averages hide the spread. A negative average September is not a September that falls every year. The distribution around these averages is wide, and the win rate is often far closer to a coin flip than the tidy average suggests.
Regimes dominate. A bull market runs straight through the summer and shrugs off the September effect. A bear market ignores the Santa rally. The macro backdrop is a far heavier weight than the calendar, and any month can be overwhelmed by what is actually happening in the economy and in policy.
Data-snooping inflates everything. These patterns are the survivors of decades of people testing every possible calendar cut across every index. Some apparent effects persist purely because so many were tested. The famous ones at least have longevity and rough consistency going for them, which is more than the obscure ones can claim.
The turn of the month
One shorter-horizon pattern is worth a mention because it is among the more robust: equities have tended to do a disproportionate share of their gains around the turn of each month, roughly the last day or two and the first few days of the next. The usual explanation points to the mechanical drip of pension and retirement contributions being put to work on a schedule. Like the others it is a tendency and not a guarantee, and the effect is small enough to be swamped by anything the macro decides to do. Still, it is a cleaner example of the one thing that separates a signal from folklore: a dated, repeating flow you can actually name.
How to use equity seasonality properly
As a context filter, it is genuinely useful. Knowing you are heading into a historically softer stretch is a reason to be a little more selective, a little quicker to bank profits, a little slower to add risk. Knowing you are in a historically supportive window is a reason to give a trend a bit more room. These are dials you nudge, not switches you flip. The moment you turn a seasonal tendency into a mechanical buy or sell date with no other input, you have converted a mild statistical lean into a coin flip with extra steps.
The same discipline you would apply to forex seasonality patterns applies here without modification. Demand a mechanism, respect the small sample, and never judge a pattern by its average alone. Before you let any of these slogans influence a position, look at the win rate and the sample size next to the average return, because that trio is the only honest way to see whether a pattern is a signal or a story. The TradeSave+ seasonality explorer puts those three numbers together per instrument for exactly that reason, so a headline average cannot hide a thin, coin-flip pattern underneath it. Tracking that kind of breakdown is part of paying attention to the journal metrics that matter instead of the ones that merely feel good.
And if you want to know whether Sell in May would have actually beaten simply staying invested, after costs and after the good summers you would have missed, do not take anyone's word for it. You can test it yourself without code and see how much of the famous edge is left once real trading gets in the way. The slogans survive scrutiny as tendencies. They rarely survive it as strategies. Keep them as background colour, weigh them against the regime you are actually in, and let the market in front of you have the final word.