Be careful when using seasonality to trade and invest


Seasonality is the simplest of historical studies. E.g. What is the AVERAGE return for market XYZ in May over the past 50 years? Over the past 5 years, what is the average return for the S&P 500 during the first day of each trading year?
The media loves to talk about seasonality. It’s easy for everyone to understand and it’s very simple to calculate. Just take an average of the market’s performance over a “season” (e.g. every October) to determine the market’s seasonality. There’s no skill involved.
Traders and investors tend to overemphasize the importance of seasonality in trading. Seasonality should be a small and relatively unimportant factor in the long list of factors that you consider before making investment and trading decisions. Here’s why.

Seasonality is often random

The first problem with seasonality is that it might be random. Just because “the stock market is historically bearish in June” doesn’t mean that there’s any meaningful reason for this seasonality.
*The S&P’s average return in June is -0.183%. This is hardly any different from zero. There are no fundamental reasons as to why June is seasonally bearish.

With a large enough pool of data, there will always be months when the market is “seasonally bullish” or “seasonally bearish” for no reason other than randomness. Some months will on outperform and other months will underperform. That’s how averages work. You need to have above-average months for there to be below-average months.
Sometimes seasonality exists for a reason. For example, the U.S. stock market is seasonally weak in February because U.S. economic data tends to underperform in February (winter + cold). There is a weak-moderate short term correlation between the stock market and economic data.
If there is a real reason for this seasonality, I would rather know what the reason is than blindly trading on seasonality. That way I can trade according to the reason and not the seasonality. E.g. if the economic data is extremely strong this February, then I won’t have to worry about weak seasonality for stocks.

Short term seasonality is even more useless than long term seasonality

Some traders like to argue e.g. “the first day of each months is more bullish than bearish”. Then they invent BS reasons such as “fund managers like to buy on the first day of each month”. These “reasons” have zero evidence and are just made up to “support” the short term seasonality.
Short term seasonality is practically useless. For starters, the market doesn’t usually move much in the short term. Short term seasonality such as “the market rises an average of 0.1% on Options Expiration Day” isn’t very useful for trading. Unless your trading strategy involves picking up pennies at a time, you can’t really use short term seasonality to trade.
In addition, short term seasonality is even more random than medium term seasonality. If you really think about it, why should any day of the week (e.g. Tuesday) be more bullish or bearish than any other day of the week (e.g. Friday)? It doesn’t make any sense. These average “underperformance” and “outperformance” is usually the result of statistical randomness.

Seasonality can change over time

The worst part about seasonality is that it can change over time. For example, the “sell in May and go Away” seasonality used to be exactly the opposite.
“Sell in May and go Away” states that the stock market is seasonally weak from May-October and seasonally strong from November-April. This was only true AFTER the 1950s. This seasonality was completely opposite before the 1950s. The stock market was seasonally strong from May-October and seasonally weak from November-April.
Here’s how some people explain this seasonality change:

The stock market tends to be strong when U.S. economic data is strong. The U.S. is now an advanced economy. The data tends to deteriorate over the summer months when people are on vacation. The U.S. economy used to be an agricultural/manufacturing based economy in the early 1900s. People worked during the summer/fall and stayed indoors during the winter/spring.

You can’t use seasonality to trade because you don’t know when it will change until it has already changed (i.e. AFTER seasonality doesn’t work).

Seasonality is based on an AVERAGE over an extremely long period of time

Some traders get frustrated when they fail after trading via seasonality. For example, the stock market is seasonally strong in April. They will be extremely upset if they go long stocks in April because of “strong seasonality” and the stock market FALLS this April.
Here’s the reality. Seasonality is an AVERAGE over a very long period of time. E.g. you will outperform the stock market if you buy every April over the past 30 years. But that doesn’t mean you won’t lose money if you long stocks this April and next April. It’s an extremely long term average.
So you cannot use seasonality to trade unless your trading time frame is 30+ years. Otherwise you might go through years of losses if the market isn’t moving according to seasonality.
Seasonality can be skewed by a few outlier cases. For example, the stock market’s seasonality is “weak” in September because a few big crashes happened in September. But on the other hand, a few large historical rallies also happened in September. A few big outliers tend to have a reason (e.g. the stock market crashed in September 2008 because the economy crashed and not because of “bearish seasonality”).
Reading too much into seasonality will be detrimental to your trading.

5 comments add yours

  1. Agree 100% Troy. Amazing how often a downturn or uptick is “signaled” due to the date on the calendar.

    • The media tries to find reasons for each and every movement, when in most cases there are none. Seasonality is the “easiest reason” that everyone can understand.

  2. Like I told u before Troy, 2/3rds of retail consumer spending make up the money flows for the economy. The psyche for the market is ‘usually” up and a big rally ensues leading up to the 4 major U.S. Holidays . Memorial Day ,4th of July, Labor Day, thanksgiving. If the market is bearish “selling off”leading up to these 4 major U S holidays . Then consumers will be less likely to spend. I have studied this pattern for years. There is a very high probability of stocks rallying leading up to the holiday. 1-2 weeks before

  3. Hi Troy,
    Could you give us your views on the effect of distribution dates?
    e.g. a lot of companies pay distributions in June, what effect does that have?
    Does it lead to a trend of rising prices in May/June?
    Thanks.

    • I think it’s mostly a nonfactor for the stock market. May-June are seasonally flat months for the stock market.

Leave a Comment