Using Seasonal and Cyclical Stock Market Patterns

by Jeffrey Hirsch

Using Seasonal And Cyclical Stock Market Patterns Splash image

“Those who cannot remember the past are condemned to repeat it,” proclaimed philosopher George Santayana. I believe that “those who study market history are bound to profit from it.”

There are three main seasonal and cyclical patterns that have stood the test of time and consistently provide me with an edge in managing my portfolios: the four-year Presidential Election/Stock Market Cycle, the Best Six Months Switching Strategy and January’s basket of indicators and trading strategies.

In this article


About the author

Jeffrey Hirsch is chief market strategist at the Magnet Æ Fund and editor-in-chief of the Stock Trader’s Almanac. His latest book is the “The Little Book of Stock Market Cycles” (John Wiley & Sons, 2012).
Jeffrey Hirsch Profile
All Articles by Jeffrey Hirsch

But first, let’s get one thing straight. While I am a strong proponent of historical and seasonal market patterns, I am always mindful that history never repeats itself exactly. I have used history as a guide for navigating current market conditions and anticipating trends with quite a degree of success over the years. What we try to get Stock Trader’s Almanac traders and investors to do is not necessarily follow historical patterns to a “T,” but to keep them in mind so they know when their radar should perk up.

Politics, Politics, Politics

What happens on Wall Street is inextricably linked to what transpires in Washington. For five decades, the Stock Trader’s Almanac has discussed and demonstrated this phenomenon. The Four-Year Presidential Election/Stock Market Cycle is the “Old Faithful” of indicators for us.

Presidential elections every four years have a profound impact on the economy and the stock market. Wars, recessions and bear markets tend to start or occur in the first half of the term, with prosperous times and bull markets in the latter half. This pattern is most compelling.

As you can see in Figure 1, the third year in the presidential term has the best performance, as there have been no Dow Jones industrial average losses in pre-election years since war-torn 1939. While pre-election years have generally had greater gains, election-year market performance has weakened, thanks in part recently to the year 2000’s bear market and undecided election and the year 2008’s financial crisis.

Figure 2 illustrates that November, December, January, March and April are the top months since 1950. Add in February, and you have an impressive trading strategy. These six consecutive months gained 14,654.27 Dow points in 62 years, up in 48 years and down in 14, while the remaining May through October months lost 1,654.97 points, up 37 times and down 25. Figure 3 shows the average change in the Dow Jones industrial average for both the best and worst six-month periods.

Seasonal Portfolio Management

Use of the words “buy” and “sell” has created some confusion when used in conjunction with our Best Six Months Switching Strategy. They are often interpreted literally, but this is not necessarily the situation. Exactly what action an individual investor takes when we issue our official fall buy or spring sell recommendation depends upon that individual’s goals and, most importantly, risk tolerance.

Highlighted rows are post-election years.
S&P 500 Index Gain (%)
Santa Claus
First Five
February Last 11
1950 1.3 2.0 1.7 1.0 19.7 21.8
1951 3.1 2.3 6.1 0.6 9.7 16.5
1952 1.4 0.6 1.6 -3.6 10.1 11.8
1954 1.7 0.5 5.1 0.3 38.0 45.0
1958 3.5 2.5 4.3 -2.1 32.4 38.1
1959 3.6 0.3 0.4 -0.02 8.1 8.5
1961 1.7 1.2 6.3 2.7 15.8 23.1
1963 1.7 2.6 4.9 -2.9 13.3 18.9
1964 2.3 1.3 2.7 1.0 10.0 13.0
1965 0.6 0.7 3.3 -0.1 5.6 9.1
1966 0.1 0.8 0.5 -1.8 -13.5 -13.1
1971 1.9 0.04 4.0 0.9 6.5 10.8
1972 1.3 1.4 1.8 2.5 13.6 15.6
1975 7.2 2.2 12.3 6.0 17.2 31.5
1976 4.3 4.9 11.8 -1.1 6.5 19.1
1979 3.3 2.8 4.0 -3.7 8.0 12.3
1983 1.2 3.2 3.3 1.9 13.5 17.3
1987 2.4 6.2 13.2 3.7 -9.9 2.0
1989 0.9 1.2 7.1 -2.9 18.8 27.3
1995 0.2 0.3 2.4 3.6 30.9 34.1
1996 1.8 0.4 3.3 0.7 16.5 20.3
1997 0.1 1.0 6.1 0.6 23.4 31
1999 1.3 3.7 4.1 -3.2 14.8 19.5
2004 2.4 1.8 1.7 1.2 7.1 9.0
2006 0.4 3.4 2.5 0.05 10.8 13.6
2011 1.1 1.1 2.3 3.2 -2.2 -0.003
2012 1.9 1.8 4.2 4.1 8.7 13.4
2013 2.0 2.2 5.0
Average       0.5 12.3 17.4
Source: Stock Trader’s Almanac.


Over the years, there has been much debate regarding the efficacy of our January Barometer. Disbelievers in the January Barometer point to the fact that we include January’s S&P 500 change in the full-year results and that this detracts from the January Barometer’s predicative power for the rest of the year. In light of this debate, we calculated the January Barometer results with both the full-year results and the returns for the following 11 months (February through December). You can see these results, along with the S&P 500’s return for the Santa Claus Rally and the First Five Days in Table 1.

The Indicator Trifecta

The lack of a Santa Claus Rally has often been a preliminary indicator of tough times to come. This was the case recently in 2000 and 2008. A 4.0% decline in 2000 foreshadowed the bursting of the tech bubble and a 2.5% loss in 2008 preceded the second-worst bear market in history. There have been several instances in which a Santa Claus Rally preceded bad years or markets, so some caution is in order. This was the case in 2011, although the market did manage to recoup most of its losses to finish the year flat.

The last 40 up First Five Days were followed by full-year gains 34 times, an 85.0% accuracy ratio and a 13.6% average gain for all 40 years. In post-presidential election years, this indicator has a solid record. Just six of the last 15 post-election-year’s First Five Days showed gains. Only 1973 was a loser, coinciding with the start of a major bear market caused by Vietnam, Watergate and the Arab Oil Embargo. The other five post-election years gained 22.8% on average (1961, 1965, 1989, 1997 and 2009).

It’s incredible just how bullish it has been when all three indicators are positive. Since 1950, all three indicators have been positive 27 times and full-year gains followed 25 times. Losses occurred in 1966 (Vietnam) and just barely in 2011 (U.S. debt ceiling and European debt). Excluding January’s performance, the last 11 months of these years were up 24 times. The market’s crash in 1987 is the additional blemish on the record. Eleven-month average gains are impressive at 12.3%.

In 2013, the S&P posted its 17th best January gain of all time, completing the indicator trifecta. The January Barometer, Santa Claus Rally and First Five Days indicators were all positive this year—increasing the odds, but not guaranteeing, positive returns for 2013.

Jeffrey Hirsch is chief market strategist at the Magnet Æ Fund and editor-in-chief of the Stock Trader’s Almanac. His latest book is the “The Little Book of Stock Market Cycles” (John Wiley & Sons, 2012).


Fred from Utah posted about 1 year ago:

It would have been helpful if you had published this article in March or April

Walter from New Jersey posted about 1 year ago:

I may be totally confused, but it appears to me that the article totally contradicts what has occurred in the market during the full 5 months of January through May of this year. Apparently the article may have been written in January or February of this year, and if so there appears to be no correlation between the author's theorems and what stocks are doing today. Maybe by the end of June the analysis will begin to make more sense. But if I'm reading this correctly, it certainly does not bode well for the remainder of 2013.

wmdietz from PA posted about 1 year ago:

Great article picking up suggestions after the fact. How about suggestions for now, looking forward. This article is just an advertisement for the Almanac. Shame on AAII.

Bruce Campbell from NC posted about 1 year ago:

The author left just enough out of this article to make you want to read the book. I recommend it highly. And the Stock Market Almanac should be on everyone's book shelf.

david harned from va posted about 1 year ago:

If the data in the stock traders almanac is correct, it would appear that the most logical approach to increase total returns would be the exact opposite of that recommended in your "more conservative" recommendation. If one wants to "buy low/sell high", it would appear to make more sense to shift out of cash into stocks in the months of May, June, August & September, and sell off stocks & go to cash in January & April. This might make some sense for someone who simply wants to "play the market" and not focus on picking quality stocks for the long haul. Personally, I think it makes a lot more sense to buy quality stocks at the right price and hold them until there is a proper sell point...which may vary considerably among positions in a properly constructed portfolio.

Daniel Ballisty from CA posted about 1 year ago:

This is a basic premise, but market fluctuations are the result of money flows, and money flows tend to follow a seasonal character. I am not certain if this seasonality can be observed at each individual equity, but it is apparent at the DIA and S&P500 indices. I've heard some call it favorable/ unfavorable seasons. This year is unique in that there is additional liquidity brought to bear on the market through QE.

Werner Emmerich from PA posted about 1 year ago:

I can think of many other such guidelines, including: Bad Thanksgiving, good Santa Claus. It's a good sign, if recovery from a bear market exceeds 50%. Bear Market bottoms are steep, Bull Market tops are flat. Second biggest economy is ahead of the biggest, (Low interest rates in Japan 13 years ago vs. US rates more recently,) etc.

Tom from TX posted about 1 year ago:

This is a good article. To those that think it's not forward looking, remember that there is a December and a January in every year. And those of you who don't get the Stock Trader's Almanac, you should.

Harold Skelton from ME posted 10 months ago:

Give a drunken monkey a machine gun and 10,000 rounds of ammunition. Have him blaze away at a barn wall until every shell is fired. Euthanize the monkey. Go look at the side of the barn. 31% of the bullets struck the upper-right quadrant of the wall. Increasingly smaller percentages struck the other three quadrants. It is accurate to say that "in the data we examined, upper-right bullet strikes were the most common outcome." The problem arises when we make the further assertion that these past results are predictive of future outcomes.

Stock market results provide a rich field of data. Looking back at those results, it will always be possible to correlate any selected result to some contemporaneous event or condition. It does not follow that there is any causative connection between the two, or that the event or condition is predictive of future outcomes.

Then again, who's interested in reading an an article entitled "Seasonal and Cyclical Stock Market Patterns -- As Likely to Be Explained by Random Outcome as by any Other Cause."

Roger Mckinney from OK posted 9 months ago:

I think all of the indicators are good, but the election indicator is probably right for the wrong reasons. Federal spending is just 20% of the economy and many estimates of the multiplier are small, so I doubt it affects the economy that much. In addition, federal spending would affect profits only, but growth in the PE ratio, which indicates greater risk tolerance by investors, account for about half the variation in stock prices.

My guess is that what he calls the election cycle is really the business cycle, which averages 4 to 5 years over the long run. Understanding the business cycle and how it affects the stock market is import for avoiding major downturns in the market.

Edward Japhe from GA posted 6 months ago:

1- Am I correct in understanding that the years not shown in Table I on page 11 of this article were those years that the three indicators were collectively negative , since all of the shown years show them to be all positive?

2- Am I correct that although all three indicators can be positive, they of course cannot always result in a positive year, as illustrated particularly in the year of 1966 when the 11 months and full year were negative by 13.5% and 13.1 % respectively?

Charles Rotblut from IL posted 6 months ago:

Hi Edward,

The table is showing when all three indicators are positive. No timing indicator (or set of indicators) is routinely perfect. More importantly, just because a pattern has existed in the past, there is no guarantee that it will continue to exist in the future.


You need to log in as a registered AAII user before commenting.
Create an account

Log In