• Print
  • Default text size A
  • Larger text size A
  • Largest text size A

Best six months switching strategy: You can sell in May and go away, but open season on stocks begins in October

  • By Jeffrey A. Hirsch,
  • Wiley Global Finance WILEY GLOBAL FINANCE
  • Technical Analysis
  • Facebook.
  • Twitter.
  • LinkedIn.
  • Google Plus
Please enter a valid e-mail address
Please enter a valid e-mail address
Important legal information about the e-mail you will be sending. By using this service, you agree to input your real e-mail address and only send it to people you know. It is a violation of law in some jurisdictions to falsely identify yourself in an e-mail. All information you provide will be used by Fidelity solely for the purpose of sending the e-mail on your behalf.The subject line of the e-mail you send will be "Fidelity.com: "

Your e-mail has been sent.

The Best Six Months is basically the flipside of the old “sell in May and go away” adage. That phrase comes from an old British saw, “Sell in May and go away, come back on St. Leger Day.” Established in 1776, the St. Leger Stakes is the last flat thoroughbred horserace of the year and the final leg of the English Triple Crown. Apparently, once the British horseracing season concludes everyone can get back to the business of buying stocks.

While the St. Leger Stakes has little to do with stock market seasonality, it does coincide with the end of the worst months of the year for stocks. Market seasonality is a reflection of cultural behavior. Back when farming was the big driver of the U.S. economy, August was the best market month. Now that farming makes up less than 2% of the U.S. economy it’s one of the worst, as it falls during a time when traders and investors prefer the golf course, beach, or pool to the trading floor or computer screen. Institutions’ efforts in the fourth quarter to beef up their numbers can help drive the market higher, as does holiday shopping and an influx of year-end bonus money. This is followed by the New Year, which can tend to bring a positive “new-leaf” mentality to forecasts and predictions and the anticipation of strong fourth- and first-quarter earnings and drives the market higher into the second quarter.

Trading volume can decline throughout the summer and then in September there’s back-to-school, back-to-work, and end-of-third-quarter portfolio window dressing that has caused stocks to sell off in September, making it the worst month of the year on average. Although there may be some shifts in seasonality, the record still shows the clear existence of seasonal trends in the stock market.

The best six months delivers

Investing in the Dow Jones Industrial Average between November 1st and April 30th each year and then switching into fixed income for the other six months has produced reliable returns with reduced risk since 1950. Exogenous factors and cultural shifts must be considered. “Backward” tests that go back to 1925 or even 1896 and conclude that the pattern does not work are best ignored, as they do not take into account these factors. Farming made August the best month from 1900-1951, but since 1987 it is the second worst month of the year for Dow and S&P. Panic caused by financial crisis in 2007-08 caused every asset class aside from U.S. Treasuries to decline substantially, but the bulk of the major decline in equities that occurred during the worst months of 2008 was sidestepped using this strategy.

Figure 1 reveals that November, December, January, March, and April have been the top months since 1950. If you add in February, you have an impressive six consecutive month trading strategy. These six consecutive months gained 14,654.27 Dow points in 62 years, up 48 and down 14. May through October months lost 1,654.97 points, up 37 times and down 25.

David Aronson, author of Evidence-Based Technical Analysis (Wiley 2006), and his colleague Dr. Timothy Masters back-tested the Best Six Months Switching Strategy using their scientific method from 1987 (one year after the strategy was published in the 1986 Stock Trader’s Almanac) to April 2006. They found that from 1987 through April 2006 the S&P 500 generated an annualized return of 16.3% during the Best Six Months compared to 3.9% for the Worst Six Months. Though no back-test can predict future results, in Aronson and Masters' opinion the Six-Month Switching Strategy was sound, valuable, and had predictive power. The returns were considered to be statistically significant, unlike any of the 6,402 rules tested for the book.

1950s seasonality reversal

Figure 2 shows the one-year seasonal pattern of the Dow Jones Industrial Average over three timeframes since 1901. By examining this comparison of the Dow’s seasonal pattern during the first half of the Twentieth Century to the pattern since 1950 a major reversal in market seasonality is quite clear. In fact, before 1950 the better strategy appears to be “Buy in May,” the polar opposite of the present day. The 25-year period since 1988 illustrates that the Best Six Months/Worst Six Months pattern endures. On average the market rises from a low in October to a high in May, and then drifts sideways to lower from May to October.

Doing the math, the November-April $674,073 gain overshadows May-October’s $1,024 loss. Just three November-April losses entered double digits:

  1. April 1970: Cambodian invasion
  2. 1973: OPEC oil embargo
  3. 2008: Financial crisis

Similarly, Iraq muted the Best Six and inflated the Worst Six in 2003. When we discovered this strategy in 1986, November-April outperformed May-October by $88,163 to minus $1,522. Results improved substantially these past 26 years, $585,910 to $498.

Figure 3 shows the percentage changes for the Dow Jones Industrial Average along with a compounding $10,000 investment during the Best Six Months November-April versus the Worst Six Months May-October. It is hard to miss the market’s tendency to be flat to down from May through October, while posting most of its gains from November through April.

Timing is on our side

Using the simple Moving Average Convergence Divergence (MACD) indicator developed by Gerald Appel to better time entries and exits into and out of the Best Six Months period nearly triples its cumulative results since 1950. In up-trending markets, MACD signals get you in earlier and keep you in longer. If the market is trending down, however, entries are delayed until the market turns up and exit points can come a month earlier. Beginning October 1, you might look to catch the market’s first hint of an uptrend after the summer doldrums; beginning April 1, considering preparing to exit seasonal positions as soon as the market falters.

The results are astounding. Instead of $10,000 gaining $674,073 over the 63 recent years when invested only during the Best Six Months, the gain nearly tripled to $1,878,557. The $1,024 loss during the worst six months expanded to a loss of $6,723. Impressive results for being invested during only 6.3 months of the year on average.

MACD Worst 6 Months   MACD Best 6 Months   Buy & Hold
Signal May-October* Investing   Signal November-April* Investing     DIJA  
Date DIJA % Change 10,000   Date DIJA % Change 10,000   Year % Change 10,000
22-Apr-50 213.90 7.3 10,730   14-Nov-50 229.54 13.3 11,330   1950 17.6 11,763
10-May-51 260.07 0.1 10,741   13-Nov-51 260.41 1.9 11,545   1951 14.4 13,453
5-Apr-52 265.44 1.4 10,891   31-Oct-52 269.23 2.1 11,788   1952 8.4 14,586
30-Apr-53 274.75 0.2 10,913   23-Oct-53 275.34 17.1 13,804   1953 -3.8 14,036
14-May-54 322.50 13.5 12,386   5-Nov-54 366.00 16.3 16,054   1954 44.0 20,207
29-Apr-55 425.65 7.7 13,340   21-Oct-55 458.47 13.1 18,157   1955 20.8 24,404
9-Apr-56 518.52 -6.8 12,433   8-Oct-56 483.38 2.8 18,665   1956 2.3 24,957
9-May-57 496.76 -12.3 10,904   29-Oct-57 435.76 4.9 19,580   1957 -12.8 21,771
16-May-58 457.10 17.3 12,790   6-Oct-58 536.29 16.7 22,849   1958 34.0 29,164
5-May-59 625.90 1.6 12,994   6-Oct-59 636.06 -3.1 22,141   1959 16.4 33,946
22-Apr-60 616.32 -4.9 12,358   7-Oct-60 586.42 16.9 25,883   1960 -9.3 30,775
21-Apr-61 685.26 2.9 12,716   9-Oct-61 705.42 -1.5 25,495   1961 18.7 36,533
23-Apr-62 694.61 -15.3 10,771   10-Oct-62 588.14 22.4 31,205   1962 -10.8 32,584
1-May-63 719.67 4.3 11,234   18-Oct-63 750.60 9.6 34,201   1963 17.0 38,123
14-Apr-64 822.95 6.7 11,986   9-Nov-64 878.08 6.2 36,322   1964 14.6 43,678
19-May-65 932.12 2.6 12,298   26-Oct-65 956.32 -2.5 35,414   1965 10.9 48,432
2-May-66 931.95 -16.4 10,281   17-Oct-66 778.89 14.3 40,478   1966 -18.9 39,259
12-May-67 890.03 -2.1 10,065   21-Nov-67 870.95 5.5 42,704   1967 15.2 45,226
8-May-68 918.86 3.4 10,407   14-Oct-68 949.96 0.2 42,789   1968 4.3 47,157
21-May-69 951.78 -11.9 9,169   16-Oct-69 838.77 -6.7 39,923   1969 -15.2 39,992
15-Apr-70 782.60 -1.4 9,040   6-Nov-70 771.97 20.8 48,226   1970 4.8 41,919
3-May-71 932.41 -11.0 8,046   29-Nov-71 829.73 15.4 55,653   1971 6.1 44,481
24-Apr-72 957.48 -0.6 7,998   25-Oct-72 951.38 -1.4 54,874   1972 14.6 50,968
26-Apr-73 937.76 -11.0 7,118   11-Dec-73 834.18 0.1 54,929   1973 -16.6 42,515
26-Apr-74 834.64 -22.4 5,524   10-Oct-74 648.08 28.2 70,419   1974 -27.6 30,792
1-May-75 830.96 0.1 5,529   17-Oct-75 832.18 18.5 83,447   1975 38.3 42,593
5-May-76 986.46 -3.4 5,341   28-Oct-76 952.63 -3.0 80,943   1976 17.9 50,200
27-Apr-77 923.76 -11.4 4,732   31-Oct-77 818.35 0.5 81,348   1977 -17.3 41,531
9-May-78 822.07 -4.5 4,519   14-Nov-78 785.26 9.3 88,913   1978 -3.1 40,224
17-Apr-79 857.93 -5.3 4,280   5-Nov-79 812.63 7.0 95,137   1979 4.2 41,910
20-Jun-80 869.71 9.3 4,678   10-Oct-80 950.68 4.7 99,609   1980 14.9 48,168
1-May-81 995.59 -14.6 3,995   14-Oct-81 850.65 0.4 100,007   1981 -9.2 43,722
4-May-82 854.45 15.5 4,614   8-Oct-82 986.85 23.5 123,509   1982 19.6 52,293
13-May-83 1218.75 2.5 4,729   21-Oct-83 1248.88 -7.3 114,493   1983 20.3 62,891
11-May-84 1157.14 3.3 4,886   17-Oct-84 1195.89 3.9 118,958   1984 -3.7 60,539
1-May-85 1242.05 7.0 5,228   4-Oct-85 1328.74 38.1 164,281   1985 27.7 77,283
25-Apr-86 1835.57 -2.8 5,081   6-Oct-86 1784.45 28.2 210,608   1986 22.6 94,736
13-Apr-87 2287.07 -14.9 4,324   4-Nov-87 1945.29 3.0 216,926   1987 2.3 96,878
10-May-88 2003.65 6.1 4,588   12-Oct-88 2126.24 11.8 242,523   1988 11.8 108,358
8-May-89 2376.47 9.8 5,037   14-Nov-89 2610.25 3.3 250,527   1989 27.0 137,570
20-Apr-90 2695.95 -6.7 4,700   22-Oct-90 2516.09 15.8 290,110   1990 -4.3 131,597
26-Apr-91 2912.38 4.8 4,926   17-Oct-91 3053.00 11.3 322,892   1991 20.3 158,338
11-May-92 3397.58 -6.2 4,620   21-Oct-92 3187.10 6.6 344,203   1992 4.2 164,948
26-Apr-93 3398.37 5.5 4,874   7-Oct-93 3583.63 5.6 363,478   1993 13.7 187,582
13-Jun-94 3783.12 3.7 5,055   17-Oct-94 3923.93 13.1 411,094   1994 2.1 191,597
23-May-95 4436.44 7.2 5,418   23-Oct-95 4755.48 16.7 479,747   1995 33.5 255,689
17-Apr-96 5549.93 9.2 5,917   17-Oct-96 6059.20 21.9 584,811   1996 26.0 322,203
27-May-97 7383.41 3.6 6,130   18-Nov-97 7650.82 18.5 693,001   1997 22.6 395,155
24-Apr-98 9064.62 -12.4 5,370   13-Oct-98 7938.14 39.9 969,509   1998 16.1 458,772
13-May-99 11107.19 -6.4 5,026   20-Oct-99 10392.36 5.1 1,018,954   1999 25.2 574,481
13-Apr-00 10923.55 -6.0 4,725   23-Oct-00 10271.72 5.4 1,073,978   2000 -6.2 539,048
11-May-01 10821.31 -17.3 3,907   2-Oct-01 8950.59 15.8 1,243,666   2001 -7.1 500,801
1-Apr-02 10362.70 -25.2 2,923   2-Oct-02 7755.61 6.0 1,318,286   2002 -16.8 416,667
10-Apr-03 8221.33 16.4 3,402   3-Oct-03 9572.31 7.8 1,421,112   2003 25.3 522,176
20-Apr-04 10314.50 -0.9 3,371   4-Oct-04 10216.54 1.8 1,446,692   2004 3.1 538,614
1-Apr-05 10404.30 -0.5 3,354   17-Oct-05 10348.10 7.7 1,558,088   2005 -0.6 535,329
3-Apr-06 11144.94 4.7 3,512   2-Oct-06 11670.35 14.4 1,782,452   2006 16.3 622,534
14-May-07 13346.78 5.6 3,709   1-Oct-07 14087.55 -12.7 1,556,081   2007 6.4 662,562
14-Apr-08 12302.06 -24.7 2,793   20-Oct-08 9265.43 -14.0 1,338,229   2008 -33.8 438,616
21-Apr-09 7969.56 23.8 3,457   9-Oct-09 9864.94 10.8 1,482,758   2009 18.8 521,076
8-Apr-10 10927.07 4.6 3,616   4-Nov-10 11434.84 7.3 1,591,000   2010 11.0 578,395
13-Apr-11 12270.99 -9.4 3,277   6-Oct-11 11123.33 18.7 1,888,516   2011 5.5 610,206
3-Apr-12 13199.55 0.3 3,286   6-Nov-12 13245.68       2012 7.3 654,751
63-Year Gain (Loss)   (6,714)         $1,878,516     $644,751
  Average -1.2         9.3       8.1  
  # Up 33         53       45  
  # Down 30         9       18  

* MACD generated entry and exit points (earlier or later) can lengthen or shorten six month periods.
© Jeffrey A. Hirsch and Stock Trader's Almanac.

Putting the strategy to work

A more conservative way to execute this switching strategy, which I call the in-or-out approach, entails switching capital between stocks and bonds.

During the “Best Months” an investor or trader is fully invested in stocks. One inexpensive way to gain stock market exposure is through index-tracking ETFs and mutual funds. During the Worst Months you could then switch into Treasury bonds, money market funds, or a bear/short fund. Money market funds may be most conservative, but are likely to offer the smallest return. Bear/short funds offer potentially greater returns, but more risk. If the market moves sideways or higher during the Worst Months, a bear/short fund is likely to lose money. Treasuries can offer a combination of decent returns with limited risk.

In my opinion, this approach can work well for retirement accounts where the goal is to build wealth over time. Of further benefit, you will probably find summertime vacations and activities much more enjoyable because you will not be concerned with stock market gyrations while your nest egg is parked in cash or bonds. Since 1950, there have only been 9 years when the DJIA Best Six Months failed to delivery market gains.

Another approach to take advantage of this switching strategy involves making adjustments to your portfolio in a more calculated manner. During the Best Months additional risk can be taken as market gains are expected, but during the Worst Months risk needs to be reduced, but not entirely eliminated. There have been several strong “Worst Months” periods over the past decade, including 2003 and 2009. Taking this approach is similar to the in-or-out approach, however; instead of exiting all stock positions a defensive posture is taken. Weak or underperforming positions should be closed out, stop losses raised, new buying should be limited, and a hedging plan put in place. Purchasing out-of-the-money index puts, adding bond market exposure, or taking a position in a bear market fund could mitigate portfolio losses in the event a mild summer pullback manifests into something more severe such as a full-blown bear market.

JEFFREY A. HIRSCH is editor-in-chief of the StockTradersAlmanac.com and the author of The Little Book of Stock Market Cycles (Wiley, 2012). He is Chief Market Strategist of the Magnet AE Fund.

  • Facebook.
  • Twitter.
  • LinkedIn.
  • Google Plus
Please enter a valid e-mail address
Please enter a valid e-mail address
Important legal information about the e-mail you will be sending. By using this service, you agree to input your real e-mail address and only send it to people you know. It is a violation of law in some jurisdictions to falsely identify yourself in an e-mail. All information you provide will be used by Fidelity solely for the purpose of sending the e-mail on your behalf.The subject line of the e-mail you send will be "Fidelity.com: "

Your e-mail has been sent.
Article copyright ©2012 by Jeffrey A. Hirsch. Adapted from the Stock Trader’s Almanac with permission from John Wiley & Sons, Inc. The statements and opinions expressed in this article are those of the author. Fidelity Investments cannot guarantee the accuracy or completeness of any statements or data.
The data and analysis contained herein are provided "as is" and without warranty of any kind, either expressed or implied. Fidelity is not adopting, making a recommendation for or endorsing any trading or investment strategy or particular security. All opinions expressed herein are subject to change without notice, and you should always obtain current information and perform due diligence before trading. Consider that the provider may modify the methods it uses to evaluate investment opportunities from time to time, that model results may not impute or show the compounded adverse effect of transaction costs or management fees or reflect actual investment results, and that investment models are necessarily constructed with the benefit of hindsight. For this and for many other reasons, model results are not a guarantee of future results. The securities mentioned in this document may not be eligible for sale in some states or countries, nor be suitable for all types of investors; their value and the income they produce may fluctuate and/or be adversely affected by exchange rates, interest rates or other factors.
Backtesting is provided for educational purposes and as examples only, and should not be used or relied upon to make decisions about your individual situation. You should not assume that Backtesting of a trading strategy will provide any indication of how your portfolio of securities, or a new portfolio of securities, might perform over time. You should choose your own trading strategies based on your particular objectives and risk tolerances. Be sure to review your decisions periodically to make sure they are still consistent with your goals. Past performance is no guarantee of future results.
1. Solutions offered by Active Trader Services are available to investors in households that place 120 or more stock, bond, or options trades in a rolling twelve-month period and maintain $25K in assets across their eligible Fidelity brokerage accounts.
641383.2.0