Ah, the old trading adage: “Sell in May and go away.”
Summertime slowdown. Lower trading volumes and volatility. The beach and barbecues are calling.
Poppycock!
Even before the age of COVID-19, this phrase did not statistically hold water when one very important event took place later that calendar year:A U.S. presidential election.
Now, I hear you thinking, “Wait, how in the world can something six months away impact that old reliable (hah!) trading wisdom?”
Answer: I don’t know the “why,” just the facts.
Especially when you extend the timeframe to the end of the calendar year, and are able to take advantage of Rydex quarterly options, instead of the usual third Friday of the month expiry options.
What the Charts Show
We’ll take a look at the two oldest indices:
- The Dow Jones Industrial Average
- S&P 500 (Note: We’ll exclude the years prior to 1958 because it was not 500 stocks)
Dow Jones Industrial Average
Our beloved bear leader’s Slope
Charts data runs back to 1900 for the most quoted stock index in the news, while thinkorswim has Options Expiration charts that go back to 1904. So we’ll begin with the 1904 presidential election for data.
(Note: Actual options trading on the DJIA began in 1997, so we’re looking at the event impact before that year versus options possibilities.)
Here are the results, with the open price being the first trading day after the third Friday in May (aka, options expiry), and the close being the end of the calendar year (EOY):
1904 – Open 48.53, EOY Close 70.05
1908 – Open 75.12, EOY Close 86.15
*1912 – Open 90.34, EOY Close 87.87
1916 – Open 92.06, EOY Close 95
*1920 – Open 87.57, EOY Close 71.95
1924 – Open 89.81, EOY Close 120.51
1928 – Open 214.33, EOY Close 300
1932 – Open 53, EOY Close 55.40
1936 – Open 151.40, EOY Close 179.90
1940 – Open 122.40, EOY Close 131.13
1944 – Open 139.40, EOY Close 151.93
*1948 – Open 190, EOY Close 177.30
1952 – Open 259.90, EOY Close 291.90
1956 – Open 496.10, EOY Close 499.47
*1960 – Open 625.20, EOY Close 615.89
1964 – Open 826.20, EOY Close 874.13
1968 – Open 899, EOY Close 943.75
1972 – Open 961.50, EOY Close 1020.02
1976 – Open 988.80, EOY Close 1004.65
1980 – Open 826.90, EOY Close 963.99
1984 – Open 1133.80, EOY Close 1211.57
1988 – Open 1952.59, EOY Close 2168.57
*1992 – Open 3353.10, EOY Close 3301.11
1996 – Open 5688.20, EOY Close 6448.27
2000 – Open 10624.79, EOY Close 10787.99
2004 – Open 9968.02, EOY Close 10783.01
*2008 – Open 11843.83, EOY Close 8776.39
2012 – Open 12369.15, EOY Close 13104.14
2016 – Open 17507.04, EOY Close 19762.60
Total: In23 out of 29presidential election years, the Dow Jones Industrial Average closed higher than the open of the first trading day after the third Friday of May (aka, May Op
Ex).
That’s a 79% win rate for the bulls.
S&P 500
Now the S&P 500 has far fewer occurrences, since the index began trading 500 stocks in 1957, near the end of the Eisenhower administration. That makes the 1960 election of John F. Kennedy the first for data collection.
And here’s what its data looks like, again with the open price being the first trading day after the third Friday in May (aka, options expiry), and the close being the end of the calendar year (EOY):
1960 – Open 55.76, EOY Close 58.11
1964 – Open 81.10, EOY Close 84.75
1968 – Open 96.90, EOY Close 103.86
1972 – Open 108.98, EOY Close 118.05
1976 – Open 101.07, EOY Close 107.46
1980 – Open 107.35, EOY Close 135.76
1984 – Open 155.78, EOY Close 167.24
1988 – Open 253.02, EOY Close 277.72
1992 – Open 410.09, EOY Close 435.71
1996 – Open 668.91, EOY Close 740.74
*2000 – Open 1406.95, EOY Close 1320.28
2004 – Open 1093.56, EOY Close 1211.92
*2008 – Open 1425.28, EOY Close 903.25
2012 – Open 1295.73, EOY Close 1426.19
2016 – Open 2052.23, EOY Close 2238.83
(Note: Actual options trading on the S&P 500 began in 1983)
That’s 13 times in 15 presidential election years where the price of the S&P 500 ended the year higher than the open of the first trading day after the third Friday in May. An 86% win rate.
Now, the folks who are more trained in statistical analysis will point out that even 29 occurrences is still far too few in general for statistical significance. But we have the data we have, and the odds favor higher into the end of year from Monday’s open.