In my previous article, I tried to show that a Sell in May strategy beat Buy and Hold for individual US stocks. Unfortunately there was a bug in my percentage gain calculation that significantly overstated the Sell in May profit.
Here, I'd like to show the details of the corrected strategy using the following high volume general market ETFs:
IWM
| Buy Date | Buy Price | Sell Price | Sell Date | Shares | Buy Amt | Sell Amt |
| 10/31/03 | 47.60 | 50.51 | 04/30/04 | 21008 | 1,000,000 | 1,061,134 |
| 11/01/04 | 53.50 | 53.05 | 04/29/05 | 19834 | 1,061,134 | 1,052,209 |
| 11/01/05 | 59.33 | 70.09 | 05/01/06 | 17735 | 1,052,209 | 1,243,036 |
| 11/01/06 | 69.79 | 76.33 | 05/01/07 | 17811 | 1,243,036 | 1,359,521 |
| 11/01/07 | 74.84 | 68.91 | 05/01/08 | 18166 | 1,359,521 | 1,251,798 |
| 10/31/08 | 51.11 | 46.93 | 05/01/09 | 24492 | 1,251,798 | 1,149,420 |
| 10/30/09 | 54.46 | 69.73 | 04/30/10 | 21106 | 1,149,420 | 1,471,705 |
| 11/01/10 | 68.34 | 85.12 | 04/29/11 | 21535 | 1,471,705 | 1,833,063 |
| 11/01/11 | 70.83 | 81.55 | 05/01/12 | 25880 | 1,833,063 | 2,110,494 |
Buy and Hold results can be calculated by subtracting a given sell price from the original buy price, dividing the result by the original buy price, and multiplying by 100.
Therefore the Buy and Hold profit for 10/31/03 - 5/1/08 = (68.91 - 47.60) / 47.60 * 100 = 45%
Sell in May results can be calculated by subtracting the Sell Amount from a selected sell date from the Buy Amount of the Buy Date, dividing the result by the Buy Amount, and multiplying the result by 100.
Therefore the Sell in May profit for 10/31/03 - 5/1/08 = 251,798 / 1,000,000 * 100 = 25%
Buy and Hold profit for the entire period = 71%
Sell in May profit = 111%
MDY
| Buy Date | Buy Price | Sell Price | Sell Date | Shares | Buy Amt | Sell Amt |
| 10/31/03 | 92.06 | 98.00 | 04/30/04 | 10862 | 1,000,000 | 1,064,523 |
| 11/01/04 | 101.99 | 107.65 | 04/29/05 | 10438 | 1,064,523 | 1,123,599 |
| 11/01/05 | 119.79 | 137.01 | 05/01/06 | 9380 | 1,123,599 | 1,285,118 |
| 11/01/06 | 133.37 | 151.06 | 05/01/07 | 9636 | 1,285,118 | 1,455,574 |
| 11/01/07 | 153.60 | 148.06 | 05/01/08 | 9476 | 1,455,574 | 1,403,075 |
| 10/31/08 | 99.69 | 98.48 | 05/01/09 | 14074 | 1,403,075 | 1,386,045 |
| 10/30/09 | 116.79 | 146.55 | 04/30/10 | 11868 | 1,386,045 | 1,739,232 |
| 11/01/10 | 148.19 | 182.43 | 04/29/11 | 11737 | 1,739,232 | 2,141,090 |
| 11/01/11 | 155.46 | 181.14 | 05/01/12 | 13773 | 2,141,090 | 2,494,771 |
B&H Profit 10/31/03 - 5/1/08 = 61%
Sell in May profit = 40%
Buy Hold for the entire period = 97%
Sell in May = 149%
SPY
Buy Date Buy Price Sell Price Sell Date Shares Buy Amt Sell Amt 10/31/03 89.29 94.87 04/30/04 11199 1,000,000 1,062,493 11/01/04 97.81 100.90 04/29/05 10863 1,062,493 1,096,059 11/01/05 105.90 115.68 05/01/06 10350 1,096,059 1,197,281 11/01/06 122.48 134.33 05/01/07 9775 1,197,281 1,313,119 11/01/07 137.70 129.98 05/01/08 9536 1,313,119 1,239,500 10/31/08 90.15 83.09 05/01/09 13749 1,239,500 1,142,430 10/30/09 98.92 114.10 04/30/10 11549 1,142,430 1,317,744 11/01/10 114.98 133.62 04/29/11 11461 1,317,744 1,531,370 11/01/11 120.70 140.74 05/01/12 12687 1,531,370 1,785,626
B&H Profit 10/31/03 - 5/1/08 = 46%
Sell in May profit = 24%
Buy Hold for the entire period = 58%
Sell in May = 79%
A total of 26 ETFs were analyzed. 13 were foreign, for example EEM, EWA, etc and 13 were US. There was no obvious advantage to using Sell in May on the foreign ETFs.
The US ETF Sell in May strategy underperformed a Buy and Hold from 2003 through May 2008. However, Sell in May clearly outperformed Buy and Hold from November 2003 through May 2012.
While XLF lost 31% during this period, Sell in May returned 22%. XLE Buy and Hold did over 132% better than Sell in May from 2003-2008 but was finally outperformed in 2003-2012 by only 12%.
Over the entire period Sell in May outperformed Buy and Hold by 15%.
ETFs that did well on Buy and Hold included QQQ, XLK, XLP and XLU. QQQ and XLK undoubtedly can credit their stakes in GOOG, and AAPL for much of this.
Disclosure: I have no positions in any stocks mentioned, and no plans to initiate any positions within the next 72 hours.
