Fosback Seasonality Timing System

We would all like to have some edge when it comes to the markets. We have discussed some stock market timing approaches as one way to mitigate the market risk. However, one limitation almost all these market timing systems have is that they are the result of some type of backtesting or simulation. The result is that while it may look good on paper, it's difficult to have much confidence in their real time performance until they develop something of a track record.

There is one market timing system that is mechanical, and can easily be reproduced by almost anyone. It was originally published in the 1970's (I first read about it in Norm Fosback's Book Stock Market Logic about 1980). It's often referred to as the Fosback Seasonality system, and it's been tracked in real time by the Hulbert Financial Digest since 1982.

Mark Hulbert publishes updates on it regularly, even though Fosback himself didn't publish it for several years after he sold the newsletters that used to carry it. Since it is mechanical in nature, it was easily replicated. As of November 2006, Hulbert claims that over the previous 25 years, the system had returned about 13.7% annually, vs 12.9 % buying and holding the Wilshire 5000, but this was done by being invested in the stock market only about 1/3 of the time, so the risk is reduced substantially over buy and hold. You can read more of Hulberts commentary at marketwatch.com (search for Fosback).

When this system was first introduced the only real vehicle that could be used to trade it was the S&P futures, but these days tt can be done using ETF's, or some of the index mutual funds like Rydex of Profunds.

Trading the Seasonality System

There are basically two components to this system. The first is to simply buy and hold the two days before any of the exchange holidays. For example, buy and hold for July 2nd and 3rd, selling on the close of the 3rd, before the 4th of July. Do this for each of the exchange holidays. The second is the month end/ beginning period. The original system had you holding on the last 2 trading days of the month, and the first 4 trading days of the next month, for a 5 day holding period. There have been variants of the system published over the years, but I believe that was the original, and the one tracked by Hulbert.

The downside to the system is of course that there are about 17 trades in an average year. Some years the first of the month overlaps with Memorial Day or Labor day, so it's not exactly the same each year. But there are still a lot of trades to be taken.

One variant on this system that I like cuts down significantly on the number of trades, and has historically done fairly well. For the year end trades, I simply like to buy on the Tuesday before Thanksgiving (one of the normal buy periods) but to then hold until the close of the 4th trading day of the New Year. This captures the Thanksgiving holiday, the Christmas holiday, the New Years holiday, the first of the month of December, and the first of the month for January all with one trade.

Recognize that no system will guarantee positive returns, and there will always be risk. But if your goal is to shift the odds in your favor, this is a system that has a 25 year track record of doing exactly that. If you are looking to boost those returns, you can trade some of the leverage index funds like SSO or MVV, but be aware you are increasing the risk as well when you do that.

Recently Norm Fosback has "updated" his system. You can read more about it at http://www.fosback.com/seasonality_system.htm