Cycles
We are due to discuss time cycles in Module 9, where we also consider the Elliott Wave. However, moving averages are bound up in cycles, so I need to mention them here. Many people have observed that prices move in cycles, and one of the most obvious cycles is the monthly cycle in the commodities market, where contracts are settled each month. In terms of trading days, this is a 20 or 21 day cycle. Cycles also tend to be related to longer and shorter cycles by a factor of two, so from a 20 day cycle you would also look at a 10 day and a 40 day. You will find this concept applied throughout when traders consider the number of days for their moving average analysis.
Moving averages are, essentially, smoothed out price lines, and they can be used to identify the cycles of a particular market. You should look for cycles on a three or four month basis, for example silver futures typically trade in a 13 week cycle. But the Dow Jones Industrial Average has a dominant short-term cycle that is four months long. For a given stock or commodity, the length of cycles is usually fixed, so you wouldn’t have a three-month cycle followed by a four month cycle. The 30 day moving average is a useful tool to smooth the price line and allow you to see where these cycles are occurring.
Weekly Rule
Talking about cycles, there’s a trend following system that’s even easier than the moving average, and that has good results on the futures market. In its original form it’s called the four-week rule, and it was recognized in 1970 when various commodity trading systems of the day were compared and it was found to be the most successful. Now this was at a time when they didn’t have our advantage of computer processing power, but they found that channel breakout systems such as the four-week rule and moving average crossover systems were the best.
The four-week rule simply states that
- when the price is more than the previous four week highs, you buy long and cover your shorts
- when the price is below the previous four week lows, you sell your long positions and enter short ones.
As you can see, you trade either long or short positions all the time, which means you will always be in one or the other. Because it is a trend following system, and markets don’t trend all the time, the original four-week rule is likely to get you whipsawed in and out of positions during a sideways or range bound period. One suggestion is to modify it by using a shorter time span to exit the trades. You would still need a breakout over the last four weeks to enter a trade, but you could exit after one or two weeks in the opposite direction. You then wouldn’t take up another position in the market until the price was higher or lower than all the last four weeks.
The reason it works is because it is clearly trend following, and requires you to be in a trade on the right side of every trend that happens. It has the incidental advantage that it doesn’t over trade your account, so you don’t waste too much in commissions. It’s like all trend following systems, in that it will never catch the absolute top or bottom of the market, but I don’t think a system’s been invented that can do that consistently. It’s certainly easy to follow, with or without a computer, and unambiguous.
Although the four-week rule works well as it seems to be related to a cycle of the commodity markets, there is nothing to stop you trying the same principle over different lengths of time. If you chose a shorter time span, such as two weeks, the system would be more sensitive and make more trades, so you need to experiment to see what works best. If you thought that the market was mainly staying in a range, you could instead extend the number of weeks required, so to eight, so that you would be looking at a genuine breakout when you did trade.
Another way to use this system is simply as confirmation for another trading signal. Before making the intended trade, you would consult this rule to see whether the market was moving in the correct direction. The four-week rule is not sophisticated, and it is easy to understand. It’s probably not worth trying to fine tune the system from four weeks (20 days) to 19 days or 21 days, but a change such as suggested above to two weeks would make a difference, and you would have to see whether it was an advantage for the market you were trading.
Join the discussion