|Using Market Cycles|
|Using Market Cycles|
|Written by Dale Gilham|
|Tuesday, 01 September 2009 13:00|
Many of you would agree that the basis of trading is to time your entry and exit so as to extract a reasonable percentage of available profit from a movement in the price of a market. However, if this is true, why then do so few traders actually use time analysis in their trading? In my experience, many traders believe they are using ‘time analysis’ to enter the market, when in fact they are relying on the more traditional methods of price and pattern to time their entry.
Undoubtedly, price and pattern do have their place in trading, but only when the ‘time’ is right. In fact, when you use time as your primary analysis tool and then apply price and pattern to aid in fine tuning your entry and exit signals, you will eliminate many of the false, late or inconsistent signals that cause traders to lose money.
Many times in other articles I have used Gann’s time projections to demonstrate how you can use this method of analysis in combination with price and pattern to trade the market. While this method of analysis works well over all time frames, it is used primarily for shorter term trading to assist in narrowing down the entry and exit points of a trade to within days of a high or low. Therefore, in this article I want to introduce you to my preferred method of timing, which although not well known, is highly effective and based on the work of Walter Bressert and Ray Merriman.
Walter Bressert is credited with introducing ‘timing cycles’ into the Futures market in the late 1960’s.
Bressert found that markets moved from low to low in measurable time frames which he called ‘cycles’. As we have identified with Gann’s theory of time analysis, cycles are intervals of time that can be repeated to assist in predicting turning points in the market.
According to Ray Merriman, timing cycles are a measurable phenomenon that occurs consistently at regular intervals of time. The phenomenon does not have to occur with 100% regularity, however it should unfold with an 80% consistency or greater during a regular interval of time. A regular interval of time does not mean on a specific date or in a specific week. It means within an interval, or ‘orb of time’. For instance, a six week cycle does not mean the cycle unfolds exactly every six weeks, every time; it means a cycle of 5-7 weeks unfolds at least 80% of the time.
Figure 1 shows how a 26 week cycle on CBA unfolded between Sep 99 and Oct 02. Notice that each of the cycles ranged in length from 25 to 29 weeks, which is well within the allowable time frame of 22 to 30 weeks for the lows to occur, which represents a consistency of 100%.
Cycles range from many years down to minutes. Common cycles include 54 years, 18 years, 9 years, 4 years, 22 months and yearly. Notice that these cycles are generally derivatives of each other and unfold to fit within the cycle of larger degree. Once we get below a yearly cycle we change the way we label cycles. For example, a primary cycle is 16-26 weeks, a major cycle is 4-7 weeks and a trading day cycle is 4-7 days. It is important to remember, that the longer the cycle the more dominant it is. For example, a 4 year cycle is more dominant than all the cycles below it.
As I indicated above, cycles are measured from low to low, and have an average time band, plus or minus a period known as the orb, which is 1/6th of the cycle length. In an 18 week cycle, 1/6th of 18 weeks is three weeks.Therefore as shown in Figure 2, the cycle has a time band of 15 to 21 weeks in which the low will occur.
To demonstrate how effective this form of analysis can be I will now apply cycles to the history of the All Ordinaries Index. In so doing I will show you how you can use this method of analysis to predict where the market is heading in the future.
Predicting the All Ords
When analysing a market or stock you need to look at the complete history to identify the major lows and then work forward in time (using smaller cycles). Because we only have reliable data on the All Ords back to 1982, I will use the Dow Jones Index to highlight how the cycles have unfolded on our market, given that the All Ords tends to the follow the Dow Jones in its major moves.
Figure 3 shows the decline on the Dow Jones from the market highs in 1929, which lead to the Great Depression, into the low in July 1932. This low was in fact a 54 year cycle low, as the previous major crash on the US market was in 1873 some 59 years earlier, which is well within the allowable time frame for a cycle low given that 1/6th of 54 years is 9 years (45 years to 63 years).
If we use July 1932 as our starting point, we can predict the next 54 year cycle low on our market, which would be due in 1986 plus or minus the orb period of 1/6th (9 years) of the cycle length. Given this, we know with high probability that the cycle low will occur sometime between 1977 and 1995. And as we already know, the next major crash on the Dow Jones occurred on 29 October 1987 or 55 years on from July 1932. The actual low on the All Ords occurred on 11 November 1987 as shown in Figure 4 below.
If we project out another 54 years from 1987, then we can expect the next 54 year cycle low to occur in 2041 plus or minus 1/6th of the cycle length (between 2032 and 2050).
The next common cycle after the 54 year cycle is the 18 year cycle. It is important to note that with an 18 year cycle low, the market does not crash like the 54 year cycle; rather it is a slow decline into the cycle low that takes place over a period of approximately 3 years. In my research I have found that in our market that the 18 year cycle is around 20 years in length. To calculate when the next cycle low is due we simply add 20 years onto the low in 1987 to arrive at 2007, plus or minus 1/6th (3.33 years) of the cycle length. Therefore, the next 18 year cycle low on the All Ords is due sometime between late 2003 and early 2011.
With this in mind and knowing that there is a high probability that the low will occur before the end of February 2011, which is the outer most limit of our ‘orb period’, then the market will need to peak no later than the end of 2007 after which we would expect a long term bear market of around three years.
Although that said, it is possible that the 18 (20 year) year cycle concluded when the market formed a significant low in March 2003. However, the decline from the high in 2002 only lasted for 13 months which doesn’t comply with cycle’s theory, because the market needs to decline over a period of approximately 3 years. Therefore, the most probable scenario is that the 18 (20 year) year low has not yet occurred, and the likelihood is that we are over half way into the move down into the low. Is it possible that March 2009 was the low? Yes as it fits within our orb period for the low, however we need to be aware the market only fell for 16 months from thehigh in November 07. For now, only time will tell whether the current move up is simply a bear market correction, or the start of the next 18 year cycle.