To understand the importance of first arriving at a sound theory before implementing and testing a trading program, we need to briefly review the characteristics of performance that indicate a robust method.
When testing a trend-following system we should expect that a trend of 100 days, compared with a trend of 50 days, will produce larger profits per trade, greater reliability, and proportionally fewer trades. As you increase the calculation period, this pattern continues; when you reduce the calculation period this pattern reverses. You are prevented from using very short calculation intervals because slippage and commissions become too large; the longest periods are undesirable because of large equity swings. There must be a clear, profitable pattern when plotting returns per trade versus the average holding period.
Each time frame has a logical purpose and is said to be modeled after Gann's concept that the markets are essentially geometric. The shortest time frame is the one in which you will trade. In addition, there are two longer time frames to put each one into proper perspective.
The patterns common to time frames are easily compared with fractals; within each time frame is another time frame with very similar patterns, reacting in much the same way. You cannot have an hourly chart without a 15-minute chart, because the longer time period is composed of shorter periods; and, if the geometry holds, then characteristics that work in one time frame, such as support and resistance, should work in shorter and longer time frames. Within each time frame there are unique levels of support and resistance; when they converge, the chance of success is increased. The relationships between price levels and profit targets are woven with Fibonacci ratios and the principles of Gann.
One primary advantage of using multiple time frames is that you can see a pattern develop sooner. A trend that appears on a weekly chart could have been seen first on the daily chart. The same logic follows for other chart formations. Similarly, the application of patterns, such as support and resistance, is the same within each time frame. When a support line appears at about the same level in hourly, daily, and weekly charts, it gains importance.
1. Every time frame has its own structure.
2. The higher time frames overrule the lower time frames.
3. Prices in the lower time frame structure tend to respect the energy points of the higher time frame structure.
4. The energy points of support/resistance created by the higher time frame's vibration (prices) can be validated by the action of lower time periods.
5. The trend created by the next time period enables us to define the tradable trend.
6. What appears to be chaos in one time period can be order in another time period.