How to Use Price Channel to Trade Trend Following Energy Futures – Technical Analysis and Trading System

In this article, we will explore one of the most popular and widely used indicators in the trading world: the Price Channel. We will delve into how this technical analysis tool works and how it can be used to identify opportunities for systematic trading. In particular, we will focus on the energy futures sector, examining Crude Oil (CL), Heating Oil (HO), Natural Gas (NG) and RBOB Gasoline (RB). These financial instruments are listed on the NYMEX (New York Mercantile Exchange) and are known for their trend-following nature. In other words, their prices tend to move in long upward and downward trends from which profits can be made by following their direction.

Want help improving your trading approach? Click here >>>

Logic and use of the Price Channel indicator

This indicator, also known as Donchian Channel, having been created by Richard Donchian in the 1950s, is formed by two parallel lines, an upper one and a lower one. The upper line is calculated by identifying the highest value of the last N bars, while the lower line is calculated in the opposite way, identifying the lowest value of the last N bars (Figure 1).

This channel can be used in different ways, but in this article we will delve into the traditional approach, that is, trend following, given the characteristics of the markets we are going to analyze.

Developing a Price Channel-Based Trend Following Strategy on Energy Stocks

So let’s start by testing a trend following strategy, which is actually more intuitive using this channel: we will buy when the upper channel is reached and sell when the lower channel is reached, since a continuation of the price in the same direction is expected. Furthermore, once the levels are reached, the position will be reversed from long to short and vice versa.

This strategy will be applied to the portfolio composed of the futures listed above, on a 60-minute time frame. To calculate the indicator, we will use the default value found on the main platforms, i.e. 20 bars. In addition, to carry out backtests and ensure the correct functioning of the platform, we will set an initial capital of the portfolio equal to $1,000,000 for purely academic purposes.

Performance of the trend following strategy based on the Price Channel

Analyzing the backtest results starting from 01/01/2010 we note, from the equity line in Figure 2, an excellent initial result, especially considering the regularity of the curve. Despite the simplicity of the strategy, over the years a net profit of just over 1,000,000 dollars has been achieved, with a decidedly limited drawdown of only 96,000 dollars.

We also note from Figure 4 that this strategy has generated profit on all four futures considered for this test. The results obtained on the Heating Oil future and the RBOB Gasoline future are higher than the results on the other two futures, which however remain positive, thus confirming the trend following nature of these underlyings.

However, analyzing the Total Trade Analysis (Figure 5), and more specifically the overall average trade of the strategy applied to the portfolio, a value of only 89 dollars emerges, which is certainly not large enough to cover the commission costs and slippage that would be incurred by operating on these underlying assets.

Optimizing the number of periods to calculate the Price Channel

It should be noted that we have used 20 as the value for calculating the upper and lower levels of the Price Channel, since they are the default values ​​on the various platforms. At this point, we are going to evaluate, through an optimization, whether 20 is actually the value with which the best results are obtained or whether it would be more profitable to use others.

Looking at figure 6, which shows the results of optimizations starting from 10 up to 50 with steps of 5, we notice that by increasing the calculation period of the Price Channel, the results in terms of net profit worsen on this basket of futures.

This is certainly due, at least in part, to a lower number of executions. Despite this, choosing one of the values ​​up to 40, the results are still satisfactory. The values ​​of 10 and 25 seem at this point to be a valid alternative but certainly not as much as 20, the value we used for the previous backtest, which shows the best result of all in terms of net profit.

Improved performance by adding an operational filter

At this point in the development, let’s try to insert an operational filter with the aim of increasing the average trade, since the value seen previously is not sufficient to be able to use this strategy in live trading.

The idea is to insert a condition so as to operate during those phases in which breakouts are potentially more efficient, avoiding false signals as much as possible. To do this we will use one of the most well-known patterns of all: the Daily Factor. This pattern will be considered on a daily time frame, therefore different from the 60-minute one, in order to have a broader indication for breakouts.

First of all, you need to calculate the body of the daily bar, that is, the absolute difference between the opening and closing of the previous day. Then, you will calculate the ratio between this last value and the range (the difference between the maximum and minimum, always of the daily bar). The necessary condition to open a trade will be to have a body less than 50% of the overall range of yesterday.

In other words, we will look for a phase of price indecision since, observing the behavior of the markets, following this we expect a phase of price expansion, ideal for a trend following strategy. Furthermore, this condition will be applied only to the opening of positions and not to their closing, since as the strategy is set up, the upper and lower levels are the only way to close a losing trade. Consequently, by also inserting the Daily Factor as a necessary condition for the exit, there would be the risk of being trapped in some trades that can lead to excessive losses.

Performance of the trend following strategy on energy futures with the addition of the operational filter

After inserting the Daily Factor as an operational filter we notice an excellent improvement in performance, starting from the equity line in Figure 7 which is more linear, especially in the final phase of the backtest.

As can be seen from the report in Figure 8, unfortunately the strategy has seen a slight decline in net profit to $930,000, due in part to fewer trades being made. However, on the other hand, the drawdown has been reduced by more than 30% compared to the previous strategy, from $96,000 to $66,000.

Instead, taking a look at the average trade (Figure 9) that is obtained after inserting the operational filter, we notice a value of about 111 dollars, which is higher than the average trade of the unfiltered strategy. This increase indicates that we are on the right track and above all that a condition of price indecision can bring improvements in trend following strategies. However, we emphasize that this is not yet a sufficient value to use the strategy in live trading.

Final Thoughts on Using the Price Channel in Trend Following Energy Futures Trading

At this point there are different ways to improve the strategy. For example, you could decide to insert a stop loss for better risk management. In fact, although the Price Channel levels are also used as exit levels, their value could be very far from the opening price of the position, leading to excessive losses, which could be better controlled through a stop loss.

In any case, the Price Channel confirms itself as a useful tool for setting up automatic trading strategies. Of course, the average trade is still at the limit to consider this strategy complete, but with the right precautions we have seen that it is possible, on certain markets, to make a more than satisfactory use of the indicator.

Want help improving your trading approach? Click here >>>

Happy trading,

Andrea Unger