The Average True Range (ATR) is an indispensable tool for excellent trading system designers and can be considered a true dark horse among technical indicators. Every system trader should be familiar with the ATR and its many useful functions. Its numerous applications include parameter setting, market-entry, stop-loss, profit-taking, and even as a very valuable auxiliary tool in money management.
I. Calculating the Average True Range (ATR)
The range is the distance between the highest and lowest points of a single candlestick chart. The true range is the maximum of the following three ranges: the distance between the highest and lowest points of the day, the distance between the previous day’s closing price and the current day’s highest price, or the distance between the previous day’s closing price and the current day’s lowest price.
When a gap appears in the daily candlestick chart, the true range and the range of a single candlestick are different. The Average True Range is the average of the true ranges. To reflect recent volatility, a short-term ATR (2-10 candlesticks) can be used; to reflect “long-term” volatility, 20-50 candlesticks or more can be used.
II. Characteristics and Advantages of ATR
ATR is a universal indicator that evaluates market price movements and is truly adaptive. The following example helps explain the importance of these characteristics. If we calculate the average price movement of corn over two days, such as $500, the average price movement of the yen contract might be $2000 or more.
If we want to establish a trading system that sets appropriate stop-loss levels for corn or yen, we will see that the stop-loss levels for the two are different due to their different volatilities. We might set a stop-loss level of $750 for corn and $3000 for the yen contract.
If we want to establish a trading system that applies to both markets, it would be difficult to equalize the stop-loss levels expressed in dollar amounts for both markets. A $750 stop-loss level is appropriate for corn but may be too small for yen; a $3000 stop-loss level is appropriate for yen but too large for corn.
However, let’s assume in the above example that the ATR of corn over two days is $500, and the ATR of yen over two days is $2000. If we set the stop-loss level at 1.5 times the ATR (i.e., the stop-loss level expressed in ATR), we can use the same standard (i.e., 1.5 times the ATR) in both markets, with the stop-loss level for corn being $750 and for yen being $3000.
Now let’s assume market conditions have changed, with corn volatility becoming very high, moving $1000 within two days; while the yen becomes very calm, moving only $1000 within two days. If we still use the previous stop-loss levels expressed in dollar amounts, i.e., the stop-loss level for corn remains at $750 and for yen at $3000, then the stop-loss level for corn is now set too close, and the stop-loss level for yen is set too far.
However, a stop-loss level expressed as a multiple of ATR can adapt to market changes, and a 1.5 times ATR stop-loss level will automatically adjust to $1500 for corn and yen. A stop-loss level expressed in ATR can automatically adapt to market changes without changing the original stop-loss standard; the new stop-loss standard under new conditions is the same as the previous one, both being 1.5 times ATR.
The universality and adaptability of ATR as a market volatility indicator are undoubtedly valuable. ATR is very valuable for building solid trading systems (i.e., trading systems that may be equally effective in the future) and can be used unmodified in multiple markets. ATR can be applied to the corn market and can also be used in the yen market without any modification.
But perhaps more importantly, you can build a system that not only performs well in historical data tests for corn but is also likely to perform well in the future even if the corn market changes significantly.
III. ATR Trailing Stop Strategy
In the past few days, the average daily fluctuation of the euro was about 80-150 points, and GBP was about 100-200 points. Unlike the commonly used percentage, this value changes during different market periods. The basic idea is very simple: we first select a reasonable starting price, then add a certain multiple of AR every day to get a trailing stop point.
The stop points generated by this method can not only continuously move up with time but also adapt to the increase and decrease of market volatility. Compared to the stop points obtained from the parabolic SAR indicator we used before, the advantage of using the ATR Ratchet is that we can more freely choose the starting price and the rate of increase or decrease.
In addition, we found that stop points based on ATR can reflect changes in volatility faster and more accurately, allowing us to lock in more profits than traditional trailing stop methods.
For example, when we achieve a profit target of more than 1 times ATR, we choose a recent low point (such as the lowest price in the last ten days) as the starting price, then according to the number of days we hold the position, we increase the lowest price by a fraction of ATR every day (such as 0.05ATR).
If we have held the position for 15 days, then we multiply 0.05ATR by 15 days and then add the product 0.75ATR to the starting price. After 20 days, we will add 1.0ATR (0.05 times 20 days) to the lowest price of the last ten days.
Unlike the parabolic SAR indicator, the ATR Ratchet can be easily used at any time during our trading process. We can start using this stop-loss strategy on the first day of entering the trade, or we can wait for certain favorable events to occur before using the stop-win strategy.
I suggest waiting until profits are realized before using this stop-loss strategy, as you and I have seen, this type of stop point will move up quickly in a favorable market environment. Increased volatility will increase the speed at which the stop point moves up, which is an important feature of the ATR Ratchet strategy.
In a fast-moving market, you will see many gaps and long candlestick charts. When the market trend accelerates, market volatility will also increase, so when our profits increase rapidly, ATR will also increase rapidly.
Since we add a certain amount of ATR to the starting price, each increase in ATR will cause the stop point to jump up suddenly, making the stop point closer to the highest price after entry. If we have held the position for 40 days, any increase in ATR will have a 40-fold impact on the stop point.
This is exactly what we want. We find that when the market gives us generous profits, the ATR Ratchet stop point will also move up surprisingly quickly, thus locking in our floating profits well.
This method has the following parameters:
1. Starting Price
One very good feature of the ATR Ratchet is that we can set the starting price anywhere we like. For example, we can set the starting price at some important low points like the parabolic SAR indicator, or we can set it at the bottom of the swing range, at the support level, at the bottom of a channel, or a place below the entry point by a certain number of ATRs.
If we wait until we have a considerable amount of profit on paper, we can even set the starting price above the entry point. This allows us to coordinate with the trading system we are using.
2. Timing of Starting ATR Ratchet
Prefer to use parameters based on time rather than price (or a combination of time and price parameters) to enable the above exit strategy.
For example, we enable the exit only when a trade has been open for at least ten trading days and has profited by more than one ATR. In general, only after the trade has reached a considerable profit target is the best time to start the ATR Ratchet.
It seems to be a good profit-taking strategy, but be aware that if you start the Ratchet before a trade is profitable, you may exit too early and miss the opportunity. As mentioned above, the most attractive point of the ATR Ratchet is its applicability and flexibility. Here’s another idea on how to enable the Ratchet strategy.
We can enable the ATR Ratchet after 15 bar charts without having to calculate the previous 15 steps. When writing program code, we can set the Ratchet to start after the 15th bar chart of the trade and subtract 10 from the number of bar charts after the trade to multiply by the unit value of ATR, or divide the number of days after the trade by a certain constant and then multiply by the unit value of ATR.
This method will simplify the calculation program of the Ratchet, especially when the exit strategy is first enabled at the beginning of the trade. Think about the ATR Ratchet and see what kind of creative thinking you can generate from it.
3. Daily Movement of ATR Ratchet
The initial daily movement of the ATR Ratchet we started to study and use proved to be too large through testing. For our trading timeframe, a too-large daily movement of the ATR Ratchet (a few percent of AR) would cause our stop-loss points to move up excessively fast.
After a period of trial and error, we found that multiplying our holding days by the daily movement of the ATR Ratchet from 0.05 to 0.10 ATR (5% to 10% ATR over 20 days) would move the stop-loss points up much faster than one might expect.
As a flexible approach to this strategy, we can initially use a smaller daily movement of the ATR Ratchet, and then once we achieve significant floating profits, we can switch to using a larger daily movement of the ATR Ratchet.
4. Length of ATR Period
As we have discovered in our previous use of the ATR, the length of the period used to calculate the ATR is very important. If we want the ATR to quickly reflect changes in the market’s short-term volatility range, we can use a shorter-term average (such as 4-5 candlesticks).
If we want a smoother ATR that is not sensitive to one or two days of abnormal volatility, we can use a longer-term average (20-50 candlesticks). Most of the ATRs I use in my work is a 20-day average unless I have a good reason to want the ATR to be more or less sensitive.
ATR Ratchet as a Profit Tool: We especially appreciate the flexibility that the ATR Ratchet brings to us as a profit tool.