Protecting your trades with volatility stops
Once you decide what to compute for volatility, you have a to set trailing stops using a technical indicator while at the same time providing a method for following a trend.
It has been said that 80% of traders are on the right side of a trade when they enter. Then why is it that 90% of those traders end up in a losing trade? Could it be that they simply don’t know where to place their stops?
The idea of exiting a winning trade in a timely manner is basic to trading success. After all, if you can’t take the money off the table while it is there, you cannot succeed at trading. Exiting a trade on time and with a profit is a fundamental concept traders need to learn.
The volatility stop provides a technically based method of stop placement. Volatility gives you prices from which you can technically place and trail protective stops. Volatility, used in this way, can show you where prices are reversing trend. volatility stop is not divorced from correlation with other indicators that measure momentum, so it is not to be used for confirmation in association with any form of momentum oscillator.
But first you need to discover what volatility is and what the volatility stop does so you can use it in your trading.
WHAT IS ‘VOLATILITY’?
Apart from natural support and resistance points, moving averages and momentum indicators, the market has only one other factor to aid in stop placement: Price volatility. The distance between a price bar’s high and low is its trading range for the period of time represented by that price bar. The arithmetical difference between a price bar’s high and low is volatility. You can use the difference between the high and low for any time frame, from one year to one month all the way down to one minute.
Once you have the volatility for one interval of time, you can do the same calculation for any number of sequential periods of the same time interval. If you want to look at the total volatility over a five-bar period, you can take the difference between the lowest low of the period and the highest high of the period.
However, you typically look at a sequential series of like time intervals by computing the average volatility over that series. To calculate for a five-bar series, you might subtract the low from the high of each bar and after five bars total the volatility for all five individual bars, then divide by five to get the five-bar average volatility. Unfortunately, it’s not always that easy.
For example, if prices were to open with a gap up and the low for the current price bar ends up being higher than the previous bar’s close, how do you account for the distance that prices actually moved (see “Accounting for gaps” right)? Would the true measure of volatility include the distance from the previous bar’s close to the current bar’s high?
Is this the true measure of volatility? Prices did move from the close of the previous bar to the high of the following bar. Or is there another way to look at it? What if you measure volatility from close to close instead of from close to high (or close to low)? Is it any more right or wrong?
A third way is to measure from the low price of one period to the high of another period.
MAKING A CHOICE
As you can see, much of how to determine true volatility is in the eye of the beholder and also depends on the use to which volatility calculations will be put. For example, the method used needs to be much more exacting in trading options than for computing the volatility stop for futures trading.
For our purposes in using the volatility stop study, we will compute the distance from each day’s low to each day’s high and average them for the last five price bars as long as the bars remain normal relative to one another. If there are any abnormally large or small bars or a large gap, we will average volatility for the last 10 price bars.
By using more bars for unusual price action, we try to compensate for the fact that the volatility stop study formulation does not enable us to change parameters in the way it computes volatility. We would prefer to include gaps but have no way of indicating them to the study.
When prices expand, increased volatility comes into the market. When prices contract, so does volatility. The supposition behind the volatile stop study is that, if a market is very volatile, a stop based upon that same volatility will give a good indication of where to place a protective stop. Keep in mind that a protective stop may be protecting a loss of profits as well as an outright loss of margin.
To compute volatility for any number of price bars (N), take the sum of the differences between the high and the low for each bar for N days and divide by N. That gives you the average volatility for the last N days.
Let’s use T-bonds as an example. If you can figure average volatility for the bonds, you can do it for anything. When you compute anything to do with bonds, you have to convert them to decimal values first. To accomplish the conversion, multiply the whole number by 32 and then add the 32nds (see “Conversion to 32nds” page 45). The average volatility for the period shown is 23/32.
Then subtract the average volatility figure from your entry price if you are long or add it to your entry price if you are short. If you were to go long bonds at 103-14, you would subtract 23/32 from your entry price and place the protective stop at 102-23. If you were to go short bonds at 103-14, you would add 23/32 to your entry price and place the protective stop at 104-05.
The volatility stop provides a way to set stops using a technical indicator while at the same time providing a method for following a trend. The parameters in the study involve the number of bars in the average range and a constant value multiplier, typically between 2.8 and 3.1.
USING THE STUDY
The volatility stop simply computes average volatility for any number (N) of days you choose, multiplies it by the factor you choose and then adds the resulting figure to the lowest close in the last N days and subtracts it from the highest close in the last N days.
The process of adding and subtracting will yield two prices, one for upper volatility and one for lower volatility. Typically, one of the figures will be for a price that is inside (within) the range of prices you see on your chart for the last N days. The remaining figure will be either above or below the range of prices you see on the chart. The study will plot only the line that is furthest away from the range of prices. We use the plot line for the stop.
“Trailing line” (left) shows a plot line with the volatility stop study set at 5,1 – five bars with a multiplier of 1. Note that you can trail stops below the plot line in an uptrend and above the line in a downtrend. In case you are wondering, setting the multiplier at 1 nullifies any effects from the multiplier. Some software allows you to offset the volatility stop study by one bar so that, at the close of the current bar, you can see where to set the stop for the next bar. If you have that capability in your software, use it.
With a seven-bar moving average and a multiplier of 2, you could have trailed the uptrend (stop, close only) to the end of the trend (see “Hanging in there,” left). A fine feature of the volatility stop study is the ability to curve-fit it to the angle of ascent or descent of a trend. An example of curve-fitting is seeing prices pull away from the plot line and then changing the parameters.
“Multiplier effect” (right) shows what happens when you change the multiplier from 2 to 1.2 If your software allows an offset moving average, it will look like this chart. The volatility line will protrude one bar forward. Notice that with the offset, you can set your stop for tomorrow ahead of time.
(Note: When the volatility stop plot line(s) is running through the price range with prices moving sideways, you do not enter any new trades.)
The volatility stop in “Staying with the trend” (right) is set to 5, 1.4. By the time the market reaches the point of the lower arrow, you have price containment by the volatility stop line. Containment means that the line “contains” – in this case, is below – all of the price action. Trail a stop one tick below the volatility stop line as prices move up. You can trail with a close-only stop or a stop at the volatility stop value of the previous price bar.
Because the volatility stop setting worked so well going up, you can try it going down. Is it wrong to curve-fit the volatility stop in this way? Not at all! We are basing what we do on what prices themselves are doing!
The volatility stop is an excellent tool for staying with the short- to medium-term trend. The ability to curve-fit it to the price action makes it very useful. However, as you review this study, you must realize that the volatility stop is not truly adequate for keeping you in a trend for the long term unless you switch to a longer-term chart along the way.
Joe Ross, a long-time trader, educator and author, teaches spread trading and day-trading techniques for futures, forex and stocks. Reach him at www.tradingeducators.com
Copyright Futures Magazine Group Jan 2004
Provided by ProQuest Information and Learning Company. All rights Reserved