Good Stop, Bad Stop: Chandelier Exits

Just as it was tough when we were children to look under the bed or in a dark closet for night monsters, it's equally tough to look at a loss and acknowledge it. It was easier to hide under the covers back then and now it's easier to adopt some defense mechanism. (The one I hear most is "Oh, that trading rule doesn't work!" as if the entry strategy caused the loss.)

After having been swept away once too often by deflating stocks, I've found that the easiest way to chase the monsters off is to never have to face them in the first place. And that means setting my stops before I place the trade. Of course, that was the advice I'd been getting from Day 1, Lesson 1 of Trading Stocks 101. But it's a long journey (into nightmares) from visualizing a hypothetical stop should the unthinkable happen, to acquiring the discipline and experience to place a hard stop on every trade I place. To quote Sweeney one more time, "Three things can happen to you [when you place a stop] and only one of them is good."

That's because stops are tricky, and even though they're damage control, they can do more harm than good. I've been taken out of perfectly good trades by too tight a stop and ended up losing money on what should have been winning trades. Or taken out way too late from a losing position. Or hit a trailing stop that's 8% off from the high. If only...

With that in mind, I've taken a look at chandelier exits.

CHANDELIER EXITS
I was originally attracted to the chandelier exit because it was unique. I hadn't heard of it before and it wasn't a standard tool on any of the charting sections in the larger financial websites that I'd noticed. Everyone likes stocks that travel under the radar of the masses, so why not a stop that does the same?

FIGURE 1: HOW TO PLACE A CHANDELIER EXIT. The stop is usually placed at three times the ATR.

To calculate the chandelier exit, start with average true range. You can find this value at Prophet.net (among other places) under their "Analyze" tab (www.prophet.net/analyze/sc.jsp). For a stop, ATR is usually multiplied by three to give an exit point. But that can be adjusted for risk tolerance, time frame, and trading style. The entry dictates the exit. If you entered a trade based on a chart pattern such as an ascending triangle, then your exit would be based on either the completion of the pattern or a stop-loss, should the pattern break down. So from the high point, keep adding in the ATR until it falls within the correct stop-loss range for your trade.

The chandelier exit offers a lot of variations--hanging the stop from lows rather than highs, changing the multiple of ATR, or increasing or decreasing the number of days used in calculating the average. The chandelier stop is so flexible that some versions allow you to use another measurement for volatility other than ATR, such as the daily range or average range. In addition, the exit can be recalculated daily, adjusting for slight changes in the moving average.

In using this stop, you give yourself a unique exit point at a volatility-based distance from your entry, calculated to eliminate whipsaws, but not so low that you're exiting at everyone else's entry point or getting lost in the mass exodus at the breaking of a major support/resistance.

FIGURE 2: NWRE WITH A 10-DAY ATR. ATR's volatility peaks at the first bottom of the double bottom formation, reflecting the quick drop in the stock's price.

FIGURE 3: AUY, SIX-MONTH DAILY. On this chart the trendlines and triangle formations are marked.

Rather than waiting for AUY to hit the bottom of its channel (Figure 3), I got into this stock on the pullback at 10.30. A nice symmetrical triangle had been forming as the stock consolidated, and gold futures had been rising after a retracement. I didn't think the stock would reach the lower trendline, but before I placed the order, I wanted to figure out a stop-loss. For a $10 stock, every 10-cent change in price is a 1% loss. By way of comparison, some of the different stops are:

1. 8% generic stop: 9.50
2. Chart pattern stop placed slightly below the bottom of the channel: 9.2
3. Stop below the largest single-day range ($1.00) for the past month: 9.3
4. Stop below the 50-day moving average: 8.82
5. Chandelier exit: 9.77 to 9.30 (ATR: 0.55, highest high: 11.42, highest close: 10.91)

My objectives in buying the stock are for a short to intermediate time frame with a target of $12. AUY's trend channel is quite large, about 2.25, from the last low of 8 to the high of 11.5, which is more than 20% of the stock's value. So I'm looking to sell the stock at the top of the trend and wait for it to retreat and buy back in. By selling at the top, I'm also limiting my exposure come the inevitable fall in the price of gold. The stock could be shorted on the way down, but I'm not doing that right now.

I could just hold onto the stock until it falls out of its upward trend, but I also want to maximize my profits. So I'm placing my stop at the shortest distance from my entry point, at the chandelier exit point of 9.77, slightly below 10, a support area for the stock. If I get stopped out, it'll be for a small loss and I can get back in at the bottom of the trend channel at 9.3 and save myself some money. My risk is no more than 5% for, roughly, a 20% gain.

FIGURE 4: CLOSE-UP OF STOPS. The chandelier exit: a mighty tight stop.

This trade went my way and I hit my target of $12 and sold my position. After I sold AUY at the top of the channel in mid-May, I placed a limit order for at 8.50, well below the bottom of the channel, and got in a few days later (Figure 5). Even though the stock had fallen out of its channel, I'd been following gold and this particular stock for a while and wanted to get back in, expecting the stock would reenter its uptrend.

FIGURE 5: AUY, DAILY, ON JUNE 16, 2006. It is likely that a downtrend with the top of the channel at 9.8 is establishing. There is a lot of support/resistance between 8.5 and 9, as shown by the volume price bars on the left.

Once in, though, I found it harder to play the stock. The market indexes were uncertain, my stock was wobbling between chart patterns, and the futures price of gold that my stock reacted to like a high-strung thoroughbred at the starting gate may or may not have been topping or bottoming out. With all the uncertainty, I shouldn't have gotten back in, but 8.5 seemed like a very good price for AUY. In this position, a system like the chandelier exit gives me a discipline to follow when otherwise I would have no clear choice.

As it happened, the stock immediately went up and I placed my chandelier stop at 8.72, higher than my entry point. With an ATR of 0.9, I needed to use four times the ATR to get a reasonable placement from the high of 12.42. I was eventually stopped out when the stock moved lower and seemed to be establishing a downtrend (Figure 4). No harm done. I missed out on profits, but I didn't lose anything in the trade either.

The shaft of the arrow had been feathered with one of the eagle's own plumes. We often give our enemies the means of our own destruction.

While I wouldn't actually go out and prove this, I believe that if you place your stops correctly, you can trade anything. I've found that if I spend more time monitoring my open trades rather than looking for new stocks to buy, my profits increase proportionally. In terms of trading psychology, using chandelier exits gives me the discipline to thoroughly review my options in trading a stock. The chandelier exit isn't a trading system in itself; rather, it is a flexible system that adapts itself to a specific trading style. I use chandelier exits as an entry stop-loss for trading chart patterns. Plus, I find ATR useful in itself as a measure of volatility.

What's the only good stop to hit? Sometimes, a stock will blow past my target. Then I can tighten the trailing stop so that at the slightest correction, I'll be stopped out. I hate to sell a stock that's going up and normally I don't like being stopped out, but anything past that is icing on the cake. And who doesn't like the sweet stuff?

Sharon Yamanaka may be reached at SYamanaka@Traders.com.

SUGGESTED READING
Gopalakrishnan, Jayanthi [2006]. "Chuck Dukas Of Trendadvisor.com," interview, Technical Analysis of STOCKS & COMMODITIES, Volume 24: June.
Pezzutti, Paolo [2006]. "Understanding Market Structure," Technical Analysis of STOCKS & COMMODITIES, Volume 24: March.
Sweeney, John [1996]. Campaign Trading: Tactics And Strategies To Exploit The Markets, John Wiley & Sons.

FIGURE 6: SAMPLE BUY ORDER

Sidebar:
POSITION SIZING: HOW MUCH DO I BUY?
How much to buy is limited by how much you have in your account versus how much you're willing to lose. For shorter time frames, the rule of thumb is to risk no more than 2% of your total capital per trade. So for a $25,000 trading account, that would be $500. For AUY, I set my stop-loss at 9.77 for a stock I bought at 10.30, which, if hit, would be a 53-cent loss per share. If I don't want to lose any more than $500, that would be $500 divided by $0.53 = 943 shares. At $10.30 per share, that's $9,713 of my $25,000 capital, or 38%. For me, that's too much of my trading capital tied up in one trade that I expect to last a few weeks, so I'd buy fewer shares. This would also give me more leeway on the stop, should I choose to move it lower.--SY

Sidebar:
THINGS TO REMEMBER WHEN YOU'RE PLACING YOUR STOP
1. Market direction. Going the opposite direction of your trade? Tighten the stop.
2. Trade setup. Chart pattern? Long or swing trade? The setup, and therefore entry, determines the stop.
3. Time frame. The longer the time frame, the larger the stop.
4. Money management. Your reward should be at least twice your risk.
(Based on "Interview: Chuck Dukas of TRENDadvisor.com," June 2006, Technical Analysis of STOCKS & COMMODITIES)