A subscriber's question about stops prompted an ongoing series of articles. This is the third in that series. The subscriber posed the question of whether stops should be based on the price of the underlying or the price of the option. Great question.

My viewpoint on stops is that options traders must first determine the original premise for our options trade before we can determine where the stop should be placed. The prior two articles delved into stops on directional options trades. Such trades are often based on the underlying's movement in relationship to a formation, trendline, moving average, or prior support or resistance in the form of swing highs or lows. Directional traders are looking for support or resistance holds or is violated. Some directional traders want to place the bounce from support or down from resistance, or some might want to play a breakout below such support or resistance, depending on their views of the market action. The same goes for trendline and moving average tests.

Whatever the premise of the original trade, a stop should be set above or below the original decision point. The stops should be an appropriate distance away so that normal market noise doesn't hit the stop. Traders should avoid setting those stops at nice round numbers, easy to pick off.

However, options prices are slippery things impacted by changing volatilities and passing time. Therefore, while stops on such directional trades should generally be based on the underlying's price action, the option's price is what determines whether the trade is profitable or losses money. Therefore, an eye must always be kept on those slippery option prices. No matter whether the underlying is moving in the wrong direction or just too slowly or with the wrong volatility in the right direction, the trader should never allow the loss to exceed a maximum preset loss that the trader set before the trade began. With options, we can be right about what happens to the underlying but if we're wrong about either the timing or the volatility, we can still run through our accounts if we're not attentive to the option prices.

If that the most appropriate stop is so far away that the maximum preset loss would be exceeded before the underlying approaches that stop level, the trade shouldn't be entered. You can't give it enough room to work in that instance. Even if you're pretty sure what's going to happen to the underlying, it's not a good practice to enter such trades. Great setups sometimes occur that have underfortunately suffer from this difficulty. Another trade will come along, but your "risk of ruin," a term we traders borrow from gamblers, rises too high if you let losses grow too large just while waiting for the stop to be tested.

The second article explored basic ideas for stops when the trade has garnered some profit. If that trade is based on a rising trendline's support, the underlying's price might have bounced way above the original entry. The option trader needs to move the stop. It's easy enough, at least at some brokerages, to then follow that moving average or trendline higher with the stop.

In cases such as these, the trader should avoid round-number stops or stops right on that moving average or trendline. If you're watching that formation, moving average or trendline, you can bet others are, too. While some option traders think in terms of the evil market makers gleefully picking off those stops before reversing the underlying again, I think differently. If I were someone trading options in the thousands or tens of thousands, rather than in multiples of ten, I would want to know whether that support or resistance is going to hold. I would need to decide whether to add more to my position on the next test or, instead, trim down my position and lock in some or all of the profit. If I had enough money to drive the markets one direction or another during low-volume times such as lunchtime, I'd run that test. I'd enter a big enough position to drive the underlying down through support or up past resistance, depending on the trade. Then I'd sit back and see if it's bought again or sold again, depending on the position. Then I'd know what to do with my trade.

Don't get in the way of such tests. How far under support or above resistance should you place the stop? That's an impossible question to answer because it of course depends on the underlying. To avoid normal market noise and such tests, you probably wouldn't need to get as far away with a low-beta stock such as IBM as you would with some herky-jerky small-cap pharmaceutical. Know your underlying or at least study it before you make such decisions.

In this and other cases, perhaps such as a directional trade based on a breakout above a previous swing high, traders might have a bit more difficulty determining whether a stop should be once the directional trade is profitable. That's where Average True Range, ATR, can be helpful to some traders. I mentioned in an earlier article that traders can use ATR to help them determine how much market noise is normal for the chart period being watched for that underlying. ATR can also be used to trail stops behind a profitable directional trade. In a February 3, 2007 Trader's Corner article, I wrote about "Chandelier Exits," a topic covered by Sharon Yamanaka in the November, 2006 issue of Stocks and Commodities. Yamanaka suggested setting stops for bullish trades by subtracting a multiple of the ATR from the highest high or highest close during the period being considered. "In effect," I wrote in that Trader's Corner article, "the exit hangs from the highest high or highest close, giving the exit its name." For bearish trades, the exit or stop is suspended above the lowest low or lowest close.

When Yamanaka adopted this plan, basing it on a 1999 book by Dr. Van K. Tharp, she had worried about setting stops too close, so that normal noise I talked about earlier would take out the trade. But what's the "normal" amount of movement in the period being considered? As those of us who have traded for many years know, that can vary widely. I have seen times when I've dialed down to 1- or 5-minute charts just to see any movement at all on the SPX, and others when we're seeing wild surges higher or lower in just a few minutes' time. It's not that difficult to find five-to-ten minute moves in the first five minutes of trading these days.

ATR expands or contracts with the changes in volatility, so it helps traders hone in on what constitutes normal movement for the time period being considered. At least in that distant-in-time article, Yamanaka believed using ATR helps keep traders from being whipsawed. It also helps traders avoid placing their stops at those nice round numbers, where the majority of stops are typically placed, making it easy for a stop run to take them out.

This is not a strategy that I've employed in my own trading, mainly because by the time I read the article, I had already gravitated away from directional trading. Because I haven't used this method, I can't guarantee how useful it's proven over the last few years, but the premise remains the same: normal market noise should grow in more volatile times, and the ATR should reflect that, at least in theory. I don't want to reprise my entire article here, so check it out in our archives if you're interested. You'll find that Yamanaka combined the ATR guidelines she set up with her knowledge of technical analysis, too, to sometimes give the trade a little more leeway than it would have otherwise had.

The takeaway from this third article on stops, as with the second one, is that once an option trade is profitable, stops must be adjusted. Trailing the stop proves advisable, with several trailing methods available to the trader, depending on the trader's brokerage.

Whatever trailing method is employed, the trader must still always keep an eye on the profit or loss of the trade. The underlying might be moving in the right direction, but, particularly in the case of a long call position, the position might not be profiting or profiting as much as expected. Declining volatility as the underlying climbs might be decaying the option's price, particularly if it was a plumped-up expensive option when purchased. In all the cases discussed in this article and the last one, setting a stop based on the underlying's actions is appropriate, but only if the trader focuses on the option's value at all times, too. That's where we'll take home our paychecks or suffer a loss, no matter where the underlying is going.

Still to be discussed are stops on other types of options trades, such as complex positions designed to be relatively delta neutral, at least at inception.