Last week's article discussed setting stops. A subscriber prompted the article after asking whether a stop should be set based on the price of the underlying or the option itself. Sometimes, it's both, the article concluded.

In directional trades, the stop should often be based on the price of the underlying, but no stop should result in a loss too big for the account size. Since weird things happen to options prices due to volatility changes and the passage of time, it's not enough to keep an eye on the price of the underlying. While the premise for the trade might still be valid, that option could have decayed until the loss is too big.

What is "too big"? Some experts advise 2-5 percent of the account's size, but I think it also depends on the type of trade and the amount of profit that can be expected from that trade. If you typically garner profits at 10 percent of the buying-power effect of the trade, then you probably ought not to set a maximum loss at 40 percent of the buying-power effect of the trade, even if that 40 percent represents only 1 percent of your total account value. If you have such unbalanced gains and losses, you'll run through your account. Therefore, if the most logical stop would result in a loss that's too big for the account size or too big in comparison to the typical gains you can expect for this type of trade, then the trade shouldn't be entered, the last article concluded.

Last week's article mentioned using the Average True Range or ATR to determine what the average price movement is during a certain period of time. The stop should logically be set outside that normal range so that the typical noise of a certain period of time won't result in the trade being stopped. Others might use Bollinger bands or Keltner channels to determine normal price movements over intraday or longer periods, appropriate to the trade. A trader who intends to scalp a few points, getting in and out quickly, might use a five-, ten- or fifteen-minute movement, for example, while a swing trader might use a 60-minute one.

What happens when the underlying's price has moved significantly away from the entry and the option trade has gathered profit?

Annotated 15-Minute Chart of the SPX on 7/23:

If appropriate, stops can trail the moving average or trendline that was the basis for setting the trade. Some of the more advanced platforms now allow stops that follow moving averages or trendlines, but not all do and not all such stops are appropriate in all market conditions. Stops can also be set at previous swing highs or lows, depending on whether the trade is a bullish or bearish one. Trailing stops can be based on the option's value so that its profit is not allowed to drop back into a loss.

July 23 was the day captured in the chart shown above. That Friday was the day when the results of the European's stress test for banks was announced. On that day, a bullish trader might have entered based on a test of the 45-ema for this time period, a moving average that sometimes serves as support or resistance on closes for the SPX on intraday charts. A trader might have felt that the first push lower would likely find support again at that moving average and attempted a bullish trade. I don't know if the 1130 would have been the best option to choose because its delta was far lower than the 0.70 I typically used to choose for scalping or day trading, but I was tracking that option for another reason and so had price points I could display.

As the trade progressed, that option would have been garnering profit. Following that moving average higher, with the stop just below it--but not at a nice round number where it might get picked off--might have been a good tactic, but what would happen as the end of the day approached? Again, this decision goes back to the premise of the last article on stops: what was the trader's intention when the trade was entered? Was this a day trade or was it meant to be a longer-term trade? If it was a day trade, then I would caution that trader not to turn it into a several-day trade. Stick to the original intention of the trade.

As the SPX still tested resistance (a Keltner channel I watch) as the end of the day approached, a stop way down slightly below the rising 45-ema no longer made sense if this was intended to be a day trade closed out before the end of the day. The trader needed a new way of setting a stop. What I used to do in cases like this was to change my stop to an option-price-based one. I would keep tightening my stop ever closer to the current option price, the closer the end of the day approached. The idea would be that you give the option's price less and less room to decline as the end of the day approaches. The option gets taken out near its high price that way. If the underlying's price doesn't retrace but zooms in the profitable direction into the close, then your stop should be a time-based one you set for yourself. I typically liked to close with at least 15 minutes left in the day. Sometimes odd things happen to option prices in the last few minutes as market makers mark the options down or up, depending on the volatility, their inventory, and other concerns. I can guarantee that you're going to find times when you close a day trade fifteen minutes before the close, only to watch the price of that option keep escalating, leaving you singing the coulda-woulda-shoulda song. I can also guarantee that if you don't make a practice of closing at or about that time period, you'll occasionally find yourself chasing prices as the underlying reverses in the last few minutes of trading. There is no absolute right or wrong, so you can't guarantee ever that the decision you make about the timing of closing a day trade will be the right one every time.

What if your trade wasn't meant to be a day trade but a swing or position trade, one you stay in for days or maybe even a couple of weeks? In this case, you don't want to tighten too much if you intend to stay in the trade for several more days at least. You may want to continue to use the rising 45-ema as the basis for your stop, with the caveat that at some point, the underlying's price will have to retreat to the bottom of the black channel and not just bounce from the 45-ema.

That doesn't exempt the swing or position directional trader from making end-of-day decisions. Some of the precepts discussed here are ones that former market maker Dan Sheridan discusses in some of his webinars. Notice how the SPX is pressed up against resistance that was holding? How much is the underlying likely to move the next day? Many charting systems will show you a standard deviation, so that you can see what the underlying is likely to move in either direction about 68.2 percent of the time. What will happen to this bullish position if, the coming Monday, the SPX moves down by that much instead of up by that much?

One charting service calculated that a standard deviation in either direction the following Monday could see the SPX ranging from about 1191.60-1115. What's the delta of your whole position? At the end of the day, that option had a delta of about 0.3280 or 32.80 if your quoting source applies the 100 multiplier before giving you the delta. What if the whole position consisted of 4 of those 1130 calls? The purchase of those calls that morning would have cost $3,480 plus commissions, with the $8.70 original price quoted on the chart. By the end of the day, taking the more conservative $11.20 price, those options would have been worth $4,480. That's a $1,000 profit or a little less than 29 percent on the original investment, excluding commissions.

Even if the trade had been intended to be a swing or position trade enduring for several days or weeks, what was the trader thinking when the trade was initiated? If the original goal was a 25 percent profit on the buying-power effect, it's there for the taking. If it's a 200 percent profit, well, the trader may be asking too much, but it's obviously not yet available. In each case, stops--both profit stops and stop losses--must always hark back to what the trader was thinking when the trade was initiated, always keeping an eye on the option price, too.

Let's suppose that the profit stop hasn't been reached and the trader had high-flying goals for the eventual profit to be collected. That trader still must keep an eye on what might happen to the option's price. We've had gappy openings lately, even on the major indices, perhaps especially on the major indices. What would happen to that profit if the SPX gapped lower to the 45-ema down at about 1095.08 on Monday morning, a 7.58-point drop? If the delta of a single option was 32.80, then the position delta for four was 131.20, which would mean that the whole position would lose more or less $131.20 for each point the SPX dropped, if it had gapped lower that next Monday. (It didn't, of course). If it had gapped down to that 45-ema, the position loss would have been more or less 131.20 x $7.58 or about $994.50. Some time decay would have occurred over the weekend, too, but not as much as models show or traders expect. Especially with an afternoon like that on July 23, with prices staying rather steady, market makers would have been lowering volatilities on those options, effectively taking out the weekend decay.

The premise of the first article was that even when trades are directional and have been prompted by a chart formation, moving average, trendline or prior support or resistance, the price of the option must still be considered. Therefore, although the original premise for this entry--a test of and bounce from the 45-ema--still proves valid at the end of the day, the option trader must consider what might happen Monday morning. Is that trader okay with the possibility that Monday's opening could gap prices down to the 45-ema, and take away $994.50 of the accumulated profit? If not, the trader has to institute a new type of stop. How much profit is the trader willing to risk? That's an individual decision, but I, for one, wouldn't want to risk that percentage of profit evaporating at Monday's opening.

One possibility is to collect part of the profit even if the intention is to keep following that bouncing 45-ema higher. Selling two to three of the profitable options would lock in $2,240 or $3,360, with the selling of three almost paying for the entire trade, so that very little was at risk. This would be reflected in the position delta, too. Selling three of those options would lower the position delta to 32.80, meaning that a Monday morning gap of 7.58 points would erase only $248.62 of the profit left in the remaining option. The trader could more easily weather that test of the 45-ema, if it were to come.

What about the unhappy case when the trade is not profitable as the end of the day approaches but the appropriate stop has not been hit, either? Again, what was the trader thinking when the trade was initiated? If this was to be a day trade, then the stop switches to the end of the day, no matter whether the original stop has been hit or not. Never, ever turn a day trade into a swing or position trade just because the trade isn't yet profitable. If the trade was meant as a swing or position trade, then the trader must go through the same exercise just tested for the case of a profitable trade. Does the position delta tell you that a gap down would bring the trade past the maximum planned loss? Then a tough decision needs to be made. When the next trading day's open could violate the planned maximum loss, then you should probably treat the trade as if the maximum loss has been reached, because it could be before you can do anything to react. Consider closing part of the trade, to lower the price-related risk, or all of the trade, if that won't bring the risk within acceptable parameters. As I mentioned with the case of taking profits at the end of the day, I can guarantee that sometimes you're going to look at a case like this and elect to go ahead and take your loss, and the next trading day's open will see prices head straight in the direction you needed them to head, and you'll sing that "coulda-woulda-shoulda" song. I can also guarantee that if you don't make these end-of-day decisions, you're going to experience some losses well above your planned maximum loss, and that's even less fun. It's more damaging to your confidence in yourself, too, and your trading abilities.

When the subscriber asked me the question about whether stops should be based on the price of the underlying or the option, I answered that the answer would be complicated, and it is. These first two articles have covered only directional trades, with the premise being that the stop must be predicated on the original reason for entering the trade. In addition, an eye must always kept on those option prices, made slippery by changes in volatility and the passage of time. A stop that's following a moving average, previous swing highs or lows or trendlines might need to be changed as the end of the trading day and especially the end of the week approaches. At that time, the trader must hark back to the original trade premise to decide whether to cinch up stops as the end of the day approaches or just close the position before the end-of-day rush hits. If it's intended to be a longer-term trade, then the trader must calculate how much harm could result from normal price movements at the open the next day, and revisit decisions about taking profit or loss at that time. The profit stop or stop loss that was appropriate at the middle of the day may not be appropriate as the close approaches.