In "The VIX Grew Up, Have You?" (thinkMoney, Winter 2012), author Thomas Preston observes that even savvy traders might sometimes ask the wrong questions (12). That observation prompted me to think about times we traders might be asking the wrong questions.
Preston's observation related to the questions traders ask themselves about the VIX. They observe the VIX's level and ask what that level suggests about the next market move. Preston suggests they should instead ask which strategy works best with the VIX at that level.
My thoughts went elsewhere. Are those asking where the markets are going to go always asking the wrong question, even leaving aside the value of the VIX?
I'll give you an example from Friday, February 17, 2012. Because the charts I'm including were snapped on that day, they'll give you an opportunity to assess whether I was asking the right questions, too.
On mid-afternoon on Friday, February 17, 2012, the OEX had pushed above the May, 2011 high. It had been climbing the 9-ema--the thin red moving average near the middle of the chart below--on the daily chart since mid-December. Daily RSI was beginning to turn down. That day was an expiration Friday ahead of a long weekend when the Greek debt crisis was due to be addressed again on the following Monday, when our markets would be closed.
What if I had been thinking of opening a trade that day? I wasn't and I didn't, but I wanted to use this as an exercise to show the kinds of questions I might be asking. I looked at a 30-minute chart to decide where I thought the OEX might be heading over the next few days. Without assigning any probability to next targets, I've marked possibilities with ovals.
OEX Daily Chart:
What if I were an iron condor trader, thinking about establishing a traditional iron condor? When I trade iron condors, I tend to sell the first call strike with a delta under 0.10 or 10, depending on the quote source. I tend to sell the first put strike with an absolute value of the delta under 0.09 or 9, depending on the quote source. However, I want at least $1.10 credit for my iron condors. On that particular day, I would have had to sell a put strike with a delta of -0.10 to collect at least that much premium if I was going to put the longs 10 points away. This is what happens when the volatility indices are low: we have to get closer in and take in less credit while the risk remains as high. When I'm trading iron condors, especially in a low volatility environment, I always spend about 10-15 percent of the credit I take in on an out-of-the-money put, so I am including such a put in this theoretical trade.
Because of the low-volatility environment, I would have taken in less credit and have closer-in strikes than I would in a higher-volatility environment. I would have had some questions to ask myself about what was going to happen the next Tuesday, as well as the rest of the time the trade would be open, wouldn't I? For example, what would happen if the OEX were to head to the upper tan oval at about 618-619 or drop to the lower blue one near 606-606.70 by a few minutes after the open on Tuesday?
Theoretical calculations don't always hold true in actual trading. However, they at least let us get an idea of what might happen. Let's look at the Analysis page on TOS with the date rolled forward to that Tuesday's.
Analysis Page with Date Rolled Forward:
Checking the white "today" line for Tuesday, February 21 shows me that at the upper tan oval's level, the position would have theoretically been down about $43.00. At the blue oval's level, the position would be about flatline.
That doesn't address the real situation, however. That Friday, I had to ask myself how volatility would change, too, in each of those situations. If the OEX started dropping quickly, volatilities would be likely to crank up. Since they were relatively low already and since they don't tend to drop on a hard rally at the open, they might not drop and might even climb even if the OEX should rally. Let's crank up the implied volatilities a bit and see how that impacts the trade. If the VIX/VXO/RVX were high instead of low, I would not crank up the vols to check the upside risk but only to check the downside risk.
What were the results? The chart isn't visually much different than the one above with some small differences I can describe, so I'm not going to include it. Cranking vols up 2.00 percent would sink the today line. At the upper target, the loss would now be about $247.00 and, at the lower, about $132.00. That wouldn't be fun to experience but it's not a devastating change.
The trade could have withstood a moderate pop or drop on that Tuesday morning, so we've answered the first question, but there are others that we should be asking. At what point would I need to make an adjustment? Does it look as if the OEX could quickly move to those levels? I haven't traded an iron condor for a couple of months, but I usually adjusted when my unrealized loss was about half the preset maximum loss I would allow on this trade. For me, that adjustment was typically at a loss of between 5-6 percent. For this trade, with this contract size, with commissions, that loss would be between about $227-272.
From our previous calculations, then, we can see that a strong upside move to the upper tan ovals, with volatilities that didn't retreat and maybe bumped up again, would perhaps have brought me to an adjustment point on Tuesday morning, at least until the volatilities settled down. Obviously, I didn't know for sure that Friday whether the upper tan oval would be reached, but I had certainly identified that as a possible short-term target along with some downside possible short-term targets. Would I have wanted to enter a trade that Friday that might need to be adjusted the first thing the next trading day?
My other adjustment decision point when trading iron condors often depended on the delta of the sold strike. When I list only the sold 600 call on a TOS Analyze chart, reset the date to 2/21, and crank up the vols, that delta escalates only to a tame 12.54. What does this tell me? Since this adjustment criteria had not been fulfilled and since I know that the early morning bump in volatility sometimes subsides on rally days, I might have been able to wait out an early jump up to the level of that tan oval long enough to see what happened next. Some possibilities included that the highs could be sustained but the vols would retreat or that prices would fall back, in which case the position might not need an adjustment. If it turned out to be a strong climbing day, however, the delta would reach an adjustment point of mine (delta of 16.00) by the time the OEX's value crossed 624. We've seen some days like that.
Maybe I might have wanted to wait to see what the weekend's developments in Greece were before I decided to enter that trade? That would depend on my risk tolerance and my belief about what would happen. Asking these questions and running the possibilities through these exercises would have allowed me to make a more informed decision.
I could have thrown a butterfly up and determined if it was a better trade for my outlook, testing what I hoped would happen and what I feared would happen. I could have done the same for a calendar or other trade, comparing what would happen at key points, determining whether one trade was more likely to get into trouble quickly than the other. And, yes, Preston would be glad to know that I would be figuring in my view of what was happening with the volatility indices as I made this decision. That's why I was not rolling volatilities lower to test upside adjustment levels. Although the volatilities can obviously go lower--this last week's action certainly proved that possibility--probabilities suggested that the were not likely to go too much lower. In conditions like these, we need to assess our risks if they rise.
It's not as easy to test trades in this manner on some other platforms, but many do allow for some testing. OptionsXpress or BrokersXpress do allow you to put a more rudimentary position analysis graph up and change the date and volatilities. The platform has an options pricer that would allow you to test how the delta of the sold strike would change under several different parameters, such as changes in date or volatility. These are the only two platforms which I've used since about 2004, so I'm not certain of how others work, but such freeware as Options Oracle also allows you to do some rudimentary testing of this sort.