Mid-morning on Wednesday, June 05, 2013, I wrote some of my trading buddies that I was SOH (sitting on hands), but that I was finding it hard not to adjust early. The chart setup showed that it was possible that the RUT would drop to 963-965 if it followed the pattern of the other indices. They had all dropped to a certain configuration of moving averages and Keltner channels that was analogous to the 963-965 region on the RUT.
SPX That Morning:
RUT That Morning:
The SPX had already dropped into the area represented by the top red rectangle. Would the RUT fall there, too?
My Trade That Morning:
The PnL wasn't accurate on this chart because I had input a volatility smile approximation--one of TOS's setup choices--rather than the standard TOS individual volatility option. Choosing that setup messes up the PnL, but, at some points in the cycle, it may more accurately reflect the deltas. At the time this chart was snapped, the loss was actually $1,575.45 or 3.94 percent of the max margin allotted to this trade. Price was jumping around a lot as implied volatilities moved so a moment later and the loss could have been a bit less or a bit more. It was well within my planned maximum loss for the trade, and the trade was well positioned with lots of theta beginning to build up.
If I was worried about that decline, why was I sitting on my hands and not adjusting? Wouldn't it be better to dump one of my call debit spreads before the RUT cratered?
My trading plan calls for me to adjust once a day until the week before expiration, when gamma becomes more negative and it may be necessary to adjust more often. Gamma tells us how fast our deltas will change. When they're negative, that tells us that our deltas are going to get more positive when the underlying is dropping and more negative when the underlying is climbing. Until that last week before expiration, I typically adjust sometime in the last ninety minutes of trading.
I adjust to keep deltas relatively flat, at about +/- 5 deltas per butterfly. (Note: I'm revising this part of my trading plan, doing back tests to determine if another delta level would be better for end-of-the-day adjustments.) This trade began with 10 butterfly contracts, so that would mean that at that time, I would have wanted to end the day with the deltas somewhere between +50 and -50. The previous day, the trade had ended up at + 20 deltas, for example. The trade had been within my trade parameters as trading opened on June 5.
Adjusting early that day wasn't within my plan. Moreover, although I was worried that the RUT might drop to that level, I wasn't certain that it would. My calculations showed that the RUT, at the day's low up until then, had been skating near a standard deviation but not much beyond it. (Those unfamiliar with standard deviations can consult the May 24 Options 101 article, "What's a Standard Deviation?") If we're going to use a standard deviation move either direction from the previous day's close as a measure of what's just noise and what might be something more, the move that morning of June 5 was just expected noise.
Also, as the RUT dropped, it was approaching the 970.63 low from May 23, where possible support might lie, with further potentially strong support near the 963-965 region. Watching the RUT drop so hard and not adjusting was almost painful. Usually in such times, I set an alert to tell me when price has dropped more than I want to see it drop. Then I busy myself with some other task, phone at hand so that the alert can tell me to come back to the computer. That lessens the temptation to adjust early.
Would I ever adjust early? Of course I would if we had a 2008-type cascade lower or a buying frenzy that brought my deltas way out of whack. And I do adjust more frequently beginning about ten days to expiration, as I mentioned previously. However, even in those cases, I would make each adjustment as conservative as possible if I'm adjusting intraday.
Also, I admit that sometimes I can't tolerate that SOH posture all day but that tends to have more to do with what's going on outside of trading--whether I'm ill, whether some other stress has appeared in my life--than with the actual trade. Those cases are few and far between. I almost always do fairly well at following my plan because my backtesting told me that it was a workable method. I wanted a trade I didn't have to tend every moment. Most days, I'm glad I adhered to the plan, as it's really how price and volatility end the day that matters, not how they traveled during the day.
My point was not to suggest that everyone should adjust only once a day, for each type of the infinite variety of trades. Not all trades can tolerate a once-a-day adjustment. I'm thinking of weeklies in particular. Rather, my point is that each trader should have a plan for that trade, should then thoroughly test that plan, and then should do the best job possible of following that plan. Back testing or paper trading the plan long enough that there's confidence that it works helps avoid seat-of-the-pants decisions during a tense moment.
Of course there are going to be times when you need to react differently. It's impossible to list all the factors that should lead to an outside-the-plan adjustment or action. As antsy as I was that morning, none of those conditions were occurring. I went back to SOH.