Dom, for you specifically since you like to use the standard deviation probabilities in your decision process, if you have time I think you should check out the CBOE webinar for What the Bell Curve Doesn't Tell You ~ Russell Rhoads. Mark I don't know if you use that tool as well but you'll remember from our Stats class how this pertains to options trading. The short version is this, in a Monte Carlo simulation of 1,000,000 30-day random stock trades using their implied volatilities, the one sigma deviation was touched or violated 54% of the time during the 30-day window until expiration. The reason this is important is that the long-term bell curve probabilities held true after 1,000,000 simulations. This means that only 32% of the time did the stocks finish outside of one sigma. So the implication here is that if you've got a trade on and your one sigma is violated, don't necessarily freak out and close the trade for a loss because the long-term odds still hold that even though your one sigma was violated you will have the same odds of expiring worthless.

Let's use a fake example. If stock XYZ is trading at $100 and it's sigma is $10, you might want to sell the $110 and or $90 strike prices. For the strangle seller, this information could be important as if the stock hits one of these strike prices, unless it's a stock specific event that you feel the move is warranted, you might just want to hang on until expiration. If the movement is outside the one sigma due to a non-even, just part of the random walk in the park movement, then don't take a loss by buying what will now be an ATM option to close it out.