Quote from trefoil:
I've been thinking and thinking about this, and I don't think you get the point: what you're saying is kind of impossible. The number of outliers, let's call them, varies wildly over time. You would have had lots in the 72-74 bear, which was actually a peak from a series that started back in the mid-sixties, then suddenly for a long time only a very few. Starting in August 1982 lots of up action that would have killed you on the call side. Then nice and trending, mostly. Then in 1987, lots again. Then again in 1989 (United Air mini-crash) 1990 (Iraq invades Kuwait) then in the late nineties they start coming one after the other (the Tequila Mexican peso crisis, the Asian currency crisis, the Russian default and LTCM, and then the dot com bust), then a period of calm where the VIX practically goes to single digits, and then in 2007 the crisis starts (the VIX shot way up at the end of Feb, which looking back is when the fun started) and you continue with that until the spring of 2009, when, by the way, you would have gotten murdered on the call rather than the put side of your IC's.
The variation of these over a long period of market history is pretty severe. We're in a period of relative calm now. (and guess what? We've got someone boasting about his credit spread returns.) For how long? No one knows. Let's just say that the frequency and the severity of discontinuities is itself discontinuous. Therein lies the problem.