I am familiar with Taleb's writing and spent 7 years in legal theft at the behest of my former employer in structured product. Pricing is done rapidly and often noncongruent features are embedded in a product. Our job is to sell a product, and more often than not the nth-order terms are of little consequence. You price it to sell at $12MM and your sales-team offers it at $18MM. You can assume that the pricing of these items are far more complex than listed equity options.
All the options on IBM are related - some closely, some distantly. Let's start with the two most closely-related options - a put and a call at the same strike. In our previous example, as the IBM 100 calls are being bid up, the 100 puts will be bid up almost the exact same amount because of their close relationship. If I can sell the 100 calls 5 cents higher than I was a minute ago, then I'm willing to pay 5 cents more for 100 puts. There's an arbitrage relationship there. The next closest thing to those 100 calls are the puts and calls at nearby strikes - the 95 and 105 puts and calls. If I sell 100 calls at .55, then .60, then .65, then .70, etc., I am now very short premium and getting nervous. I can reduce my risk considerably by buying premium at the 95 and 105 strikes. So even if the volume in those strikes is zero, my bid for puts and calls at those strikes has risen considerably. Less related to those 100 calls are puts and calls at distant strikes. By buying those, I can somewhat cover the risk I've exposed myself to by being short all those 100 calls. Another way to cover my risk is to buy puts and calls in other months. So you can see that option premium on IBM - also known as time value or volatility on IBM - is an asset class. If I have exposed myself to risk by massively selling 100 calls - being massively short IBM volatility, I'm now very anxious to reduce my risk by buying IBM volatility somewhere else. I'm willing to pay a lot more for EVERY IBM option than I was yesterday, before I sold all those 100 calls. Therefore, the IV on all IBM options will rise.
I'm sorry, but this is nonsense. The illiquid option is orphaned or it isn't. Models exist to price the curve. Many participants would prefer to deal in limited liquidity scenarios.
Yes, of course. I was just referring to what I do, which is extremely liquid options on extremely liquid securities. I have no idea how to benefit from the illiquid options. It might be the whole new game. Unfortunately I have never spent any time on it at all. I was trying to point out that the "swarming" effect is best noticed in something like S&P options. That is all. You obviously have the "first hand" knowledge about OTC options. I respect that. I, on another hand, have no experience with them.