Quote from laputa:
Nitro,
Thanks for sharing your knowledge. It is post like this with high content value that makes people keep coming to ET.
A question pops up my mind after reading your post - If almost all option MM use such model then to what point would profit be arbitraged out? It seems to me if competition get really bad it would all boil down to execution speed or maybe customer order flows for MM/Broker firms like IB. Of course unless someone come up with a better model and could tighten up the spread further and assume higher risk without blowing up. But then the other guy would come up with still better models and faster executions. Would this ever end? Maybe in the end it's all about having customer orders and no one can make money without having a brokerage operation sitting on top of MM operation?
Is that so? Or maybe the market is just so inefficient?
YW.
The thing that separates one option trader from another can be summed up in three concepts,
volatility, volatility of volatility, and the error term in the estimation of the volatility of volatility .
If your estimate of volatility is generally better than others, you will make money trading options, assuming you know how to put positions together. You have to know what tools (straddles, risk reversals, put/call spreads, ratio spreads, butterflies, etc etc etc) to use. When you are handed edge, how to put the pieces together as you get legs filled, or how quickly you hedge naked risk away.
There is very little arbitrage in index options. Since our theos are generated from our estimate of vol, and someone else's is based on their estimate of vol, we may both think that we have an "arbitrage" and be on opposite sides of the trade. In equity options arbitrage is far more common, but not in index options.
MMs are
constantly working on their models. But the key to a good model isn't that it divines the future because that doesn't exist. Good models are highly calibrateable, so that a talented trader can adjust vol, skew, kurtosis, etc as he reads flow coming into the markets he makes. By doing so, he is making markets more aggressively on some strikes or months, and how he does that may depend on the position he already has on. The position manager tells you what your risks are, and you try to balance that out each day. I can't over stress how important the technology backing this operation is. 99% of the software we use is made in house, tailor made. Problems get turned around in minutes, and rarely longer than a day.
I can't stress enough how mathematically precise options MM is, but what most do not know is that behind that precision there is a great trader, called the vol fitter, that makes it all work. It is a high art and probably cannot be automated, although that is debatable.
I guess I didn't really answer your question. All I can tell you is that trading, when done right, is nearly as certain as getting a paycheck, a big fat one.
Insitutions need insurance on their portfolios. MMs provide that at a cost. Is that inneficient? I have no idea. Both sides meet and do business each day.
nitro
) So I guess it's more a matter of directional bias with better odds, as per nitro's 60/40 coin toss analogy. But as the competition intensifies with other such sophisticated players, as you suggested, perhaps the hypothetical 60/40 begins to move closer to 51/49. Of course, at this point I'll believe anything.
