yeah, but the closer to the realised median/median of that distribution over time, when estimating parameters, the better you would do right? or even better if you find the mispricings, then you would crush it on those wins. this assumes at least 1000's of transactions a year, but that's probably easily realistic at an IB.
ah.. dude something just clicked... yeah i can see how i was confusing you. since its a vol bet, really only the greeks do matter, cause those estimate relative price. but i think my point still stands that making sure you are getting solid estimates that would make a big difference in long term gains, since you could be right in the middle of that vwap and maximize your outcome and charge the right premium etc., etc.
yeah sorry about that. anyways, possible ways to crush... 1. over time, take directional bets that are normalized to the mispricing and make more on those directional bets than just betting vol. in this case you would capture market vol + mispricing excess = vol excess. normally you're just getting market vol. 2. as a market maker, if your greek estimates are industry standard, but your starting price point estimate is superior, then you would absolutely have stat arb advantage in volume pricing and fees. would you not? that's just inventory management for any business.
The model won't tell what shares exhibit vol of vol upside v volcorr with index, etc. I can tell when people are new to vol-markets as they tend to focus on models like traders applying TA. Find a numerical method, open-form; it's not going to have an impact to PNL in listed vanilla vol. You're asking the wrong questions. The model is the least of it.
You just won't quit. NVDA May 17 880C is 51.00 mid. 1W stress at current vol vs. -100 beeps = $100 in PNL on a $5,100 outlay. The fcking decay is $510. All numerical models I've seen converge to BSM so what are you going to run that is better? You're asking us to prove a negative bc you're pot committed to this stupid argument out of inexperience and ignorance.