How legit is options selling?

Is option selling legit

  • Total voters
    19
  • Poll closed .
nooby_mcnoob said:
Let's say the thesis is that implied volatility is usually over-estimated compared to realized volatility, thus leading to over-priced options. If this is the case 60% of the time, and you sell risk-limited options at strikes outside 1 standard deviation, it seems like a no brainer, similar to how insurance companies work.
More...

This is a common thing people think about options selling. There are a lot of problems however here.

First, a 1 SD movement is great if the tails of your distribution are small enough. This typically is not true in general. The probability of experiencing a large movement in price is much larger.

It is true IV is generally overstated in the price of the options leading up to binary events. However, this doesnt mean you can safely straddle into earnings. There are other risks involved.

Finally, the insurance analogy is flawed. It comes from the realization that selling options (tail risk) is basically what insurance companies are doing. If you sell enough insurance to enough healthy people your overall risk is small. Its that one guy who has a quadruple bypass and brain surgery two-for-one special at the hospital after buying his new car that bankrupts you.

In the stock market the amount of "people" having quadruple bypass and brain surgery two for one specials is significantly higher than the human case. More to the point (and most importantly imo) is that as an insurance company you can reinvest your premiums in order to mitigate some of your tail risk. The money you make on premiums sitting there is basically the "free" capital you are thinking of when insurance companies make money. You can't do that in options.

I like trading options. It's fun. Not as much fun as futures, but it's a mathematically satisfying problem. Don't be swooned by the optionsalpha/tastytrade/etc "sell options for easy profit" trash. It can be done, but it is not nearly as easy as it looks. If it was, Tom Sosnoff's entire staff wouldn't be tits up on shorting puts.
 
nooby_mcnoob said:
Let's say the thesis is that implied volatility is usually over-estimated compared to realized volatility, thus leading to over-priced options. If this is the case 60% of the time, and you sell risk-limited options at strikes outside 1 standard deviation, it seems like a no brainer, similar to how insurance companies work.
More...
The insurance analogy is flawed on multiple levels.

First, let's take the index options seller. It's not the risk of a single company that you are selling but the risk of the whole market tanking. An insurance company would never get involved in a business where they have only one client since diversification is impossible. Predictably, it has some positive expectation since, unlike almost any insurance product, you are taking a risk that you can't diversify. While you are likely to get compensated for it, it's going to hurt a lot once when you are wrong (and you will be eventually).

Things are different for a single name option seller. Your primary risk is that of negative selection - vol looks rich, but it's rich for a reason that's unknown to you. An insurance company takes great pains to avoid negative selection due to information asymmetry, such as forcing customers to undergo the health exams etc. Unless you have a massive information-gathering operation, there are always people who know more than you do about any single company.

Finally, there is a matter of market efficiency. The barriers for entry into the insurance business are pretty high, while any Sweet Bobby can start selling puts until he's making crying videos for YouTube. Obviously, that depresses volatility and makes selling vol a bad proposition a lot of times.
 
Last edited:
gaussian said:
This is a common thing people think about options selling. There are a lot of problems however here.

First, a 1 SD movement is great if the tails of your distribution are small enough. This typically is not true in general. The probability of experiencing a large movement in price is much larger.

It is true IV is generally overstated in the price of the options leading up to binary events. However, this doesnt mean you can safely straddle into earnings. There are other risks involved.

Finally, the insurance analogy is flawed. It comes from the realization that selling options (tail risk) is basically what insurance companies are doing. If you sell enough insurance to enough healthy people your overall risk is small. Its that one guy who has a quadruple bypass and brain surgery two-for-one special at the hospital after buying his new car that bankrupts you.

In the stock market the amount of "people" having quadruple bypass and brain surgery two for one specials is significantly higher than the human case. More to the point (and most importantly imo) is that as an insurance company you can reinvest your premiums in order to mitigate some of your tail risk. The money you make on premiums sitting there is basically the "free" capital you are thinking of when insurance companies make money. You can't do that in options.

I like trading options. It's fun. Not as much fun as futures, but it's a mathematically satisfying problem. Don't be swooned by the optionsalpha/tastytrade/etc "sell options for easy profit" trash. It can be done, but it is not nearly as easy as it looks. If it was, Tom Sosnoff's entire staff wouldn't be tits up on shorting puts.
More...

A trader who I cannot recall the name said the problem with options trading is there are two pieces to it. Retail traders are better off buying calls and puts. Options selling is more geared to big investors and institutitions with their algorithymic applications. They sell retail traders into selling options instead of buying it! And there are those who advocate not even hedging their positions since, they supposedly, win 90% of the time. Optionsellers.com demonstrated to us the fallacy of following such a strategy.
 
Well the dog threw up so the boy woke me up at 3am and this thread is the best gift I could ask for since I can't go back to sleep.

I agree that the insurance company analogy is not directly analogous because of the lack of information and many other reasons. That being said, here is the thought process:

1. Use IV to find interesting strikes (high IV relative to historical volatility, low-ish delta/probability of exercise).

2. Use these strikes to sell a call/put and hedge with a lower/higher call/put

3. Do this enough times and the probabilities work out often enough to give you an income. Not a great income, mind you, but an income.

So if you create a strategy hundreds of times a year that has a 80% chance of making you a profit (not the max profit, just a profit), and the amount you lose is limited by your hedge, you will end up making a profit overall. This is why I'm drawn to the approach, and I'm trying to understand why it's bullshit:

1. Lots of transactions = great for brokers, not so much for you

2. IV may be wrong

3. ????
 
it's not necessarily bullshit. the reason is the same as the rationale for buying stocks: when you buy stocks you earn the equity risk premium. Ditto if you sell vol, you earn the vol risk premium. Neither will give you smooth bank account like returns and you can/will have down years even if the chance of making a profit on any equity buy or vol sell is quite high. Neither will make you rich beyond your wildest dreams because you'd be mad to over-leverage them and blow-up. But the ERP exists and the VRP exists so returns in excess of the risk-free rate are there to be had. Just don't get too excited...it's not a secret and if anything the "buy equities they always go up" trade is relatively crowded. Same with the "sell vol it always works" too nowadays.
 
I cant sleep either noob! I dont see the difference in PnL between the strategy you mentioned and just being long SPY. Plus long equity will give you a nice tax break.
If you are looking to do better tha buy and hold spy. You need to be able to predict something. In your case you want to be able to predict hv better than the markets model (IV).

If you can find different assests
whose volatilities are uncorrelated in both good and bad times and the assests have a definite risk premium attached to them. You can make some good money!! Please notify me if you do find such assets.
 
You must log in or register to reply here.
Back
Top