Developing a naked puts trading system

[Windlesham1]

Dumb money will always try to find a cheap way to profit-bring it- I like being subsidised

 

[ironchef]

Idiots trade naked-and just because QE has held the fake markets up for so long does not mean they won't crash and hit the skids for a decade or two. The global economy is BS- so much debt and money printing nobody knows what's real any more, and the market does not need another who dilutes option prices even further. Do something intelligent and productive

https://www.researchgate.net/publication/228289846_Why_are_Put_Options_So_Expensive

I don't think you can blame QE for this. And if OP can find a profitable way to trade naked puts, I would be celebrating with him knowing it can be done.

 

[i960]

Naked isn't the issue - it's how far out of the money that one is selling and exposure to gamma that is.

The $PUT index is based on ATM 30-day puts with proceeds placed into treasuries. Last I checked it had a better sharpe than SPX.

 

[sle]

I like being subsidised

Everyone does

PS. Off topic, but is that your rabbit? Very cute

 

[Zazys]

Instead of naked puts, try selling vertical spreads. Less profit, but time works for you. Maybe the same system that doesn't work with naked puts would work with vertical spreads...

I second this approach. Challenge is to set a large number of trades to distribute the risk. Needs a large pipeline and a low cost broker.
Wouldn't like going into a naked trade without position sizing. It is too risky.

 

[Zazys]

Consider this more of an academic exercise about trading system development than anything else for now. If the results look good then I'll think about viability.

RedDuke and ironchef--I have option analytics software that gives me backtesting data to 2001.

Handle123--requirements are significantly lower in a margin account.

I want to focus specifically on how to conduct the backtest and I'll fill in a couple blanks. First, I will enter on every trading day because I want a large sample size. Second, I will sell the first put under X delta to keep probability of profit constant throughout the 15+ year sample. Third, I will exit when premium drops under $0.30 or just before the close on the final trading day. Fourth, this will be a daily backtest and I will collect all data at 15:30 ET.

Any further suggestions would be appreciated.

Have you tried Quantopian?

 

[drmark27]

I second this approach. Challenge is to set a large number of trades to distribute the risk. Needs a large pipeline and a low cost broker.
Wouldn't like going into a naked trade without position sizing. It is too risky.

What do you mean by position sizing?

 

[Zazys]

Position sizing = Not having too many eggs in one basket and knowing what is the maximum damage that can be caused to a single egg.

Some rules could be:

(1) Using up to 50% of total available margin = Overall Position Size Limit, and,

(2) Limiting maximum loss per trade to less than 0.1% to 0.5% of Overall Position Size Limit.

• For Nakeds establishing maximum loss is tricky. One could establish that max loss to be 2 Standard Deviations away from current price.
• For defined risk trades this is not an issue.

What was your definition of position sizing?

 

[drmark27]

...

(2) Limiting maximum loss per trade to less than 0.1% to 0.5% of Overall Position Size Limit.

• For Nakeds establishing maximum loss is tricky. One could establish that max loss to be 2 Standard Deviations away from current price.
• For defined risk trades this is not an issue.

What was your definition of position sizing?

Now you're getting at the reason I started this thread. Is there a best way to handle this?

For starters, what about selling one contract every day?

 

[mushinseeker]

drMark27.. Put that criteria (selling 1 contract every day) to a backtest first, then add whatever filters you have on the same strategy (ie selling puts). Taking that first step is a good way to establish a baseline of whatever strategy you are trying to implement. After that, start adding layers to your criteria to enhance trade yield . As an example, you probably don't want to be selling puts right after a ticker just violated a 50d moving average? Look out below! ... stuff like that.. I just finished an article on my site that did exactly what you are contemplating.. selling 1 slightly otm put spread on the QQQ at 12 IV... won 80% of the time ... but....