Base strike on probability or fixed percentage?

[dyrmose]

When executing my monthly systematic options strategy I'm curios about your thoughts on whether to select strikes based on:

• A fixed percentage; eg. 15% OTM.
• A fixed standard deviation; eg. one standard deviation away.

It's natural that the efficiency will depend on several factors. My thoughts are that one would be able to get more consistent outcomes basing the strikes on standard deviations (using this to create a confidence interval) rather than fixed percentages due to one taking market expected moves into consideration.

Any thoughts on pros and cons on either approach would be greatly appreciated.

 

[The_Krakenite]

Rarely do I go for 1-month options. I prefer 2-month horizons and even further. Have dealt with 2-year long leaps as well.

Usually, I don't go by % but choose SD moves in my strike selections. But this is an overly-simplistic measure...

 

[dyrmose]

I didn't state that the options had 1 month to expiry, just that I executed each month.

"this is an overly-simplistic measure"

Can you elaborate on the above quote in terms of what other quantitative aspects you consider in addition to SD? Furthermore, how do you weigh these aspects?

 

[guru]

Fixed percentage if you don’t mind owning the stock at specific price, or have a feel for the stock and can project specific max %drop just like a non-option based stock trading strategy might, and/or are in the trade long-term. Because long-term all the greeks and SD will fluctuate over time, together with vol surface, so a static SD estimate may be invalid and may need to be adjusted over time.

Otherwise, say short-term, you may want to use option probabilities, not because of estimating an expected move (move of what anyway, stock price?), but maybe because of being able to estimate the expected immediate reaction to any stock price movement, before the options become repriced. Most option strategies use delta because the delta reflects the current vol surface, skews & iv smile, providing an estimate of option price change to the underlying price movement short-term. But after significant price movement, the options can be repriced and greeks can change, which is the basis of continuous delta-hedging.

 

[guru]

Btw, this may also come down to whether you’re using options as an aid to trade the stock and risking the value of shares (for example selling a $400 SPY put using $40K capital instead of buying 100 shares. In which case you’re betting on specific stock movement and using amount of cash similar to trading the shares. Then you can use percentages.

But when you’re trading options for the sake of trading options, for example risking $100, then you’re playing a casino-type game and making bets with the odds reflected by option greeks and probabilities, in which case you may want to use those probabilities.

 

[dyrmose]

Thank you for the explanation, guru.

With fixed SD I mean that every month I would go short / long the same SD away, derived primarily from the implied volatility at that point in time - to ensure that the strike level would change dynamically.

You have a good point regarding focusing more on delta, as it is based on more factors. However, I also think that simultaneously makes it hard to identify what parameter specifically makes delta be at set level; a delta on the same level could be perceived to be the same situation but with impactful confounding variables being relatively different. With that in mind, I think it would be better to base one's strike level on fewer parameters, however, gaining more transparency. Do you have any thoughts on that? And given I wanted these fewer parameters, what other parameters, in addition to implied volatility, would you consider?

 

[newwurldmn]

When executing my monthly systematic options strategy I'm curios about your thoughts on whether to select strikes based on:

• A fixed percentage; eg. 15% OTM.
• A fixed standard deviation; eg. one standard deviation away.

It's natural that the efficiency will depend on several factors. My thoughts are that one would be able to get more consistent outcomes basing the strikes on standard deviations (using this to create a confidence interval) rather than fixed percentages due to one taking market expected moves into consideration.

Any thoughts on pros and cons on either approach would be greatly appreciated.

the decision is either is highly dependent on your strategy or it doesn’t matter much at all.

With so little information no one can tell you what is reasonable.

 

[dyrmose]

the decision is either is highly dependent on your strategy or it doesn’t matter much at all.

With so little information no one can tell you what is reasonable.

I understand your points. However, I was looking for a more general opinion on the pros and cons.

 

[newwurldmn]

I understand your points. However, I was looking for a more general opinion on the pros and cons.

Here’s a a general opinion: it either matters to your strategy or it doesn’t.

For me sometimes it’s critical so critical it defines the profitability of the trade and other times its so irrelevant that it doesn’t matter if the strike is below spot or above spot.

 

[.sigma]

Personally I choose strikes via voluminousity and liquidity.

rarely do I trade ITM strikes, usually always OTM (unless I’m opening a ATM natural fly).

If you have a systematic system (non discretionary) why is this a question? Doesn’t your system trade the strikes that the system deems worthy (regardless if it’s fixed strike or any other way?)