In our quest to understand trading options for income, we often consider using the option chain deltas or probability of expiring OTM (out of the money) when selecting the short strikes for our credit spreads. When we do this, we are impacted by skew.
What is skew? In an option chain, the premium received for Calls and Puts with strikes equidistant from the current price of the underlying usually differ noticeably; the Puts generally receiving the higher premium. This reflects the market's concern that prices in the underlying drop faster than they rise, leading to a higher IV (implied volatility) for Puts.
This differential in IV (and hence, premium) between Calls and Puts is called skew. The greater the differential, the greater the skew.
It is often debated that skew is more representative of the future price action of the underlying; and therefore, should be relied on for determining the position of our short strikes. This is also a position Tasty Trade has taken, despite their support of the Probability Model (which relies on standard deviation, or expected move).
Let's recognize that when we rely on the probability of expiring OTM (POTM), skew will understate the POTM for Calls while overstating it for Puts. As a result, our short strike for Calls will be closer to the current underlying price, or ATM (at the money), while the Put strike will be further from ATM.
Tasty Trade recently ran a test over a 6 year period on the SPY (from June 2008) comparing the expected move (non-skew) vs. skew for the monthly only. The test was preformed only on the Put side (naked short) set at 1 standard deviation (84% POTM).
The results: the skewed Put outperformed at all levels. With 72 total trades, the skew had a higher P&L ($582 vs. $341), higher percent winners (94% vs. 89%), and lower max loss ($2,200 vs. 2,773).
It would seem that skew would be the approach to use, but there are some problems with this test that are misleading.
First, simply testing at 1 SD (standard deviation) does not provide us with an accurate picture of the impact of skew over other levels of risk (for example, 1.5 SD and 2 SD).
Second, by focusing on the Put side, we are unaware of the far worse results that occur on the Call side when we rely on skew (or POTM).
And third, simply testing the Monthly leaves out the advantages of the Weekly. Research has shown that the accuracy of the Probability Model increases as duration decreases; and average daily premium increases as duration decreases; both indicating that the Weekly performance would outperform the Monthly.
At Options Annex we performed a similar test using the SPX over a 5-year period (starting on May 2008). We compared the results of expected move vs. skew at 3 levels of risk (1 SD, 1.5 SD, and 2 SD). We also compared the results of Weekly vs. Monthly.
Our analysis covered short Calls and Puts, and $25 wide credit spreads. Given the data for Calls and Puts, we were able to determine the P&L for short Strangles and Iron Condors (for those who like to play both sides concurrently).
In addition, to provide a more complete picture of risk, we also compiled data for both Calls and Puts on challenged (or touched) positions and losing positions.
Here are the results of our analysis...
1) Regardless of whether Weekly or Monthly, the Expected Move had lower or equal Total Losers at all SDs than the Skew
2) The Weekly significantly outperformed the Monthly at all SDs with lower risk.
3) At 1 SD, the Skew P&L is slightly better than the Expected Move P&L for both Weekly and Monthly, but at significantly greater Risk. At 1.5 & 2 SDs, the Exp Move outperformed the Skew in both P&L and Risk.
4) Based on Total Losers, the Skew is equivalent between Weekly and Monthly at 1.5 & 2 SDs; higher by 60% at 1 SD for the Monthly.
5) Using the expected move and a $2.5k spread (IC), the P&L is about half to a third the size of the Short Strangle; the Weekly ROC is significantly better than the Monthly ROC.
Weekly ROC: 1 SD = 1,003.4% (25,042 / 2,500); 1.5 SD = 838.5% (20,962 / 2,500); 2 SD = 548.3% (13,708 / 2,500)
Monthly ROC: 1 SD = 236.7% (5,917 / 2,500); 1.5 SD = 225.4% (5,636 / 2,500); 2 SD = 98.0% (2,451 / 2,500)
NOTE: when comparing Challenged and Losers, multiply the Monthly by 4.33 (the average number of Weekly trades per month) to get an equivalent value for proper comparison between the two.
In conclusion, when a more exhaustive and comprehensive test is performed comparing expected move to skew, Monthly vs. Weekly, we come away with far different results than Tasty Trade. It is now clear that going further out to 1.5 SD and 2 SD offers far lower risk while sacrificing some premium. And, we also learned that the Weekly offers far greater annual P&L than the Monthly (which Tasty Trade favors). This underscores the need to corroborate test results of third parties by doing your own independent research.
If you would like to learn more about options, and how to generate consistent weekly income trading options, go to Options Annex.