Search the Community
Showing results for tags 'spx'.
Found 3 results
And the same is true with determining which strikes to use when selling puts. As it turns out, over a relatively long period of time risk and reward are related, as we would expect. Below are the results of three backtests simulating the monthly sale of SPX put options from 2001-2017 at various delta levels (16, 30, and 50). Each test assumes you enter approximately one month from expiration and exit approximately 3 days prior to expiration (results have minimal variation if entries and exits are slightly modified). No active trade management is involved. My tests also assume no leverage is being used, with cash fully secured by 1 month T-bills. In other words, the returns of put selling can be thought of as the yield on cash plus the net results of the option trades. No commissions, slippage, or taxes are included, so real results would be slightly lower. Given the size of the SPX contract and its liquidity, this isn't a serious issue. The strategy passes the test of real world investability. Click on the image for greater clarity Several interesting observations can be seen: 1. All strikes deliver returns similar to the underlying asset (using SPY as a proxy). The at the money strike (50 delta) even slightly outperformed SPY. It should be noted that every backtest is sensitive to the start and end point, and if we started our test in 2009 SPY would substantially outperform. This is what we would expect given the limited profit potential of put selling. CBOE has data going back to 1986 showing that at the money put writing has delivered returns comparable to owning the underlying. 2. All strikes delivered less risk than the underlying asset, measured with standard deviation and max drawdown. This results in higher Sharpe Ratios. The Sharpe Ratio is one way to measure how well we are being compensated for the risk taken, and a higher number is better. Put selling having lower risk than owning the underlying asset is something that we can expect to persist in the future. This is often counter to the perception that many have about selling options being "risky". Leverage is what creates risk, not product/strategy. 3. The farther out of the money strikes deliver higher Sharpe Ratio's, with 16 delta options producing an extremely high 0.92 Sharpe. This could be noise in the data that may not be likely to continue out of sample, but many other researchers have found similar results on additional data and underlying assets leading us to believe it's not random. Those who believed this will continue to persist may choose to modestly lever their notional exposure to produce higher returns instead of selling strikes that are closer to the money. 4. If you look closely at the chart, you can see that put selling can make money even in a declining market (especially with further out of the money options, of course). For example, during the 2001-2002 bear market, 16 delta put selling was continuing to put in new equity highs. This was also the case during the first several months of 2008 where the market was declining, but not too far/too fast. When the crisis hit in the fourth quarter, the speed and magnitude of stock market losses was too great for any of the strikes to endure. No surprise there, as insurance must pay off from time to time to attract buyers. 5. After a crisis period like 2008, put selling recovered quickly as option premiums were substantial. SPY didn't reach a new high until 2013, while put selling recovered it's drawdowns in 2010. This is important given the nature of our human discomfort with losses and our ability to stick with a strategy. You may even draw additional conclusions of your own from this data. So with all this being said, which strike should you sell? I don't think a binary decision needs to be made here anymore than a binary decision on if you should sell puts or just buy the ETF directly. There are advantages and disadvantages of both, but hopefully this has made it clear that put selling is worthy of at least a partial allocation for a particular asset class like US large cap equity. Below I'll present one additional chart that is 50% SPY, and 50% put selling with equal allocation to our three strike levels (in other words, about 16.66% each), rebalanced monthly. Given the certainty of our uncertainty about which method will be best in the future, as well as the simplicity and tax efficiency of a traditional ETF allocation, this approach could be ideal for many sophisticated investors. Click on the image for greater clarity Jesse Blom is a licensed investment advisor and Vice President of Lorintine Capital, LP. He provides investment advice to clients all over the United States and around the world. Jesse has been in financial services since 2008 and is a CERTIFIED FINANCIAL PLANNER™ professional. Working with a CFP® professional represents the highest standard of financial planning advice. Jesse has a Bachelor of Science in Finance from Oral Roberts University. Jesse oversees the LC Diversified forum and contributes to the Steady Condors newsletter.
Study Methodology Here are the specifics of the test: Time Period: January 2007 to Present Trade Setup: On each trading day, locate the standard expiration cycle with 30-45 days to expiration. If the front-month standard expiration cycle did not fall within that time frame, we simply did nothing and proceeded to the next trading day. When the front-month expiration cycle fell between 30-45 days to expiration, we “sold” the at-the-money put in that cycle. In the following standard expiration cycle, we purchased the put at the same strike price to complete the calendar spread. Position Tracking: Each position’s profit/loss as a percentage of the debit paid was tracked on each trading day until the day before settlement. We did this to avoid any unwanted outcomes as a result of the SPX settlement process. Overall, 1,200 SPX calendars were tracked. Once all of the profit/loss metrics were gathered, we filtered the trades into four buckets based on the VIX Index level at the time of entering the trade: VIX Below 15 VIX Between 15 to 20 VIX Between 20 and 25 VIX Above 25 Each bucket had a similar number of occurrences. The Results: Profit/Loss Frequencies for 30-45 DTE SPX Calendar Spreads Let’s take a look at the data! We’ll start with the percentage of calendar spreads that reached profit levels between 10-100% of the initial debit paid: There are some key findings from these results: Over 50% of the trades reached returns of 40% or more on the initial debit paid, with 75% of the positions reaching 10-20% returns on the debit paid. The low IV (VIX below 15) and high IV (VIX above 25) had the highest percentage of trades that reached each profit level. High IV calendar spreads in second place? How can that be? In super high IV environments, the VIX term structure (SPX volatility) goes into backwardation. As a result, the near-term expiration cycles trade with significantly higher implied volatility than longer-term expiration cycles. When volatility comes back down, the front-month implied volatility will fall with greater magnitude than the back-month implied volatility, which can lead to quick profits on a calendar spread if the underlying hasn’t moved too much. The only problem is that in high implied volatility environments, realized volatility tends to be high, which is not ideal for delta-neutral calendar spreads. So, in high IV, long calendar spreads become more of a term structure reversion trade, and less of a time decay trade. Let’s move on to the loss frequencies: In low IV environments, the SPX calendar spreads reached each loss level less frequently than in higher IV environments. Again, this is most likely due to the fact that realized volatility tends to be more significant in times of high implied volatility. When you compare the calendar spread profit frequencies to the loss frequencies, we can see that the biggest gap (profit frequency – loss frequency) is typically in the low IV (VIX below 15) trades. For example, in the low IV bucket, over 85% of positions reached a 10% profit, while 65% reached a 10% loss. In the 15-20 VIX trades, 75% of trades reached a 10% profit, while 85% reached a 10% loss. Of course, you’d like to see a large gap between the percentage of trades that hit each profit and loss level (more trades reach the profit level as opposed to the loss level). To visualize this, we simply subtracted the loss frequency from the profit frequency at each level: As we can see, the calendars entered when the VIX was between 15 and 25 reached the loss levels more frequently than the same profit level (resulting in a negative value). In the low (VIX below 15) and high (VIX above 25) volatility environments, the SPX calendar spreads reached the profit levels more often than the same loss level. Summary What can we learn from this calendar spread profit and loss data? First and foremost, calendar spreads typically perform very well in extremely low implied volatility environments, but they also have potential when front-month IV is inflated at a significant premium to the back-month IV. However, there are strategies that may be more suitable for time when implied volatility is high. Second, based on a positive spread between profit and loss frequencies, calendar spreads have historically had positive expectancy at each profit/loss level, as profits have occurred more often than losses of the same magnitude. On a final note, it’s important to keep in mind that this is a backtest based on closing values. As a result, it’s likely that there were trades that hit profit or loss levels intraday, but ended the day less than those levels. Additionally, these positions were not managed, which is the key to success when trading calendar spreads. At the very least, the data discussed in this post can help to guide management levels, and expectations for achieving certain profit/loss levels when trading SPX calendars. About the Author Chris Butler is the founder of projectoption, an options trading education and research website. Chris primarily trades non-directional options strategies in equity indices (SPX, RUT, /ES Options), but is also active in volatility-related ETNs (VIX Options, VXX, XIV).
This year's sideways market has been very kind to calendar spread trades.We booked few very nice winners with SPX and RUT calendar spreads. When we opened another SPX calendar spread on August 5, I expected another nice winner. But the market had very different plans. The strike was 2100, which was right in the middle of the range for the big part of the year. Last Thursday, August 20 SPX was at 2061, and the trade was still in decent shape, down only 8%: However, as the selloff accelerated towards the end of the day, the trade was down around 20-25%. At this point I considered different adjustment options but didn't find an efficient and inexpensive hedge. So my plan was just to close the trade around 30% loss. I tested different adjustments and didn't find them too efficient, so considering the fact that we also had a butterfly trade that was expected to offset the loss, it was an acceptable result to me. However, SPX really collapsed at the last hour on Thursday, and the trade went through the stop loss in matter of minutes. I assume that most members wouldn't have time to act on the alert sent 5-10 minutes before the close. Spreads also became very wide, and I doubt we could close it anywhere near the mid. On Friday SPX gaped down another 20 points and the trade was down over 50%. By the end of the day the loss was 70%+. We used yesterday's rally to reduce the loss and closed the trade for 60% loss. To put things in perspective, SPX went down 130 points in 3 trading days. Last time it happened was 2011. So what could we do differently? In the wise words of one of our mentors, Dan Sheridan, "just buy a stinking put!" Dan survived on the floor of the CBOE trading options for over two decades, so he's experienced it. Lets see how things would be different with the put. When SPX went through our adjustment point, we could buy the 1750 put for just 0.75. This is how the P/L chart would look like with the put: As we can see, the chart looks much "smoother". But even more important, it significantly increases the vega, which helps in case SPX continues down and volatility increases. Fast forward to Friday morning: That's right. Instead of being down 50%+, we would be actually UP 41%. So what can we learn from this trade? The most important thing is "don't assume anything". Gaps happen and should be taken into consideration. If the market went down 60 points and became oversold, it doesn't mean it cannot go lower. Don't let your opinions impact your risk management. When in doubt, cut the loss or "just buy a stinking put!" This trade emphasizes once again the importance of position sizing. In our model portfolio, we recommend allocating 10% per trade. Which means that this trade had a 6% negative impact on the overall portfolio. Not pleasant but not catastrophic and allows us to leave another day. After closing 9 consecutive winners in August, we are still having a great month, while most major indexes are significantly down. Related posts: How We Made 23% on $QIHU Straddle in 4 Hours How Position Sizing Impacts Your Returns How we trade calendar spreads We invite you to join us and see how we manage our portfolio of non-directional strategies. Start Your Free Trial