May 3, 2021
Welcome back to a new installment of the options series! In the discussion following the previous post (Part 6), a reader suggested the following: In recent history, the index has never lost more than 50% over the span of one year. Then why not simply write (=short) a put option, about one year out with a strike 50+% below today’s index level? Make it extra-safe and use a strike 60% below today’s index!
So, let’s take a look at the following scenario where we short a put option on the S&P 500 index slightly more than a year out and with a strike about 60% below the current index level:
- Trading date: 4/30/2021
- Index level at inception: 4,181.17
- Expiration: 6/16/2022
- Strike: 1,700 (=59.2% below the index)
- Option premium: $11.50
- Multiplier: 100x (so, we receive $1,150 per short contract, minus about $1.50 in commission)
- Initial Margin: $4,400, maintenance margin: $4,000
In other words, as a percentage of the initial margin, we can generate about 26% return over about 13.5 months. Annualized that’s still slightly above 23%! Even if we put down $15,000 instead of the bare minimum initial margin, we’re still looking at about 6.8% annualized return. If that’s a truly bulletproof and 100% safe return that’s nothing to sneeze at. A 6.8% safe return certainly beats the 0.1% safe return in a money market, right? Does that mean we have solved that pesky Sequence Risk problem?
Here are a few reasons to be skeptical about this strategy…
Problem 1: During the Great Depression, the index lost 70% over a 12-months window!
If we extend the range of the chart to include the Great Depression, we notice that there was indeed a 12-month window where this strategy would have lost a lot of money and probably would have wiped out the account after 1 year. The S&P lost a staggering 70.1% between June 1931 and June 1932! And there was another 50+% loss later that decade. I am not saying that we have to fear another Great Depression right around the corner. But a once-in-a-hundred-year event that can wipe out the entire portfolio is certainly something to worry about. Recall that if you make 6.8% every year for 99 years and then lose 100% in one year, the average return is not 5.8% (6.8%*0.99-100*0.01) but it’s compounded via [1.068^99*0]^0.01-1=-100%. One single accident like that can wipe out an entire century of gains! It’s less of a problem when that happens after 100 years when I’m already dead. It’s more of a problem if that accident happens over the next 30 years. A 30% risk of blowing up the portfolio over a 30-year horizon is not acceptable!
Problem 2: That pesky Option Delta (again!)
The false advertising in this strategy – as well as a lot of similar “easy money” option strategies that are marketed (shilled?) to unsuspecting retail investors out there – is that because we are (somewhat) certain that the index will stay above the strike on the option expiration date we will also have a smooth ride along the way. Nothing could be further from the truth! To understand why this is really dangerous logic, look at the following analogy. I once took a flight from Atlanta to Zurich, and the weather in Atlanta was nice and the weather forecast for the destination was mostly sunny. That didn’t prevent bad weather along the way, including several people around me filling up their barf bags. Some more than once!
So, how susceptible are we to “bad weather” with this short put option? One gauge (not the only one, see below for more details) is to look at the option delta. The option delta is the slope of the option price with respect to changes in the underlying, all else constant. It can also be understood as a very rough approximation of the market-implied probability that the underlying drops below the strike (see Part 4 for the details). I calculate the Delta of this particular option as -0.01464. Thus, a 10-point drop in the S&P 500 index, all else equal, will likely result in an increase of $0.1464 in the option price. And thus a $14.64 loss in your short option position (100x multiplier).
Thus, a 0.24% drop in the index will still cause a drop in your portfolio by about 0.1% if you keep $15,000 in margin cash. You are still holding a portfolio that is subject to market volatility. The claim that you can generate 6.8% completely risk-free goes out the window!
Also notice that if you only hold the initial margin requirement of $4,400, then a $14.64 loss is now 0.33% of that portfolio value, and thus your options portfolio is indeed more volatile than simply holding a 100% S&P 500 index portfolio. What’s more, because you have only a slim margin cushion of $400 ($4,400-$4,000) it takes only a 273-point drop in the S&P index (=10*400/14.64) for you to get a margin call, which means the exchange would liquidate the position unless you contribute more money and bring your account back to compliance with the margin requirements.
Of course, people will now object that the risk of a margin call is much reduced if you simply hold about $15,000 in margin cash. But even that is based on flawed logic, which brings me to the next point:
Problem 3: The other Option Greeks!
Someone who understands the “Option Delta” and only the Option Delta might get a false sense of security with the calculations above (Dunning-Kruger Effect). The problem with the Delta calculation is that it is valid only as a local linear approximation, i.e., for relatively small shocks to the value of the underlying. That’s because the Delta itself will change as the underlying moves (Gamma effect) and when the market drops we also observe a boost in the implied volatility of the option (Vega effect). Both the Gamma and the Vega effect will boost the price of the put option and thus your losses when shorting it. And a large enough stock market drop in the first month alone can still wipe out the entire account, even if a 12-month return of -60% never materializes. Sunny weather in Zurich and barf bags while flying over Iceland!
To gauge the impact of how badly our portfolio might get dinged by just one month of stock market volatility, let’s look at the following thought experiment: Starting with the option above with a price of $11.50 on April 30, 2021, I calculate hypothetical option prices for May 31, 2021, after applying a variety of monthly stock returns and monthly changes in the VIX. Instead of using completely hypothetical SPX and VIX changes, I use the historical monthly changes observed since 1990. So, for example, in March 2020, the S&P 500 price index dropped by 12.5%, while the VIX increased by about one-third (+33.5%). What if that same blowup happened again in May 2021? Let’s take a look. In the calculations below, I reprice the option as of 5/31/2021 after dropping the SPX to 3,658 and scaling up the IV by a factor of 1.335. The option price would shoot up to $65.12 for a total loss of $5,362. More than the initial margin and even with $15,000 in your account initially, you would have lost 36% of the account after a 12% drop in the index. So much for a risk-free investment: your short option account would have lost three times as much as the index!
Side note: This calculation relies on the assumption that a shift in the VIX would scale up the entire Vol curve proportionately. I don’t have the historical quotes to prove that, but my personal experience certainly supports the idea that the vol smile/smirk certainly stays intact and might even be exacerbated during financial stress periods. Maybe my blogging buddy spintwig can take a shot at simulating this with historical options quotes?!
OK, so this was one single month, how about all the other monthly SPX and VIX changes since 1990? Let’s plot that in the chart below. Historically, the worst monthly P&L would have been a loss of almost $28k in 2015. That wasn’t even during a bear market. The surprise Chinese Yuan devaluation came out of “left field” and sent the stock market tumbling (-6% for the month) and the VIX more than doubled in one month. But other stock market crashes would have also caused some nasty losses. Quite intriguingly, out of the six losses worse than $10,000, only 3 occurred during a recession/bear market, the rest were simply short corrections:
- August 1998: Russia Crisis, LTCM (no recession, no Bear Market)
- Both September and October 2008: Global Financial Crisis (recession+bear market)
- August 2015: Chinese Devaluation (no recession, no Bear Market)
- October 2018: Q4 scare, growth concerns, monetary policy uncertainty (no recession, no Bear Market)
- February 2020: Global Health Crisis (recession+bear market).
Also, 40% of the monthly SPX and VIX changes would have caused losses in the short put position. About the same “hit rate” as a plain-vanilla equity index fund!
Problem 4: It looks even worse when using intra-month data!
Note that the worst monthly drop during the 2020 bear market came not from the dramatic decline in March but the prior month. March 2020 was “only” the 17th-worst month when measured by the short put option P&L. Does that mean we didn’t have to worry too much about the crazy market moves in March that year? Not exactly. The bottom of the bear market was on March 23, 2020, and the market subsequently recovered quite nicely in the last few days. The S&P 500 recovered 15.5% by 3/31 and the VIX also declined quite noticeably.
But if I were to feed in the 31-day move between February 21, 2020, and March 23, 2020, with a 33% drop in the S&P and a 261% increase in the VIX, the numbers would look much worse. With implied volatility all the way at 153%, the P&L from the option would have been, wait for it, -$77,167. This would have wiped out your $15,000 more than five times over. How crazy is an IV of 150%? Not that crazy. During the volatile weeks of March 2020, I routinely traded my short puts with implied vol numbers between 120 and 150%. The VIX index exceeded 80 during the month of March, and due to the volatility skew, it’s certainly realistic that a way-out-of-the-money put option will easily have an IV twice the value of the VIX. For example, on 4/30/2021, the 1700-put had an IV of 42.62% when the VIX was at 18.61!
The same criticism applies to many of the options strategies marketed to unsuspecting investors!
The option gurus that want to sell you courses on how to make 30%+ annualized returns make the same mistakes:
- They exaggerate the expected returns by using the minimum margin requirement in the denominator. No serious options trader would ever calculate the expected returns this way! If you see anybody calculate the expected return this way, and most option gurus on youtube do this to market their 30% or 60% return claims, you should not walk, but run away as fast as you can. These guys have more experience in selling courses than trading derivatives. To hedge against even moderate market moves you need to have a margin cushion much larger than the initial/maintenance margins. And longer-dated options have a lot of Gamma and Vega risk!
- They exaggerate the expected returns by using the gross option revenue in the numerator without accounting for losses along the way. In my personal experience trading my strategy for about 10 years now, I’ve sold options with a combined premium of over $1,000,000, but I also paid out around $370,000 when my short puts ended up “in the money.” I currently have a 63.9% “hit ratio.” But to be on the conservative side, I budget a 40% hit rate just to be on the safe side in case I was just plain lucky during the last 10 years.
- They ignore the delta/gamma/vega risk along the way and only show you where the underlying has to land on the expiration date. That might look like a safe bet, but ignores all the things that can go wrong in between – think of the barf bags in the airplane!
Similar mistakes also resulted in the optionsellers.com meltdown and the UBS Yield Enhancement Strategy losses. For example, most of the Nat Gas call options that optionsellers.com shorted indeed expired worthless in the end. They would have made a ton of money in February and March 2019 if they hadn’t suffered catastrophic losses in November 2018, causing a total loss of all customer accounts, plus additional losses, with the clients holding the (barf) bag!
My personal options trading involves ultra-short-term put writing with a little bit of leverage. I target around $140,000 in margin cash for every short put option. As I detailed in previous posts, the rationale is the Central Limit Theorem from basic math/statistics: By averaging my investments over 150+ (mostly) independent bets every year, I can generate relatively stable income without too much downside skewness. This suits my retirement cash flow needs perfectly. Even though I need to sell with strikes much closer to the current index level and I suffer occasional losses I view this as the safer strategy. The market can meltdown much worse over a 1-month period than over a 2-day period. The return profile of the 1-year-out short-put option I studied above is actually so horrendous over the first month of the contract, it looks more like a candidate for a cheap hedge, i.e., go long that put to insure against a “black swan” event.
If you want to employ one of the strategies involving short put options with a long time to expiration, make sure you don’t put all eggs into one basket. Maybe stagger the contracts so that you sell a put every month using only 1/12 of your money. But a back-to-back meltdown like September/October 2008 or February/March 2020 could still wipe out your portfolio. Another idea would be to use credit spreads (vertical spreads), so you hedge the downside by going long a put with an even lower strike. But that eats into your already-low profits.
Sorry if I unloaded on that poor guy who left those comments last week. I want to end this post on a consolatory note and wish him good luck, though. If Miguel’s portfolio doesn’t blow up soon we will all have a smooth ride going forward! Cheers to that! 😉
Hope you enjoyed today’s post! Looking forward to your comments and suggestions!
Other posts in this series:
- Trading derivatives on the path to Financial Independence and Early Retirement
- Part 1 – Intro
- Part 2 – Extended Intro
- Part “2.5” – Trading like an Escape Artist: October 2018 update
- Part 3 – Strategy details as of 2019
- Part 4 – Surviving the 2020 Bear Market!
- Part 5 – A 2018-2020 backtest: Guest Post by “Spintwig” (plus a quick update on last week’s volatility)
- Part 6 – A 2018-2021 backtest with different contract sizes: Guest Post by “Spintwig”
- Part 7 – Careful when shorting long-dated options!
- Part 8 – A 2021 Update
- Part 9 – A 2016-2021 backtest: Guest Post by “Spintwig”
Title Picture Credit: pixabay.com
114 thoughts on “Passive income through option writing: Part 7 – Careful when shorting long-dated options!”
This is a great writeup! It just so happens that I have some relevant data 🙂
TLDR: a portfolio can indeed blowout without a position ever going ITM.
One of my private research clients asked me to do an empirical study on a tail hedging strategy that was performed by another researcher using theoretical data (they authorized publication of the research so I’m free to discuss openly). The vehicle of choice was a 90DTE put on SPX rolled at 60DTE.
Here’s the link: https://spintwig.com/spy-long-put-90-dte-10-and-30-otm-tail-hedging/
The thesis was that buying a very far OTM position is 1) less expensive, 2) and offers a greater degree of hedge, 3) while having a similar amount of certainty that the hedge will “activate” vs a far OTM position. 30% and 10% OTM were the respective definitions of “very far” and “far.”
The premise is that Volga (the rate of change of Vega) is materially higher on a 30% OTM position than on a 10% OTM position while all the other greeks are roughly the same. This would cause the position to experience explosive growth in Vega while delta would experience a more linear behavior.
Said another way, the repricing of risk is a better hedge (i.e. experiences more price action) than a delta hedge.
ELI5? One can make (or lose) a ton of money without a position ever going ITM.
It turns out this is indeed true.
If we look at the charts in the linked study, a quick glance reveals that a sudden left-tail-like move in the underlying (Feb/Mer 2020) caused the 30% OTM position to experience a much greater response than the 10% OTM position. In fact, the 30% OTM position offset all historic losses and even turned a small profit!
Now, imagine if we were on the other side of this trade and had sold that far OTM put. Ouch!
Seems you’d want to turn on this approach several years into to market run up, and then turn off after the big drop. Maybe timing isn’t possible, but after x years or y% above market trendline growth might make sense
Another approach would be to buy the hedge to prevent SoRR. While we always like positive tail events, cutting off both tails seems to be a fine compromise in retirement.
The count of occurrences is far too small to build a signal with a respectable confidence interval. Continuously running it would be a requirement.
This strat didn’t do much during the slow grind down of 2008/9. SoRR would still be an issue. This is only good for sudden, severe shocks.
Good point. # of observations is too small. That’s why I like the 3x weekly trading. 🙂
That hedge indeed looks attractive as SoRR insurance!
Reminds me of the issue raised by someone in an earlier post: LEAPS premia for ATM Calls 2+ years out are only 4-5% p.a. That would be another nice way of hedging that risk!
Is this correct….
In execution with 420k portfolio, you’d buy the 1 SPX ATM call with maximum DTE and invest rest in mix of munis, preferred, and other safe-ish items.
The call is ~4.5% drag and the investments are a ~4.5% gain, so limit the downside and participate in the upside.
Certainly will be even better when VIX is down near 12 again. Different strategies for different climates.
Yes, that’s correct. Now the yield is around 4% for the ATM call for Dec 2023. But note: you also lose the ~1.6% dividend yield in the index. So, the drag is a little bit bigger. But then again, you get the yield from your Muni/Preferred portfolio
Curious what you think about a deep ITM SPX call with maximum DTE. Looks like drag is much less (<1%). I realize that a significant drop in S&P could cause loss of premium and premium is quite hefty (e.g. 50%). Any other factors in the risk part of the equation? In a bull market environment, seems to mimic S&P but with very low drag.
You mean going long an ATM or ITM call with a log DTE? Not a bad idea. The premium is pretty low (~4.5-5% p.a. for ATM 2 years out), and that’s a small premium to hedge the downside in retirement.
Still in accumulation phase so thinking deep ITM SPX calls with 2 year DTE as way to get abut of S&P leverage (perhaps 2x). I’m thinking small percentage of overall portfolio as a way to get my feet wet. For retirement phase – I like ATM calls like you said to hedge downside of retirement. By the way – really appreciate everything you do. You’ve opened my world to other possibilities and if anything it has been a great learning experience and intellectually stimulating.
Good strategy. One could also do a hybrid strategy. Say do (100-x)% equity index fund and the rest through ITM LEAPS (2+y time to expiry). That way, the leveraged part has a floor, as a safety net.
For large accounts, it’s also best to stagger the long LEAPS, i.e., always have multiple LEAPS with multiple staggered expiration dates.
But: only recommended in accumulation phase when you still have aggressive contributions! 🙂
Thanks for the link! Yes totally agree! The vol blowup does more damage to the short put than the Delta/Gamma. And it makes sense looking at how tail event probabilities change: the strike might be 3 standard deviations out of the money initially, with a very small probability. Double the IV and suddenly that’s only a 1.5-sigma event (not even counting the impact from the index drop), and the probability just went up 50x.
Unloading on Miguel? – I don’t think so. You basically saved his life (financial) – “He gets to keep his money” and that is the biggest reward.
I don’t remember if you mention this in previous posts but what I like about the short terms options is that I get to turn the same money three times a week. So $10 in normal weekly options turns to $30 every single week.
May be there is more money to collect on weeklies but why in the world would I have 5 days at risk instead of ~2 days at risk? For me this strategy has a very long future. I think it is also important to mention that one should have extra capital reserves ready to take take on opportunities when there is fast slide down.
I talked to another friend that does weekly options on various companies and he won’t do short term options because he is afraid of market melt down, which is an oxymoron as in most cases single company carries lot more risk than market itself.
Haha, that’s a very consolatory way of putting it!
Yes, and definitely, I make very little money, but making it 3x a week, 156x a year adds up. 🙂
I used to do only Friday->Friday but after going to 3x a week I never looked back. During the bad market blowups, it only takes a few days of negative momentum to cause serious damage.
It seems like all this analysis is assuming that you’re selling naked puts. Why not sell a spread, or some other options strategy that has a tiny hedge built in? You could buy some very far OTM puts for a cheap price, and that eliminates most of your tail risk.
Here’s what I’m thinking:
1) sell options that are close to expiring, when theta is highest
1) buy long term, very far OTM options, which are cheaper and have lower theta
Did I just come up with the perfect plan to make infinite money with no risk? Let me know…
Commissions will eat you alive. Here’s the breakdown on commissions for a 45 DTE short put spread. Shorter durations will simply have no meat on the bone to absorb the transaction costs.
I want to add to the conversation:
1. It seems to me that the brokers are always inventing new setups to make more on commissions.
2. Then I ask myself – How in the world would a reasonable person manage these positions? Because sooner or later we will make an honest mistake, which costs money. So, effectively, you could die by “thousand paper cuts”
3. I personally want to live my life and do other things than watch every single minute the market or do very complex setups and than have to think of every leg every single time.
I think – this is a perfect situation of “less is more”.
All true. I trade 3x a week because it’s fun. I can sometimes do all my trades on a single ski lift run (~3 minutes) and be done for the day. And have fun!
There is no requirement that you need to watch this all day.
In the accumulation phase is it safe to use 50% of the proceeds from income selling in the SPX to buy more shares of the whatever is in your portfolio (more index fund shares)?
But: For me it was the other way around. I liked to slowly transition to more options trading, so I plowed my index fund dividends into my IB account.
I was actually thinking of buying puts with a DTE of one year or more, so the commissions would be negligible. The spread is more of an issue I think. Well, that and the inherent time decay of buying options. But might still be worth it, if it allows you to leverage more without the risk of a sudden crash + margin call.
One year out and more is not a bad idea. I never implemented it but it sounds like a good insurance against vol spikes. The key is the right ratio between short pus with 2-3 DTE and long-dated long puts.
You can do spread trades (as I suggested in the conclusion) but that reduces your profit and you’d have to use even more leverage.
But I agree with the suggestion: BUY this put as a tail hedge against a market blowup in my SHORT-TERM options. The same is true for long VIX futures, which would be an even purer hedge against vol only without the time decay in a 12-m short put.
But again: the tail hedges eat into your put selling profits and they are never a perfect hedge. There is no infinite money without risk. But I’m sure you knew that! 🙂
All cash-secured put selling is, to some extent, a story of pennies and steamrollers. In the event of a 30-70% correction, I would not want to watch my long positions go down AND have some short puts blowing up. As volatility went up, those puts would rapidly shoot up in price. I’m not a big fan of covered calls for the same reason. If my underlying tanks, I might look for a consolation prize in my far-OTM short calls, only to find they didn’t go down as much as I’d hoped because volatility propped up their prices.
If we’re looking to reduce portfolio volatility/SORR, perhaps we should consider complimentary bear spreads.
Suppose I buy XYZ at $10, and also enter a bear call spread at $11/$12 for a credit of 20 cents/share.
> If XYZ goes down, I outperform the alternative of only holding XYZ by 20 cents, or 2%, no matter how bad it is. Meanwhile, rising volatility works in my favor as a force compressing the value of OTM spreads.
> If XYZ goes up, but no more than 10%, I keep all the appreciation and get to keep the 20 cents too. Outperform again.
> If XYZ goes up between 10% and 12%, I match the performance of only holding XYZ, but I did it with 20 cents less risk.
> If XYZ goes up >12% my position underperforms the alternative of only holding XYZ. But this underperformance is capped at 80 cents ($1 spread risk minus 20 cents received). I still can’t complain. After XYZ has appreciated beyond $12 I resume participating in the upside dollar for dollar.
All the while, I did not have to maintain a margin reserve. My broker sees it as a covered call plus long call.
In this case, losing the spread bet means winning the long stock bet, and losing the long stock bet means winning the spread bet. Sure, the hedge is a small one, but the typical outcome would be collecting the premium of 2% of the stock’s value. Every time a spread expires, I lock in a gain and roll again for another gain. Early assignment would be an double bonus, because in addition to pocketing the remaining time value on the short call, I could sell my long call plus its time value, buy XYZ again at the market price, and repeat.
If someone wanted to retire ASAP but was 20% away from their 3.5% WR FIRE number, this strategy might accelerate their trip to the finish line, at the expense of some of the chance of overshooting it, which doesn’t matter as much.
The numbers you posted, are they based on legit option quotes? Is this for a year out? If so, the plan would work beautifully if you pocket the 2% extra return and you lose the upside less than 1 in 5 times. Right now, the yield p.a. for the May 2022 calls SPX4600C is 2.27% and for the SPX5000C is 0.68% for a net yield of 1.59%, not 2%.
But there are still some hickups. If the S&P drops 50% in one year and then recovers 20% for a few years after that, you lose the opportunity to recover. A SoRR nightmare.
So, yeah, there are environments where this works great. But I wouldn’t bet the house on it.
The numbers are for illustration only, but go out far enough in time and a 20% return is available on a spread 10% and 20% OTM. In the case of SPY at today’s volatility, the 455 and 500 calls are $8.77 apart, if my after-hours quotes can be trusted, at the June 17, 2022 expiration 409 days out. This would be a 2.1% consolation prize on one’s SPY position if it failed to go up at least 10% in that time.
That’s not much, but it is enough to buy a put. In a strategic sense then, selling unneeded upside could solidify one’s planned retirement date.
Note that if I used index options for the spread, my broker would see it as separate from any S&P500 ETF I might own and place a maintenance requirement on it!.
Are you sure about your first point ? Rising volatility is bad for a short call spread position since your net vega will be negative
This is hardly passive if one has two open, close, and monitor positions 5 days a week. We might as well call landlording with a manager passive then.
Barrier to entry to intelligent options trading is quite high as well. Only a few truly interested would bother.
Sorry, you misread the posts.
1) you trade 3x a week (M/W/F).
2) if you like you trade only once that day. Takes me 3 minutes. A ski lift ride.
3) I don’t trade to close any positions. The options simply expire
Using TD Ameritrade.
When I sell, on Monday, put options expiring on Wednesday, the margin required doesn’t go away at 3:59PM on Wednesday. They put options do not expire at 3:59 PM on Wednesday.
Why is this relevant?
Because I cannot sell, on Wednesday, new put options expiring on Friday. I have to wait until the next day. I have to wait until Thursday morning, for the Wednesday put options to expire. Only on Thursday morning does the margin required go away. Only on Thursday morning, can I sell new put options expiring on Friday.
Am I using the wrong brokerage? Am I missing something?
Of course, you have to satifsy margin for the 2x puts.
With portfolio margin, though, the puts expiring the same day should already have a lower margin requirement if it’s a minute before expiry and they are 100 points OTM.
So, with Portfolio Margin at IB I never ran into constraints. Maybe TD is not set up for serious traders.
I mentioned “Portfolio Margin” to the help desk and they explained it to me.
Portfolio margin is available to qualified TD Ameritrade clients who currently have a margin account and meet the requirements outlined below:
* $125,000 in current equity
* Must achieve a score of 80%, or better, on an option test
Until yesterday, I had no idea that there were two ways to calculate margin: portfolio margin uses a more real-time calculation of margin requirements, while Reg T margin which is inflexible and uses fixed percentage margin.
I have submitted my application for Portfolio Margin.
I keep learning more from your blog. And also, keep EARNING more. You have my deepest gratitude.
There’s also SPAN margin which is a third kind that’s pertinent for futures and options on futures.
Hedging the SPX puts is not cheap and can really backfire. I have a real world example that bit me recently. When the VIX hit it a 12 month low in early March, I decided to buy 4 VIX futures contracts expiring in mid April as a hedge.
I figured the VIX wouldn’t drop much further which was very wrong. Over the next month, I lost $30,000 as I watched the VIX continue to plummet. This one hedging attempt nearly ate up the entire amount of premium I had earned so far this year selling SPX puts.
It has been about a year since any of my SPX puts has lost money, but that hasn’t that stopped me from trying to “improve” on Big Ern’s strategy.
Try it again if the VIX drops to 12. 🙂
Also, not sure how many short puts you had, but 4 VIX futures sounds like a lot.
One should also slowly phase in and do staggered VIX futures. Not all at once.
Thanks for the suggestion! I was hedging 8 SPX puts, so I thought 4 VIX futures seemed about right, but maybe I overdid it a bit. Live and learn!
That’s definitely too many VIX futures. I believe the rule is 1 /VX per 100k. If you’re using portfolio margins, that’s approximately 1 /VX per 3-4 puts.
Back when the S&P 500 was around 3000, I was targeting around $125,000 in margin cash for every short put option. But now with the index at 4200, I target around $170,000 in margin cash for every short put option (so as not to exceed 2.5x in leverage).
With the CAPE at stratospheric levels and the VIX having fallen to a more “normal” range, I’m even considering targeting a higher margin cash per put option (in the 2x leverage range.) Basically, I would be selling 6 SPX puts instead of 8. Too timid?
I’m still doing the number of contracts but target a lower premium (and thus delta). I’m still targeting about $1.00 premium for a M->W option.
That has increased my leverage to about 3x ($140k per contract). But I generate enough income to live comfortably in retirement!
Okay, makes sense. I have been targeting $1 premium options as well rather than 5 delta. That saved me on Wednesday. All 6 of my puts had strikes of
3065[4065, corrected], so instead of making $600, I lost $600. Would have been much worse if I had sold 5 delta puts on Monday.
That’s a nice “point landing”! Great job!
Strikes were 4065
Curious Big ERN if you would be interested in an Options based post for us mere mortals on our way to FIRE? This was mentioned by a poster in part 6, but I believe it carries a lot of weight. Some of us want to hold more index, and could certainly be paid the premium, plus the bonus of buying it low (dollar cost averaging down?) when we do go ITM. I realize it’s not your current strategy, but perhaps combined with the smaller contracts you mentioned, you could highlight a path to a smaller portfolio on its way up.
Additionally, I noted an Australian poster. As I’m Canadian, I’m wondering if you, or a guest poster could highlight not only fees (as per part 6) from different brokers around, but perhaps at home indexs that may be appropriate for us (Canadian here) from around the world (Or the comparison to the flexibility of the SP500 option runs). Might be a far out consideration, but would be wonderful all the same if you could.
For Canada, I’ve found quite a bit of information from the Montreal Exchange’s “Option’s Matter” Blog (Chicago’s equivalent). Great for Canadian content, contract schedules and whatnot.
Others have written about covered-calls and I may or my not weigh in on that. It’s on my maybe-to-do list because I’ve been playing around with that. But with only limited success. But maybe that’s a story writing about, too.
Also, I don’t want people to get the wrong impression: while some option math is essential, it’s not like you need a Ph.D. to do this. It’s a relatively straightforward strategy.
If someone wants to do a guest post and look into alternative indexes, I’ll be open for that.
Even though, most non-US traders will likely still have better success with the US indices.
Always great to see another installment in the option trading series. 🙂
Glad you liked it! 🙂
I feel I have to share. I am fairly new to put writing and this week was my first ever vol spike (yeah it wasn’t that big of a spike, but still), and I must say, it wasn’t a pretty sight. On Monday morning I sold 3 contracts of 405 SPY Puts at $0.1. I had the pleasure of seeing those contracts jump to $1.2 at the peak of the market turmoil. Of course, this is not a lot of money, and even if they ended ITM the net loss would have been fairly small. But I wasn’t ready to buy 300 shares of SPY, and it kept me up at night. Ultimately I bought to close the contracts at $0.15 minutes before the market closed, fearing that the market will move against me after hours and preferring to take the very small loss to being assigned.
-3 contracts is too much for me right now. If it keeps me up at night, it’s not worth it.
-If I intend to write more than one contract, then one in the morning and one afternoon.
-More margin cash.
-This episode reminded me how human investing is. I’ve been rocking a 100% stock portfolio for years now, and I learned to ignore the noise, but this was all very new to me.
(I imagine I wasn’t the only one here whose options went ITM this week.)
Wow, good timing! It’s important not to lose your nerves! Good lesson!
And with a 100% it’s easier to ignore the noise. A drop one day will recover eventually.
With options you could lock in that loss and it takes a while to dig out of the loss. Sometimes, shorting puts is a lot harder ton the stomach than just buy-and-hold stocks! 🙂
Thank you for this post. I will have to re-read it several times but the information is extremely valuable and really illustrates the convexity of long dated OTM options. I have been seeing several people lately who have been selling OTM short dated options the past few days take severe losses due to the gamma increase. A .20 or .30 cent put option can jump to $4.00 in an instant and that wipes out weeks worth of profits. I am looking forward to reading more article. I do enjoy specifically options strategy and selling options. Btw, I went to Purdue had no idea you were also there too.
Wiping out weeks of profits is not a problem if that happens only 3x a year. 🙂
Anyone else get totally hammered this week selling puts? or just me… The market really tanked at the end of day on Wednesday, and I suffered about a 50 point loss on each put.
I did. Though not 50 points ITM. Not sure how you got strikes at 4110+. My strikes for the Wed expiration were 4060-4080 and the 4065-4080 lost money. About 2.5 weeks worth of income. Well, the first loss this calendar year. Not so bad.
heh, simple answer for how I had high strikes… I was experimenting with different deltas and time periods. Got greedy, I guess.
Gotcha! It’s good to experiment with different deltas. It’s a tradeoff, though: higher premium income means more frequent ITM events. 🙂
Probably you opened puts just before close on Monday. If you open before the meltdown started you would been 59 points ITM easily. Do you open traded during specific times of the day or any other criteria?
It was more like the middle of the day on Monday. Right when SPY was peaking. Then spy bottomed out right at the end of the day on wednesday, ugh. Over $15,000 lost in moments, while regular spy has mostly recovered. Not really feeling the “low volatility” right now…
I try to avoid the very beginning and end of the day, but I don’t really have any criteria for timing, I’ve just been doing it whenever it’s convenient for me and there isn’t a huge spread.
Yeah, I had bad timing, but in the end, it worked out with Only a few $ ITM, on average. Will make the money back by early next week! 🙂
I did better than that. I hammered myself! On Friday 5/7 some of my covered calls were assigned on hundreds of thousands of dollars of ETFs. I entered Monday 5/10 very cash-heavy, which is not where I like to be in a bull market. In hindsight, this was all a gift from the options gods, but I failed to notice it. Had I gotten busy or delayed just a few hours, the equivalent of a new car could have been mine.
Yet, I did notice the markets were starting to droop a little so I figured that would be a good time to sell some 5/14 ATM puts and grab that falling knife like a boss. Within a couple of hours, five-figures of bright red numbers appeared on my account. Oh well, it’ll all work out! My short puts might break even by Monday.
Yeah, I got the same with some of my stock lots. Sometimes you can enter again at a lower cost! 🙂
I had 8 puts that were 8 pts ITM! My first loss since beginning last September.
That’s manageable! $8 after so many gains! Let’s hope the next loss doesn’t come for another 8 months! 🙂
I hope so – fingers crossed!
Thanks for the informative post. The way you narrated the post is good and understandable. After reading this post I learned some new things.Keep posting. Please let me know for the upcoming posts.
I was wondering what you thought of something like the SWAN etf in retirement? It uses 90% treasuries plus 6-12 mo ITM calls for what it calls a 70% exposure to equities. Its been around for 2.5 years with decent returns (35% total) and very low volatility. It barely (<10%) declined in Mar 2020 for instance.
I suppose the biggest risk is when bonds and equities decline together in a stagflation scenario, though the equities downside would be capped by the calls
I like that idea in general. I’d propose the 2y+ out ATM calls plus the rest in bonds. If you do everything yourself with a margin account you can save the expense ratio and further custom-tailor it to your needs, e.g., stagger the contracts.
So, yes, as a SoRR insurance in retirement this would be a viable “option”, pun intended.
Would it be better tax wise instead of the long ITM call to hold the underlying equity etf & buy a leap put for basically the same exposure & downside protection but not have to realize the gains on the upside of the underlying equity until you want to rather than when the ITM call would expire. For many retirees, they are in 15%+ LTCG while accumulating and many times 0% while retired.
This is a great article you post here its help so much thanks for sharing with us.
I have been selling long term puts (ES, SPX) for a while (before pandemic) and it works. between 90 to 150 DTE. Extra 2-4% yield. It’s not much but it’s nothing to sneeze about. The key is sizing. If you go big, you will blow up. There are a few things that I think now is actually good time to do this.
1. VIX.. VIX is still pretty high. Not sure people remember but for a long time (like 2015 to 2020 before pandemic), most of the time VIX was less than 15, in fact 12- 13.. At that time, probably long term SPX put was real cheap and it was easy to blow up if you don’t hedge. 2018 Feb was the example. Many people blew up. But now especially longer term has pretty good juice, implied volatility of SPX is around 20-22.
2. Due to long term SPX volatility being high and in a smaller position, you can close the position if it goes against you. Longer term moves slower than shorter term. I had /ES short puts during 2020 crash and lost but nowhere near to blow up because my position was small.
3. I try to open short position only when SPX is significantly down. Now that’s pretty subjective but I try to do when SPX touches certain moving average (usually it tends to bound those lines). There is no certain rule, on average I sell maybe once a month or two. And limit
4. Sometimes I buy very cheap options when SPX goes up a lot. I didn’t experience how much protection it will give in real world but on simulation, it’s not that bad. Although I don’t think I need to buy puts, but just to be extra cautious in case astroid hits or nuclear war breaks out.
It’s not my main strategy but it has generated about 2-4% annual (for about 3 years) with little effort and reasonable risks This year has been real good but will see.
Glad you were successful with this strategy. What is the 2-4%m though? That’s the gross option revenue before accounting for any losses?
That would seem awfully low.
Net gain in the whole account. For example, account 100k size, net profit would be 2000 – 4000 at the end of the year. Loss rarely occurs as expected unless it’s really quick and swift downturn. I lost 2020 crash but not 2018 october to december. I just waited and won 🙂
Like I said, it’s not my main strategy but 2-4% this kind of dividend I am happy to take it.
Thanks for the info. If you make enough income with the rest of the portfolio and you add another 2-4% from a little bit of occasional options trading, then that’s very impressive! Congrats!
FYI I do other strategies and have core stock index mix. This is just to squeeze a little more return. Even a couple more percentage a year makes a huge difference in the long run.
Understood. I have too much equity exposure in my other portfolio and prefer to do my daily options trading with only the fixed income portfolio to hold the margin cash. That explains our difference in appetite for DTE and gamma risk. 🙂
Right now, when you check around 100 DTE SPX puts, you can easily get about 5.0 per SPX short contract on delta 1 or 2. It’s heaven right now. Especially, when you wait SPX down day, like more than 1%, VIX goes up to almost 20 and SPX long term premium is real juicy. I even go lower end of delta 1. Before Covid crash, it was hard to get even 1.00 🙂
Nice! For that kind of strategy, yes, you probably want to time the vol, just like you proposed.
So I’ve been following the ERN strategy for about 4 months now and one issue that I struggle with is timing of execution. Due to gamma risk, the exact time of trade execution throughout a trading session has a major impact on my P/L and strike choice. For example, it is common for me to write a 3-delta SPX put with 2 DTE for around $1.0 premium, only to to see the option’s price jump to $4.0 half an hour later due to the index making a lousy 0.30% move. Had I only logged in a bit later, I could have sold the same option for 4x the premium or receive the same premium at a much lower strike.
Now, I realize that I can set a limit order above the the current price and hope to catch these moves in the option price, but that risks getting into a bad position in case the market actually drops, and risks losing on possible premium if nothing happens.
From what I understand, it is well established that the market tends to go up during pre and after market hours, and down during daily sessions. Is this something that we option sellers can harness? Is there an edge in opening positions at the end of the day?
That scenario can cut both ways. There are many times when a position is opened first thing after the opening bell and is closed within minutes or a few hours, having hit profit targets.
As for after-hours outperformance, that trend reversed circa 2016: https://spintwig.com/making-money-in-your-sleep-a-look-at-overnight-returns/ (pardon the rough nature of this data / post, it was one of my early non-option backtests). AAPL, the largest constituent of SPY by weight, depicts this reversal more clearly: https://spintwig.com/aapl-long-day-trade-equity-backtest/#Discussion
Very true! Most of the time, I already make a substantial portion of the option premium for the Monday option by the Friday close!
You are probably talking about the September 10 small market meltdown at the end. Part of this is Gamma. It’s probably more Vega, i.e., change in the prce of the option due to changes in implied vol.
That said, Friday’s move is the exception not the rule. I’ve seen many cases where the market drops after I sell my puts intra-day, but due to time decay (aka Theta effect) I still made money on the Monday options by the Friday close.
I sold my Monday options with strikes between 4350 and 4400. Should all work out in the end,
I think you just have to learn to live with the strategy sometimes working against you like what happened on Friday. Most of the time it still finishes OTM, but of course sometimes it doesn’t and you lose weeks of premium.
Well said. That’s my philosophy.
How do you deal with the fact that Interactive Brokers underestimates IV?
You can calculate the IV directly from the options prices and you can see that IB consistently underestimates the IV that they show in their UI.
The 6 deltas you have been selling are probably more like 8-10 deltas if you go by the Black-Scholes model.
Indeed, the delta value displayed by IBKR is consistently lower than what all the other platforms show.
This is believed to be a function of their date math. Same-day expirations are the most understated while longer durations are negligibly impacted.
Due to the opacity of the calculations, it’s unclear who has the “correct” calculation. However, the fact that all the other platforms agree and IBKR is the outlier causes us to reject the IBKR value.
My understanding is that people use another data provider and base their trades on that data. Eg: have ToS open in parallel and select their strikes accordingly.
True. I use my own IV calculations for exactly that reason!
How do you model the magnitude of the skew / smile?
You don’t need to model skew to find IV, just plug in price and other given BS inputs get the implied vol. It can be done in excel or one of the many online calculators
But once you calculate for IV, you can find skew easily as well.
But we don’t know what model BigERN is using. He could be using BSM. He could be using quadratic splines. Or he could be winging it 😀
Most brokers use some kind of [quadratic] spline model in practice (maybe IBKR isn’t and that’s why their stuff is different). BSM is the academic foundation it’s all based on for ATM European options, sure. Pricing a far OTM position with that model alone yields silly results without applying a function for skew.
Consider researcher Corey Hoffstein’s work exploring using vega as a vehicle for downside protection as opposed to delta (https://blog.thinknewfound.com/2020/06/tail-hedging/).
He uses theoretical pricing via “a quadratic curve to log-moneyness and implied total variance for each quoted maturity.” That’s a fancy way of saying: “uses a model to generate theoretical pricing.”
I’ve independently replicated his work at the bequest of a client using empirical data (https://spintwig.com/spy-long-put-90-dte-10-and-30-otm-tail-hedging/) and achieved similar results; Corey’s pricing model is sufficient.
Curious what BigERN’s model consists of.
You’re confusing two different things, modeling the volatility smile (quartic or some other curve shape) and what to do with that volatility smile (plug it into Black Scholes or some other model).
If you’re using black scholes to find the implied volatility of an OTM option that you know the price of the ATM volatility is irrelevant.
When you look at the IV’s on IB or some other platform, they are taking prices (like mid price) converting them to IV’s (using some model like BS) then finally fitting a curve through it using something like a spline.
I think we’re saying the same thing.
Question remains: I’m curious what BigERN’s methodology is.
He’s said it before, using the option excel calculations from the whaley book excel CD, but if you’re just looking for IV that’s easy to find without the CD.
My point is you don’t need to model skew to find the IV of a single OTM option if you have it’s price.
Skew modeling is something that people look after they have the IV.
I’m asking specifically about how he models skew, hence my question “how do you model skew?” 🙂
I’d be willing to wager a freshly-minted $.25 coin that the answer is not in the Whaley CD.
Modeling Skew is interesting for other reasons but not necessary for finding IV or deltas in this case
Very low-tech. You guys are overthinking this! 🙂
That’s my approach. I look at the IB Deltas (knowing that they are a little bit too low) and the premium and just sell following my gut feeling.
Haha, winging it with the Whaley book.
Also keep in mind that I don’t need to compute or model any smile. I just sell puts. I target a specific premium and off we go. And for the contracts I sell I also compute the IV, Delta, etc.
I use the Whaley Excel Add-In. It’s not so much an issue with the formula itself but IB might be using different inputs and that messes with the results.
I come up with my own formula on how to deal with intra-day trades. I also make an adjustment for options held over a weekend.
Any chance you can share the secret sauce?
e.g. do you consider trading days only or include the weekend as well?
I will in a future post!
TL/DR: Market open is +0.5x trading day, linearly decaying to the market close.
Really looking forward to that post!
Yeah I bet the biggest difference in IV’s among short dated OTM options on different platforms is what time to expiration they are using. Some platforms may use fractional days, some may use whole days, some may use actual/365 day calendars, some may use 252 trading day calendars, etc.
This strategy’s risk-adjusted return was so good!!
I made some adjustment:
Mon: sell put expire at Wed , Delta around 4 (or at least 2.5% OTM)
Tue: sell put expire at Fri , Delta around 3.5 (or at least 3.5% OTM)
Wed: sell put expire at Fri , Delta around 4 (or at least 2.5% OTM)
Thu: sell put expire at next Mon , Delta around 3.75 (or at least 2.5% OTM)
Fri: sell put expire at next Wed , Delta around 3.5 (or at least 3.5% OTM)
So I have options expired at Mon:Wed:Fri rotio= 1:2:2
I can open options everyday to lower bias
But I have longer DTE
Never get hit for months
According to your experience
What are the advantages and disadvantages?? Will it easily blow up?
That sounds like a good plan. I’d caution about this rules-based put selling, where you pile on more delta on Tue & Thu when you don’t really know yet what happened to your Wed & Fri options. If there’s a big enough blow up on a Wed or Fir you might suffer some unexpectedly large losses. I’d condition the Tue & Thu trades on how much existing delta you have.
Couldn’t this all be avoided by just using a European style option?
SPX options traded on the CBOE are European Options. Always keep in mind that even though the options can’t be exercised before the 1-year expiration, the price will still move and you have to keep your margins along the way!