January 14, 2025 – Happy New Year, everybody! I hope you had a quiet, relaxing Christmas season and a great start to the New Year. As I’ve done in prior years, I want to update you on my options trading strategy: How was the performance in CY 2024? Are there any strategy changes? How did I deal with the volatility in August and December? I also want to share some general thoughts and observations to rationalize the long-term profitability of my options strategy.
Let’s get started…
Options Trading Strategy Details
I didn’t implement any new features, but for readers who are new to this topic, here is a quick recap of what I’m doing:
- Every trading day, I write (=short) 1DTE CBOE puts on the SPX index. This has been my bread-and-butter strategy for the last 10+ years because it accounts for most of my trading profits.
- In addition to the overnight puts, I monitor the market open and write additional SPX index 0DTE puts, which expire on the same day. I would usually take a break from the 0DTE puts if the market opens down significantly, where my overnight puts are in danger of losses, and I don’t want to add even more downside risk.
- I write 0DTE calls on the SPX index. I hinted at this innovation during the 2023 update, and now I have a full calendar year of returns, and the results look pretty promising. I also write 1DTE calls very, very occasionally. But I would only do so on Thursdays before the third Friday of the month when I can write calls for the AM closing.
- Occasionally, I write longer-dated vertical put spreads, usually 30 to 180 days to expiration. I normally do so after a significant drop in the index when implied volatility spikes. They are often wildly OTM, i.e., sell a 1600 put and buy a 600 put. I try to generate a premium around $1.00-$2.00.
Discontinued features: I had mentioned in earlier posts and comments that I was planning to hedge the downside with some long VIX exposure. I bought a few VIX Call options about 6-8 months out, usually at a 24 to 25 strike. I usually did so when markets were calm and the premium looked cheap. But alas, that didn’t do it for me. Even during the market blowup on August 5, only the August contract VIX call saw a spike attractive enough to call this a hedge. The VIX term structure went into extreme backwardation, so the impact of the vol spike on the, say, December 2024 VIX call was tiny. To hedge against a vol spike, I would have to sell such a massive number of VIX calls that I would “throw away the baby with the bathwater.”
Some stats on the contracts I traded in 2024
People frequently ask me what exact options I sell every day. It varies daily, and I target no fixed percentage or index point number above or below the index level when picking put and call strikes. But here are some stats for my fellow options traders. I use the same chart format as in earlier years:
- The median overnight puts premium was 10 cents. The average was 16.3 cents. The range was $0.05 to $4.60. That’s not a typo; after the 3% intra-day drop on December 18, following the FOMC release, I sold a few “haymaker” puts for this awesome premium and about 6% out-of-the-money!
- The same-day puts and calls had mostly 10 cents premium.
- The same goes for Calls: most had a 10-cent premium, though I also sold a few $0.05 contracts earlier in the year. I’ve now moved to almost exclusively $0.10 calls.
- The mean deltas were 0.0043 for 1DTE puts, 0.0068 for 0DTE puts, and 0.009 for the calls. And I know the puts have negative deltas, but I plot the absolute values here for easier comparison.
How far out-of-the-money, and how much implied volatility (IV) do I get for my contracts? I plot those in the next set of charts:
- I wrote overnight puts with strikes around 5.3% (median) or 5.7% (mean) away from the current index. The range was 1.9% to 23.5% (not a typo; that was after the August 5 market shock!)
- The 0DTE puts were about 2.1-2.3% OTM, while the 0DTE Calls were about 1.4-1.5% OTM
- The median implied volatility (IV) for the 0DTE puts was 33.4%. Positive skewness pushes the means a bit higher to 36.1%. Occasionally, I sell puts with a 100%+ implied volatility, especially during the stress periods in early August and late December. The maximum IV for the 1DTE puts was 191.5%!
- For the 0DTE contracts, the implied vol was in the low 20s for the puts and around 15 for the calls.
When do I trade my contracts? And how does my implied volatility compare to the then-prevailing VIX index?
- No surprise here: I trade most of the 1 DTE puts right around closing time. But occasionally, I already “double-dip” and sell the next-day puts during the morning hours.
- I sell most of the 0DTE Puts and Calls at the open. But sometimes, I trade a few more contracts, usually around 10:00 and 12:00. The shortest time to expiration for a call was only 2 minutes and 9 seconds.
- I sell most of my overnight puts trade at an IV, usually 2x the CBOE VIX. The same-day puts about 1.5x and the Calls at 1x VIX. That’s about what you’d expect due to the negative skewness. Specifically, large upside moves are less likely than large downside moves. And to be sure, I know the 30-day VIX is not an ideal comparison for the 0-day and 1-day vol selling. But I still love that chart!
0DTE/1DTE Options Trading Performance in 2024
My 0DTE and 1DTE put strategy returned just under $100,000 in 2024. Notice that these returns are in addition to my portfolio returns. So, I make the gains, dividends, and interest in my Stock/Bond/Preferred Share portfolio, and the options trading supplements that. In the chart below, I plot the cumulative daily gross option selling revenue (black), the net profit (green), and the losses (red). Loss is the difference between the actual and maximum gain. The net loss is, of course, a little bit less because you made the premium. But I like to track the gross losses, i.e., everything less than making the entire option premium. As we can see from the chart, the 19 loss events (when the red line moves a step up) are spread out pretty evenly throughout the year. Every calendar month, I had at least one loss. April and December had three. The longest stretch of no losses was between the August 1 and September 30 losses.
Here are some more detailed stats; see the table below:
- I traded just over 11,000 contracts. Notice that these are not 11k individual trades. One trade is anywhere from 1 to 25 contracts.
- $124k+ total premium collected, net profit $95,861.10, which is 77.3% of the gross revenue. I call this the PCR = Premium Capture Rate. It’s not the best PCR ever, but certainly nothing to sneeze at either.
- The overnight puts are my bread-and-butter business, accounting for almost half of the gross revenue and two-thirds of the earnings. The minor “loss” was due to a small trade error: I sold two puts too many and bought them back immediately. The 2-way commissions plus the $30 in premium not earned caused a $35 gross loss, and the net loss was about $5.00. But everything else worked out just fine.
- I also traded a few overnight calls, but only occasionally, when the expiration was in the AM, i.e., on the third Friday of the month: $650+. Not bad.
- The 0DTE puts performed poorly this year, with only a 32.2% PCR. Although this figure looked better all year, the December losses dragged it down.
- The 0DTE calls compensated for the poor 0DTE put performance: they earned me an extra $21.5k with a PCR of 61.5%.
- Also, isn’t that intriguing!? The S&P 500 was up 23% (price index) and 25% (total return index), and my 0DTE short calls (i.e., selling the upside risk) did better than the 0DTE short puts. In any given year, the general direction of the stock market predicts very little about the profitability of the 0DTE put vs. call selling!
How to deal with volatility: August 5
Monday, August 5, 2024, was a day to remember in the short-volatility world. It reminded me of the February 5, 2018 meltdown. On August 2, when the S&P 500 closed at 5346.56, I had sold puts with strikes between 4800 and 4975. Between 7% and 10% out-of-the-money felt safe at the time. The premiums were also quite impressive, between $0.40 and $1.15. But on Monday, the market opened at 5151.14, already down almost 200 points or about 3.66%. My short puts had rushed well past the max price I usually set before throwing in the towel, even though their strikes were another 175 to 350 points from the index level.
So, I consciously decided that the market was likely overreacting to the Bank of Japan’s interest rate move and decided to do nothing. My instinct was that it didn’t make sense that a surprise central bank move in Japan would derail the US economy and financial markets. It ended up being the right thing to do. The index bounced around in the 5,100s but ended at 5,186, far away from my strikes. I made the entire premium, even though intra-day, I had wiped out about four to five months of options-selling income.
How to deal with volatility: December 18-23
December 18 is probably still fresh in everyone’s memory. The FOMC met that day and announced its interest rate decision. And rocked the boat by throwing cold water on investors’ rate-cut optimism. Within the last two hours of trading on December 18, the S&P 500 fell by 3%. And it knocked out my stop-loss orders. Even worse, I got a two-for-one loss that day because, after the first stop loss was executed, I sold more puts, further out of the money, only to get stopped again on those as well.
Next, I had a loss on my 0DTE calls on December 20 and another on the 0DTE put side on December 23. What saved my December performance was the fact that the 1DTE puts made so much money around that time to compensate for those losses. Notice how the black line shoots up during that time in the cumulative return chart above. Well, you win some, you lose some.
Longer-term put spreads
Despite my stern warning in an earlier post, I occasionally sell longer-dated vertical put spreads (some call them “credit spreads”). I always target 30 to 180 days out. Specifically, I’d sell one put far out of the money and buy back one even farther OTM with the same expiration date. The long put helps with margin management. I target a premium of at least $1.00 and typically sell such spreads when we go through a volatility spike. I plot the cumulative returns below.
I made a total of about $9,700 extra from those trades. Not bad! But this portion of put selling was definitely a bumpy ride! August 5 sent me down a cool $7k, but the vertical spreads quickly recovered. In fact, on August 5, in response to the big vol spike, I sold some additional spreads with net premiums between $3.35 and $6.25 for December and October, with strikes for the short put between 1600 and 2000 and 600 and 1000 for the long put. It didn’t look very likely that a central bank surprise halfway around the world would take the US stock market down by 3000+ points or 62% in just a few months. So, I’ve now gotten into the groove of selling between $1,000 and $1,500 worth of put spread premium expiring monthly. It’s small potatoes compared to my other options trading activities, but it takes very little time, and every dollar matters.
The overall portfolio in 2024
How did the rest of the portfolio perform? First, let me display the CY 2024 return stats; please see the table below. I report stats for bonds (10-year Treasury benchmark), stocks (S&P 500 total return), and a 40/30/30 benchmark, i.e., 40% stocks, 30% bonds, and 30% cash (i.e., short-term safe assets, like T-bills). Why 40/30/30? My taxable options trading portfolio at Interactive Brokers is about half of my financial assets, while the other half, all in tax-advantaged accounts, is 100% equities. Since I like to have a weighted overall allocation of 70% equities, 15% bonds, and 15% cash to hedge against Sequence Risk, I’d need to target 40% stocks, 30% bonds, and 30% Tbills portfolio in my taxable account.
The put selling contributed 4.8% to the annual return. 18.3% came from the other part (equity mutual funds, preferred shares, bonds, cash). The total return was 23.9%. That may be a bit smaller than a 100% equity portfolio (+25%). But that’s an inappropriate comparison because I’d never want to be 100% equities in retirement. Compared to the 10.5% I could have gotten from the 40/30/30 benchmark, my 23.9% looks spectacular.
Also, notice the impressive put writing risk vs. return stats: The annualized risk was only 0.4%, and the Information Ratio was 10.62, which is fantastic. Also, notice that the monthly put writing returns actually had positive(!) skewness, negligible betas with both bonds and stocks, and extremely low correlations to the other returns. Of course, over longer horizons, say the last 10 years, my options trading IR was a bit more modest. “Only” 3.0, but that’s still very fantastic. There aren’t too many hedge funds with an IR/Sharpe Ratio of 3!
Interesting Research on Options Trading Strategies
The idea of selling insurance for profit is neither my invention nor new. If you don’t believe some random guy on the internet like me, I want to provide some additional reading material to support the philosophy underlying my strategy.
AQR’s Paper “Covered Calls Uncovered”
In 2015, there was a great article aptly titled “Covered Calls Uncovered” in the Financial Analyst Journal (FAJ), the in-house journal of the CFA Institute. Two brilliant researchers at the famed hedge fund AQR wrote it: Roni Israelov & Lars N. Nielsen (2015) Covered Calls Uncovered, Financial Analysts Journal, 71:6, 44-57. Available online: https://doi.org/10.2469/faj.v71.n6.1.
The article is free to CFA Institute members like myself but likely behind a paywall for everyone else. But I like to summarize it quickly: The authors first simulate the returns of a covered-call strategy with at-the-money strikes. It’s not even close to my strategy in at least two dimensions (strike and days to expiration), but bear with me. The authors then split the covered call returns into different components:
- A constant 0.5 Delta Equity allocation. (=Passive Equity) Why 0.5 Delta? At inception, the covered call has a net 0.5 Delta, +1.0 from the equity portion, and -0.5 from the short at-the-money call.
- A Delta-hedged short-call position. The delta-hedging is daily. Notice that the short call option had a Delta=-0.50 at inception, so we would pair this with the remaining 0.5 equity allocation. But over time, that Call delta might change. Because we want to distill the pure short-volatility component orthogonal to any equity beta, we adjust the equity exposure to make the net equity exposure precisely zero. (=Pure Short Volatility)
- The difference between the initial call Delta and the dynamic Delta-hedging in Part 2. Notice that component 3 picks up the “slack” from component 2 and will guarantee that the sum of the three components will always be the simple covered call strategy, i.e., 100% equities minus a Call option. (=Equity Timing)
Why is there (implicit) market timing in a covered-call strategy? A covered call position will have varying deltas on its way to expiration. And the deltas look like a “valuation” strategy (not to be confused with value investing, the topic of last month’s post). For example, if the index falls, your covered call’s net Delta rises. So, you’re essentially betting on mean reversion, i.e., after the fall, you throw in more equity exposure and hope for a recovery. Likewise, if the index rises, your delta declines, so you take less equity risk. Again, this looks like a valuation or mean-reversion strategy.
In any case, the authors calculate the return stats of the covered call strategy and its three components. I have included a screenshot below:
Analysts Journal, 71:6, 44-57. Fair Use permission.
Some observations:
- The covered call strategy has a slightly higher Sharpe Ratio than the passive equity strategy. Notice that the Passive Equity strategy is only a 0.5x equity exposure, so the excess return of a passive equity strategy would have been around 7.0% (simple) and 6.4% (geometric), beating the covered call strategy. However, 100% equity would also have a much higher volatility (17% vs. 11.4%).
- Short Volatility had an impressive Sharpe Ratio of 0.98. You get a 1.9% extra return for only 1.9% risk! Also, people always wonder how statistically significant an alpha is. If we want to test the Null Hypothesis (H0) that the short-vol strategy doesn’t add any alpha, we can calculate the t-stat as IR times the square root of the years. In this case 0.98*sqrt(18.75)=4.25. We can easily reject the H0 of zero alpha (p-value = 0.00001)
- Equity timing adds a lot of volatility with very little return. The IR is only 0.1, so the slight outperformance is not even statistically significant. (The t-stat would be only 0.43, with a p-value of 0.433, so we cannot reject the H0 at any reasonable significance level.)
The takeaways from this article are threefold. First, the pure and distilled put selling part, orthogonal to any equity beta and equity timing, has a desirable risk vs. return profile. A Sharpe Ratio (actually, Information Ratio (IR) would have been the better terminology in this context) of just under 1.0 is quite impressive. Recall that this 1996-2014 simulation includes two of the worst bear markets in recent history, with a 49% drawdown in 2000-2002 and 56% in 2007-2009!
And importantly, this is only the most basic options strategy, i.e., plain-Jane monthly covered call selling with at-the-money strikes. With my more sophisticated approach and selling the even more attractive options, i.e., deeply OTM puts and calls, I have generated an IR of about 3.0 over the years. So, keep a portfolio of productive assets (stocks, bonds, preferred shares, etc.) and then add the pure option-selling/short-volatility portion only, but not the noisy and risky and almost return-free market-timing portion. Covered call writing, cash-secured puts, etc., are for the unanointed masses and YouTube Fin-fluencers. Savvy investors should do what I do!
Second, to get the best risk vs. return profile, it’s recommended to set the DTE of your options as low as possible, tantamount to the Delta-hedging in the AQR paper. In other words, you never want any day-to-day varying delta exposure from your short options. Instead, reset your bets every day and start anew. And just to be sure, I don’t need to do any additional Delta hedging because a) I sell my puts so far out of the money that they have negligible Deltas at inception, usually below 0.01, and b) they all expire after one day anyway.
Third, the article provides the perfect rationale against attempting to make back your money after a put-selling loss, which is often called the “Wheel strategy.” See my post from last year: Part 12 – Why the Wheel Strategy Doesn’t Work. You don’t want to mix options selling and market timing, and the AQR paper has the receipts: the market timing part is very volatile and has low or even zero expected return. You’d ruin your Sharpe Ratio when chasing after past losses. After 13+ years of experience trading this strategy, I recommend that a loss is water under the bridge; it’s best to reset your bet every day and try again.
Robert Litterman’s “Who Should Hedge Tail Risk?”
The second link I want to share is a great paper by Robert Litterman, also published in the FAJ: Litterman, R. (2011). Who Should Hedge Tail Risk? Financial Analysts Journal, 67(3), 6–11. https://doi.org/10.2469/faj.v67.n3.7. It, too, is likely behind a paywall. However, the almost identical content is also available on YouTube; please check out Robert Litterman’s video here. If you don’t know, Robert Litterman, a former partner at Goldman Sachs, is the guy from the famous Black-Litterman model. And yes, the “Black” in Black-Litterman is Fischer Black from the Black-Scholes options formula. Small World! And a fun fact and bragging right: I know Dr. Litterman. We’re both University of Minnesota Economics PhDs:
Anyhoo, Litterman makes the excellent point that hedging against tail events is often prohibitively expensive. Indeed, it is so costly that instead of hedging against tail risk, it can be preferable to simply reduce exposure to this risk factor and then sell the tail risk insurance for an extra return.
He presents an example of an investor with a 100% equity portfolio (=portfolio 1). The 2005-2011 window would have been a frustrating experience: Only 0.9% annualized returns but 24.5% annualized volatility; see the table below. Hedging the tail risk with a 10% allocation to a long-VIX ETF (ticker VXX) as in portfolio #2 would have reduced volatility, but you also gave up 1.55 percentage points in return. You would have done better simply scaling down the risky equity position to 75% and holding 25% T-bills as in portfolio #3: Your volatility is lower, and your average return would have been higher than any of the #1 or #2 portfolios. Finally, reducing your equity position to 50%, shorting(!) the VIX ETF, and holding the rest in T-bills would have been the best option (portfolio #4) in terms of both return and risk.
What I’m doing in my taxable account is precisely what Robert Litterman proposed! I can’t hold 100% equities because that’s too much tail risk, especially in retirement when you have the added headache of Sequence Risk. One could do a 75/25 portfolio, but that’s still not ideal. We should do 50% equities and add the short-vol as another return-generating engine while keeping the rest in lower-risk, diversifying assets. I don’t think the 50/10/40 split is gospel, but the direction toward less equity beta and more volatility selling is undoubtedly optimal.
Robert Litterman also had another great nugget. For example, he lists this question to be considered before buying tail-risk insurance:
“Is the governance structure […] leading management to consider buying tail-risk
insurance […] to protect itself, which may not be in the long-term best interest of the fund’s beneficiaries?”
What a great point! Part of the tail-risk insurance demand is likely due to people working at and running institutional investor portfolios, such as pension funds, endowments, sovereign wealth funds, etc. The managers likely buy insurance to protect their interests, often to the detriment of the funds’ ultimate owners and beneficiaries. This leads to market inefficiency due to a principal-agent problem. As long as this inefficiency exists, it is payday for us options traders! So, the profitability of selling tail-risk insurance is not just a statistical fluke. Instead, there are some fundamental economic reasons to justify this rich options trading premium.
As a side note, I understand my options trading differs from the VIX selling Litterman proposes. But I claim that options selling is an even better route to selling volatility. If you remember, I always warned against shorting VIX because it seemed to have too much tail risk for my taste and an unattractive risk vs. return tradeoff; see my 2017 blog post on that topic. So, Litterman’s simulations likely underestimate the benefit of selling tail risk.
Podcast appearances
Finally, I want to give a shoutout to the podcasters who waded into the options trading topic and were generous enough to feature me in 2024:
- Jason at Two Sides of FIRE had me on the program to chat about my options trading strategy.
- Jason also featured David Sun, one of my option-trading buddies. I also recommend checking that one out—it has lots of great information from a guy who runs an options-trading hedge fund. I’ve learned a lot from David! David also has his own podcast, “The Trade Busters.”
- David Baughier at Forget About Money interviewed Brad Finn and me about our respective options trading experiences. The interview is an excellent introduction to options trading if you’re new to the subject.
Conclusion
OK, we’re at 4,500 words; it’s time to wrap up. 2024 was a terrific year for options trading, both in dollar and percentage terms. I sold the 1DTE puts far enough out-of-the-money that all puts made their full profit. My same-day puts had a bumpier ride, but I’m glad I had the supplemental income, as I am now taking less risk with my 1DTE contracts. While the percentage return dwarfs compared to the 25% total return in the S&P 500, it’s nice to have that additional return on top of my other portfolio.
How was your options trading experience in 2024? I look forward to your comments and suggestions below!
Please check out the Options Trading Landing Page for other parts of this series.
Title Picture Credit: WordPress AI
If there is a 30% move down in the market and you have 3.4x leverage, since you don’t have an overnight stop, doesn’t that mean there is a risk of the portfolio going to zero or negative due to overnight gaps? (30*3.4 = 102). Although I have used a small amount of leverage in many years ago, I always felt uncomfortable with that risk and others like those associated with rehypothecation, so I switched back to cash accounts.
No. I don’t sell my 1DTE puts at the money. I sell between 5 and 20% OTM.
Also, the market has never fallen 30% overnight. Not even on 10/19/1987. Even the 21% drop on that day happened during the day, not overnight.
Thanks for your response Karsten. Makes sense re: the OTM reducing the potential down-draw in scenarios where the market declines even with the leverage. I am aware that the market has never fallen 30% overnight, however, the risk remains nonzero.
More practically speaking if using such a leveraged strategy, I’d be worried about outcomes like the market falls 20% or 30% overnight due to outbreak of war or a meteor strike or other catastrophe, or perhaps a similar decline happens over several days. Meanwhile, I’m imagining financial markets might be closed, or perhaps some rehypothecated shares (that one agreed to rehypothecate as part of the IB margin account contract) cannot easily be returned to the account from wherever they ended up in the financial system as some firms with problems suddenly find they want to argue about their contractual obligations, and margin standards are suddenly tightened. I personally do value investing and in those scenarios after figuring out safety aspects of my life then investing would likely be better than normal whenever the market reopens because of volatility. But if I had a leveraged margin account I could imagine there might be all sorts of weird scenarios where the account could be stuck in some bad state.
Of course value investing also has its associated risks, particularly, picking a bunch of companies that all go to zero. But it works for me. It may also just be a personality thing since I don’t want to trade every day. And it may also be a philosophical thing since I don’t regard the expected value of options as equal to the output of option pricing models. Thanks for sharing your trading strategy. Cheers, C
There have been multiple wars that started, i.e., Ukraine (both 2014 and 2022) and Israel/Gaza (2023). Nothing bad happened in my leveraged portfolio.
Of course, the USA could be drawn into a war, but I don’t think that’s very likely. I don’t think WoKanada or Mexico will invade us anytime soon.
maybe asteroid hit or alien invasion 🙂 you never know.
Personally I use spreads so even 100% loss won’t wipe out my account AT ALL. For 5-10% return, no reason to do naked or something similar, at least for me.
Asteroids will be visible for years before the impact. Aliens, too, I hope. 🙂
What’s your ballpark spread width and how much leverage do you use?
Aliens can be descended out of thin air I assume, guess they won’t “warn” us, :). Sometimes I shudder to think what would happen to stock market if they really suddenly appear on earth. Now we have circuit breaker etc but still can’t imagine volatility in that case. So probably too paranoia but with naked puts, I really cannot sleep :).
Last August, I’ve seen some of people in the group that I follow wiped out completely (some of them had 500k to 1million, god forbid!!! ). Granted, they levered too much with naked components, but everyone knows they shouldn’t over leverage but it always happens, always, as you know because last April, 1-1-1 or 1-1-2 traders are wiped out again !!!
My width is 25 on SPX (short vertical spread), delta about 10 to15. Going about 150 to 250DTE (move slowly and easy to adjust). Usually 200-300 premium per spread. Planned capital per one spread is 5k to 10K (use 50% of total planned capital max – if alien invades, I would lose close to 50% maybe but not wiped out).
I have been doing this since the end of 2021. Drawdown was long in the middle in 2022 but by the end of December 2022, I was up significantly, approximately 12%. Then around 10% last two years.
This year has been tough because my size got too big (aimed 15-20%) so the drawdown was big and rythme totally messed up, and am still recovering since April, unfortunately. At least I didn’t lever with the nakeds, so I guess I survived, unlike other people!!
Yeah, aliens can come suddenly. But I would hope they announce themselves, like in the movies (e.g., Independence Day) so we have a few days warnings before they attack.
Your strategy: It will blow up when the conditions are right. I suspect it would have failed in 2020. Failure as in -50% as calibrate it, so you would need several years to dig out of it, i.e., about 7 years at 10% p.a. return. My strategy made money in 2020 and ever since. But still, good luck!
Well, I backtested 2020, it was pretty good. Size is what really matters, just like your strategy (or any other). Annual 5-10%, I believe, can be definitely achieved by defined risk strategy as well. Anyways, good luck to you too, and take care.
Yep, sizing is crucial. I’ve been through a lot of vol events, so I’ve finetuned that part. Cheers!
Who needs aliens when we have AI? 😉
Very funny! Though, I think even if that’s true, the impact of AI will come slowly over time, and not in the 6 hours 30 minutes holding time of one of my short put 0DTEs.
But I think AI will eventually a force for good and higher productivity.
Moody’s downgrade after-hours Fri isn’t a nice way to start the weekend. S&P ETF -1% in after-hours trading.
ERN, any perspective on whether this could get ugly on Sun night/Mon for put sellers?
The 2023 Fitch downgrade seemed relatively mild, S&P down 1.3% or so in the next trading session. VIX went from 14 the prior close to 15-16 the day of.
I don’t see this as a big deal. 1% already, maybe another 1% by the open on Monday. I sold my puts with strikes much farther Out of the Money.
Saw this yesterday, thanks for sharing.
I ended up closing 5/19 5700 puts for a small loss Sun evening. Jan 27 memories are too fresh. Though they traded as high as 6x original premium before everything calmed down early Mon. Looks like VIX calmed down significantly from pre-market to open, and ES never dipped much more after Sun open. Should have held onto them!
Unrelated, I need to re-enter email, name, etc. info on every comment. Is there a different way to go about that? “Save name, email and website” doesn’t seem to work.
At 6x, that wasn’t a large loss, so no worries.
I’m bad at tech support. Maybe check the settings of your browser to allow for auto-fill.
I’ve followed your options strategy for the past year and I’m a firm believer in it. I was originally doing it without Stop orders and I’ve learned my lesson that I need Stop orders to cap losses. But that has me fearing overnight trades and AM-expiry trades the 3rd Thursday of each month. I feel a lack of control over the cap of my losses on those trades as I don’t believe Stop orders protect you overnight or throughout the morning on AM-expiries.
I’m wondering if you’d try to convince me my fear is unwarranted. I’ve moved to just trading 0DTE puts and calls. Obviously you favor getting the ~95% premium capture rate of your 1DTE trades. And I have seen your excellent analysis that almost always throughout history the delta/premium at close reflects the volatility experienced the next day so you sell puts far enough out of the money. Still, I have this fear that “I don’t need risk I can’t control on leverage.” I don’t believe there’s a good option to cap my risk for puts overnight?
I’m FI going through One More Year syndrome and don’t want a black swan event to take me out of FI. For someone in my situation looking to cap risk, do you think trading exclusively 0DTE puts/calls with a 20x Stop is a reasonable approach or would you suggest something else?
There is the risk of a big gap down over night. I occasionally monitor the futures and the put prices at night. I think I’m safe from a total disaster that way. It worked so far with all this volatility over the last few years.
But if you prefer 0DTE only, that’s fine, too. It will make you money and you sleep better.
Another option would be to trade maybe 60-120 days out, which I usually do in response to a vol spike. Sell some puts 60%+ OTM for a $0.75-$1.25 premium and keep a staggered set of puts. Some overnight vol may cause some small losses, but they are only temporary. I don’t think a a trade war will sink the stock market by that much.
Thank you for your kindness in fielding questions. I do have another one on the ever popular stops question. There have been follow up questions on 0DTE = ~2-3 weeks of premium for stops and ~3mos for 1DTE. Just to keep the math simple and to help me understand how things “add up” let’s assume the following:
2×5 day weeks = 10 Days
10 x 0DTE sold daily for 10 days = 100 contracts sold
$10 per contract x 100 contracts = $1000 premium collected
It’s Monday and time to sell 10 new 0DTEs and set stops and I’m just going to sell 10 contracts all on the same strike. Do I:
1) Set the stop to equal a $1000 total loss (e.g. 10 contracts set to stop out at 100 dollars loss per contract)
2) Set the stop to equal a $10000 total loss (e.g. 10 contracts set to stop out at 1000 dollars loss per contract)
I assume option #1, buth thought I would verify.
You sell at premium P and set the STP to 10 to about 15 times P. You tie the STP to the price, then it doesn’t matter how many you sell. This lloks more like option 1. Not sure how you got the $10,000 total loss in option 2.
Thank you…and your first sentence was way more precise and clear than what I wrote! Yep, option 1 as I described it.
I’m very fascinated by this strategy. I’m a long time Boglehead follower and curious to hear your thoughts on:
1) Bogleheads usually prefer rules based plans (i.e put $X in market every 2 weeks from paychecks in a 3 fund portfolio in 60/30/10 split). Do you think your strategy could be better with rigid rules (i.e every day sell X put with strike Y based on VIX/delta at time T) or theres an art to what you do that can’t be captured algorithmically? I read you don’t do the same thing every day but would it better to be more rigid in your approach?
2) Historically, what options are the most mispriced (where implied vol >> realized vol)? Is it always kind of in the same place OTM (when considering VIX/delta/premiums, etc) relative to ATM or it varies?
3) Kind of scared of a big big crash wiping out everything. Remember what happened to XIV a few years back? Is there any chance of a bankruptcy? I’m thinking from low to super high VIX in the same day with a big crash.
4) Do you need portfolio margin? I’m trying with a magin enabled Robinhood account and a IBKR paper account and they have pretty harsh margin requirements. I can’t leverage up much and am lucky to be able to sell one contract.
5) Why do you use SPX? Would this work well with SPY or any other stock/ETF?
For 4) and 5)
4)You indeed need it – that is, portfolio margin and a nice, long, stable portfolio underneath. Otherwise, if you just use Reg-T rules, your margin will be calculated for each position alone and you’ll end up with very harsh requirements.
5) You need European, cash settled options. This way you can ignore assignment risk, pin risk, and most likely a few tax headaches in most tax residencies. You’ll also want to do this on indexes, as the probability of seeing a “too many sigmas” move against you are lower (the tails are not too fat in well-know indexes). This makes SPX the ideal candidate.
Good comment!
About item 5: SPY is also on the index. But since this is on an ETF, you still have the tax and record keeping headaches. So, it’s essential to do this on the SPX index. Options on the ES e-mini might work, too, but have the headache of physical delivery of the futures.
1: this strategy doesn’t work with rigid rules. Even if I had a rigid rule I wouldn’t publish it here because I don’t my perticular trades to get too crowded and/or other front-running what I’m doing.
2: There is no rigid formula for that. It changes every day.
3: I wrote about XIV here: https://earlyretirementnow.com/2017/10/25/returned-over-100-percent-year-to-date-still-not-buying-it/
I don’t think that naked put options face the same kind of skewness risk
4: Portfolio margin is very helpful. I would even call it essential.
5: SPX has the preferential tax treatment (S.1256). Lower taxes and less record keeping requirement, i.e., you don’t need to itemize your trades during tax season.
Without a rules based strategy, is there a risk of your own emotions or market timing affecting the results?
I still follow certain rules: Always use STPs. Always trade at least some contract. Always look for the best premium to risk tradeoff.
Researching this, I’m finding that the margin requirement on one 1DTE SPX put 20% OTM would be about $60,000. That seems exceedingly high to get $10 premium (ignoring transaction costs).
Formula 1: 20% of the underlying index value, minus the out-of-the-money (OTM) amount, plus the premium.
Formula 2: 15% of the strike price, plus the premium. (Some sources mention 10% of the strike price.)
Formula 3 (Minimum): $2.50 per contract
Formula 2 is the greatest with SPX at 5,000, 15% * 4000 [20% OTM] + $0.10 = $600.10 * 100 shares.
With a stop-loss of 20x the $0.10 premium, the potential loss is $4k, which is tiny compared to the $60k I’m calculating. Do brokerages take this into account?
Back to the $60k, with a $1.2M portfolio, one could sell 20 contracts, collecting $200/day in premiums before fees.
Does this seem ballpark correct?
I don’t use a fixed % number for the OTM target. It depends on the volatility environment.
Either way, the exchange will demand $60k margin because if you forget to set the STP and he trade goes sour they’d be on the hook for a loss that exceeds your capital.
Finally, yes, if you max out your margin that way, you could make about $200/day gross. If you also sell on the call side, maybe double that.
If the puts you sold are almost worthless due to a market move up, would it ever make sense to just close them and sell a new batch to capture the premium again? This would also let you get more independent samples during a year, helping the central limit theorem case.
Sometimes that’s a good idea. But I usually let them expire and collect even the last remaining $0.05 and I save the commission as well.
my broker requires 100% margin for selling puts. to reduce this, what do you think about doing a put credit spread with the higher leg at the usual premium say $0.30 and buying lower leg at some really low price like $0.05 to reduce margin requirements? then I can do several of those and still be within margin requirements
Spreads might work. I would change my broker. Remember: you can move your assets without tax consequences through ACATS.
Have you tried using a backtester to replicate your results? Option Omega and Option Alpha are platforms I’m looking at. Here is a 3 year backtest with next day expiration on SPX, entry point at 345pm, a short put spread (they don’t allow short puts only), stop loss -500%: https://postimg.cc/zVDRtpNx (parameters are on the right hand side). The 2024 data doesn’t look at smooth. Tried some different parameters like using various deltas or credit amounts as the entry point and they’ve never been super smooth.
I’ve played around with options omega with a free trial. I still prefer my method. My 2024 ride was definitely much smoother.
Those back testers can allow you to modify parameters in a way that closely resembles what you did such as daily selling of 0DTE SPX puts with a 0.2 credit. Have you been able to replicate your results that way?
In my limited testing, I find that it’s not so easy replicating your results. I would have expected that general daily SPX put selling would do well since implied vol >> realized vol and that it’s not so sensitive to the parameters in the back test.
You can’t replicate my results because I don’t sell $0.20 every day. Look at the range of premium revenue in the bar charts. The premium ranged from $0.05 to over $4.00.
Hi ERN! I have been enjoying this blog on and off for years, thank you! Regarding having to manually monitor and stop out positions out of regular trading hours, in IBKR there is an option to check a box to attempt to fill stop limit orders outside RTH. Any thoughts on the pros and cons of using this to automatically stop out positions overnight?
The Cons is that you have very illiquid markets over night. I would not trust that STP to fill at a reasonable price.
Hi,
Thanks you for this amazing post series, I’m trading this methods for a couple of months now with lots of success.
I did notice that:
– Commission fee by IBKR for selling SPXW options ranges in 8-12% of the sell value, I couldn’t make any sense why this rate changes (maybe when orders are filled in parts?)
– Exposure fee is also applied and account for around 16% of the total income, also fluctuates a lot
Can you share what is a normal expectation from these fees as they take almost 28% of the income pre tax?
Also is there anything I can do to reduce them?
BTW: I’m using EU based IBKR
Yes, for 0.10 premium and $1.20 commission, you make only $8.80 per contract.
I would avoid the exposure fee unless the premiums are particularly rich that day, e.g., 0.25 and above. And even then I don’t max our my entire margin.
One way to avoid the exposure fee: Trade only a smaller number of contracts over night. Then, the next morning at the open, fill up the remaining margin with 0DTE contracts that don’t count toward the exposure fee calculation.
Thanks for the tip! How much gap to you leave in order not to max out?
The gap will depend on your underlying portfolio and its sensitivity to a market drop, and it depends on how far OTM your strike is. You can look at the calculations under Performance and Reports -> Other Reports -> Exposure Fee Calculation. Best is to experiment on size, see if it triggers the Exposure Fee, and review the calculation the day after.
Great answer!
Matt-C already answered this. One more comment to add: the further you sell out-of-the money the more you relax the exposure fee constraint. So, if during normal days, I sell 14 1DTE puts, I can expect to sell 16+ puts without EF when the market is particularly volatile and I sell 10%+ out of the money. After a while you get a pretty good feeling for how many 1DTEs you want to sell.
Thank you, much appreciated 🙏
Hi ERN,
I notice in IBKR that the SPXW options that expire OTM does not have the margin released until around 4:00AM ET the next day. It means that if we run 3x leverage on these positions, the the margin required held overnight are actually 6x until 4:00AM ET as we have 3x from new positions we just open and 3x from the held margin for expired positions.
Does it mean we won’t earn interest on the held-up margin overnight? Assuming $60k margin for 1 SXPW contract, at 4% interest rate, it will be roughly 7$/day that we miss.
I am asking because if I trade ES options that pretty much release margin instantly after expiry at 4PM the same day, I can earn this extra interest accruals (with the trade-off that transaction cost will be higher as I need to trade 2 ES options vs just 1 SPXW option).
Thank you very much.
My account uses Portfolio Margin and the margin is released about 20-60 seconds after the market close on the same day, so I can still trade until a 15min after the NYSE close before the options market closes. That’s plenty of time for me.
If you don’t have portfolio Margin you should get it ASAP. if you have Portfolio Margin and your margin still comes back with such an unacceptable delay, you should complain to IBKR.
Has anyone looked at the SPUT ETF, the Innovator Equity Premium Income — Daily PutWrite ETF? It’s the closest fund to ERN’s strategy that I know of. It’s 50% S&P 500, 25% treasuries and 25% dedicated to the put selling strategy. They buy leveraged ELNs that track the Goldman Sachs Enhanced Daily Put Writing Index (5 Delta) that cover the equivalent of 100% of the fund’s NAV.
Thanks for the link. I don’t think this is a useful replication of my strategy, though. It’s too early to tell definitively, but I don’t like the performance chart. SPUT went down almost as badly as the S&P 500 in April, but the recovery was only muted. That’s not how my strategy performed if you had added it to a 50% stocks+50% fixed-income portfolio.
Hello ERN, when you sell a long-dated put spread with 1,000 points between the low and high strikes, doesn’t it consume a considerable amount of margin, preventing you from selling your usual number of 0DTE and 1DTE positions?
A 1,000-point difference represents a max loss of $100K for margin calculations. That might end up making you sell fewer contracts than you intended during that period, meaning you might actually make less money than you could.
Am I misunderstanding this? Or maybe you never planned to use your maximum available margin for 0DTE trades in the first place (even though, in theory, 0DTEs are safe because their stop orders are always activated).
Try it out. It doesn’t take much margin. Sometimes only $1k-$3k, depending on the strikes.
Gonna try it out when vol will spike.
Thanks.
Exactly. Only when vol spikes. Which means, this sub-strategy has been running almost dry for months now. My most recent entry points were on:
6/13/2025
7/31/2025
8/1/2025
8/4/2025
… all with only iffy, moderately high VIX and each time with only very small quantities. Waiting for the next VIX at 30+ to go in again.
I noticed you increased your primary targeted 0DTE premiums to $0.10 this year after starting them at $0.05 last year. I’m wondering if you think $0.15 might be a more profitable premium sweet spot? I agree with not reaching too high for premium, keeping the payouts at “catastrophic insurance” and “lottery ticket” levels. It’s just that with premiums after fees being about 57% higher at $0.15 over $0.10, I suspect this might be the more profitable premium for 0DTE despite having a lower premium capture rate.
Curious on your take of selling $0.15 vs $0.10 0DTE puts/calls for people who target static premiums (not interested in analyzing premiums vs deltas) with a 20x stop. Do you suspect on average the higher premium would overcome hitting stops more frequently with a higher payout? Is there a program to backtest this?
Also curious about this. following along.
I’ve never consistently sold only 0.05. Especially on the put side, I’ve avoided 0.05 because that just takes margin away and pushes me from the pre-market close to post market close where I tend to lose 0.05 in premium revenue while waiting for the margin to come back. Not worth it.
I did start with 0.05 on the call side, but since increased it to mostly 0.10.
I will monitor the 0.15, especially on the put side. Might be worthwhile to increase the revenue a bit, even if that means I get more false alarms through the STPs.
Why monitor 0.15 “especially on the put side” when the call side has shown a higher premium capture rate? I assume you suspect the dropoff on premium capture rate to be much sharper as you go closer to the money on the call-side, making calls likely to be less profitable at 0.15 than 0.10?
On the 0DTE call side, even at 0.10 I’ve had lots of losses and even more “close calls” (pun intended) this year. I will stick with 0.10 for most of the time. Unless there is a special opportunity. But that’s rare.
Hi Ern,
I’m curious how your calls have been doing over the last few months given the bull market and (from what I can see anyway), low IV and premiums?
It looks like you’d need to sell much closer to the money for the same premiums, though I assume with your strategy this still keeps you far out of the money.
Calls have been the worst PCR (premium capture rate) this year. But they are still profitable.
Last year the 0DTE calls we better than the 0DTE puts.
I’m curious about the 0DTE strategy because it de-risks you from the overnight black swan. Looking at your breakouts, 0DTE seems like it could match your 1DTE net profit if you scale the N by 5x (~2000 to 11000) and you are able to increase PCR to match 1DTE PCR (30% ->70%+). Have you considered looking at optimization strategies for 0DTE (lower premium higher PCR etc)?
Well, the 0DTE puts this year have a PCR of 65% (and only 32% last year), compared to 90+% for the 1DTE. I know, the black swan risk is there, but the income from the 1DTE is still more solid.
I’m guessing you can’t do this with put credit spreads.
I tried this option strategy yesterday (1DTE) with SPX put credit spreads. The best I could do was 20 cents per put credit spread. That came out for $20 per $500. I wrote three credit spreads, risking $60 per $1,500.
Then I realized that the put credit spreads were within 1% of the current SPX price. IBKR said they were 98% chance in the profit. SPX proceeded to drop another additional .4%, which put these spreads within striking distance.
I’m feeling lucky today, but betting $500 to get $20 in premium? I’m going to pass the next time. I don’t have the margin required for a straight writing of a SPX put, unfortunately.
Thank you for sharing your option strategies with us, though. I wish you success!
Why would you do that with only a 5-point spread? Normally, people use a wider spread further out of the money.
Hi Karsten. I currently am selling 1DTE far out-of-the-money short put options on the SPX index, near the time of market close each day, without the usage of stops. The SPX options premiums seem to correlate well with both current and expecte next day volatility events, such as FOMC meetings or other volatility measures such as the VIX. In other words, in a low volatility environment, the SPX strike for a $0.10 premium short put option will be significantly closer to being in-the-money than on a high volatility day. I’m not sure if you have personally maintained or have access to historical records to check this, but would selling a 1DTE $0.10 premium short put option near market close ever have ended up in-the-money at expiration of the option the next day? Thank you for sharing your valuable insights.
0.10 would have been scary for the 1/27/2025 expiration. I sold them on Friday at the close. The S&P got close to the strike and I closed some of them. The index never went below the strike, but you might have still suffered a loss that day.
Hello, I cannot find the comment button so I am riding on one of the threads.
Thank you for generously sharing your strategy and knowledge. I am new to options writing and find your blog very very useful.
I have been using your strategy for the past 2 months with great success. Due to time differences, I cannot sell puts at market closure, so I sell at market open. I understand there is greater risk but at the same time this limits me from selling too many overnight puts so the risk is lowered. I can earn about 3% pa based on this strategy.
I hope to fully retire soon, and I read in some of your posts that you thoroughly recommend this strategy as a way to alleviate sequence risk for early retirees and also increase income for a smaller portfolio.
Could I clarify that this means that I can do the following:
1. Create a 1M portfolio of assets in IBKR e.g. 50%diversified etf + 50% bonds
2. Withdraw 2.4k per month i.e. Swr of 2.4%
3. Sell 1dte and 0dte options for an income of around 3k i.e. 3%
So this means that I have lowered my swr to a super failsafe rate, hence alleviating sequence risk, while not compromising on income since I supplement with put writing that is done based on the margin of same portfolio that I withdraw from. And you also have high conviction that this strategy works even in bear markets so therefore the income stream should not dry up in a bear scenario.
There is no risk of assignment so I do not have to put aside large amounts of cash. But I should put aside a certain amount e.g. 7k per put in the event of cash loss.
Are there any other pitfalls that I should consider before committing to this?
Thank you for your advice!
2.4k per month is 28.8k per year, so 2.88%, but close enough.
If you generate another 3% p.a. from the options trading you would have a great retirement outlook.
I’ve been through 2 bear markets: 2020 and 2022 and many other vol events: 2011 debt downgrade, 2015 vol, 2016 Brexit, 2018 scare, etc. and my experience has been that the higher-vol periods are actually more lucrative. So,
About the cash cushion: yes, you may need a few $1,000 for occasional losses. But you can go into a negative cash balance and pay back the occasional losses with the new gains after the vol shock.
Pitfalls: don’t overdo the leverage. Don’t go to max leverage overnight!
Good luck!
Thank you for the positive reply.
Yes the numbers are wrong. What I meant was
For a 1M portfolio
– withdraw 2k per month (2.4%swr)
– earn 2.5k on put writing per month (3% pa)
This should be a sufficient spending amount for my retirement.
Because I sell the 1dte puts at market open, this means that I cannot sell the maximum number of 1dte puts due to the previous day 1dte puts that are still open. I am fully leveraged during the day i.e. 1 put per 120k but when market closes, i only have 1 put open per 300k of portfolio. I do not get charged any exposure fee.
Thank you once again.
Very cool! Glad we’re on the same page.
Thanks for your reply Karsten. It’s interesting… looking at my own trading records, I sold some 1DTE short puts around the time of market close on Friday 1/24/2025 (expiring on Monday 1/27/2025) for a $0.05 premium; the strikes were between 5,300 and 5,425.
On Monday 1/27/2025, the SPX opened at 5,969. It had an intraday low of 5,962, and closed at 6,012. It felt pretty safe to me that day because the SPX never dropped anywhere close to my strikes.
I’m curious what were the strikes for your 1DTE $0.10 short puts that you sold on Friday 1/24/2025? I’m wondering if there was a large discrepancy between the strikes for my $0.05 puts and your $0.10 puts.
You also mentioned that you closed some of your puts because the S&P got close to your strikes. Did you decide to close your puts at a loss after the market opened down on Monday 1/27/2025, or during the pre-market hours? Unfortunately, unlike Interactive Brokers, my brokerage doesn’t allow trading outside of normal market hours.
Thank you again!
My strikes were: min=5775, max=5900. I liquidated some of them. Too close for comfort.
Big difference between 0.05 and 0.10 contracts!
When volatility is higher, you mentioned being able to sell at higher premiums. I’ve noticed that too. On days with higher VIX, I can go further out of the money, at smaller deltas and still make higher premiums. I’m curious about how you set stop losses on those days. When premiums are higher, are your stops also at much higher amounts? For example, say a typical say you earn $0.10 premium and set stop at $1.00. Now with high VIX you sell at $0.70 premium, would your stop still be $1.00 or actually $7.00? If it’s $7.00 and you hit it, then you would need potentially 70 days to recover (assuming in subsequent days the premiums quickly go back to their long term median).
I’d move up the STP when the premium is higher.
Hey Karsten, in the Two sides of FI video (which is great btw), you mention several things that caught my attention:
– what did you mean when you said that you were maxing out margin on the next day with 0dte? Are you really going to the maximum limit of options you can write or do you have some target buffer that your risk model advises you not to breach?
– You mention splitting the trades between 1dte and 0dte to avoid the IBKR exposure fee. You also mention that the shorter the maturity the better. Following those two arguments, would it not be better to trade only 0dte every day?
Thanks!
1: I use maybe 50% of my margin on the 1DTE, then fill up the remaining margin with 0DTE the next morning.
2: No. Because with only 0DTE, you would have zero exposure between the market close and market open.
1. Doesn’t that make your leverage level skyrocket on the next morning (maybe in the 6x range?)
2. Is the argument that 1dte have more risk exposure than 0dte? If so why not prefer 1week dte that have even more exposure? I suppose in the end you’ve settled on 1dte as the best trade off?
1: Yes.
2a: Yes. You can take more risk within the trading day and less risk overnight.
2b: No, I don’t want to do weeklies. I have written about this issue at length: I like to use the shortest possible DTE to take maximum advantage of the Central Limit Theorem.
This is where I’m confused
I read the posts you mention and it feels like focusing on 0dte only (the shortest possible one) would enable you to do even more trades and benefit more from the central limit theorem
This is why I assume there is some kind of tradeoff between how far you are willing to go in terms of shortening the deadlines vs how much juice there is in the premiums
Everybody needs to find the sweet spot of what they want to
Very true. Some folks do only 0DTE, but I think they’re missing out on the juiciest premium overnight.
Everbody needs to figure out which maturity, or mix of maturities, works for them best, but there’s no denying that a lot of the premium comes from exposure for when the market is closed. You can look at the option prices between 4.15pm and 5pm EST, and compare them to when the market reopens at 8.15pm. Even if only a couple of hours separate the two, 50% of the premium may be gone already.
Thanks!
I agree that 1 DTE has a lot more premium than 0 DTE. You can sell a 1 DTE further a lot out of the money and still make substantially more premium than 0 DTE.
What I don’t know is, since you can’t stop out our position if the SPX falls overnight, what the potential loss might have been. A 0 DTE crash would be limited to somewhere around your stop loss, which I currently set at 10-15x the premium. A 1 DTE fall can even land you in the money at market open where the loss is much much higher.
Do you have any sense of what that loss could be based on historical data?
Highest overnight/weekend drop: -7.5% on March, 16, 2020 from 2,711 to 2,509. VIX d-1 was 75 so not “out of the blue”.
Yes, thanks for checking. That overnight drop was during the height of the pandemic. Did not mess with my puts at all. March 2020 was the highest-profit month so far.
Yes. I have data on close-to-open movements and I can calculate how likely it is that my positions go in-the-money. I still find the 1DTE options to be the most attractive. But again, due to the exposure fee and just out of an abundance of caution, I don’t go overboard with the 1DTE. I leave plenty of excess margin. Which I then fill up with 0DTE puts at the open.
Exactly. Most of the time I’ve made almost all the premium when the market opens for the off hours at 17:15 Pacific Time.
Hello Eduardo and Karsten,
You mentioned that the largest overnight/weekend drop was -7.5% on March 16, 2020, with the VIX the day prior being 75. I’m wondering how far back in time your analysis covered. Did it go as far back as the 1980’s?
Karsten previously published useful charts demonstrating the worst S&P 500 daily returns since 1987 vs. the implied volatility (VIX/VXO) the prior day. For risk management purposes, it would be super interesting and helpful to see a similar chart that just focuses on the largest historical overnight/weekend drops.
Thank you!
I use the 1987-2025 daily data in my analysis. I don’t exactly trust the “Open” quote provided by Yahoo Finance for the early data, though. But for the last 10 years, it looks legit.
No. You’d miss out on the overnight premium.
Hi Karsten
Why are u comfortable with maxing out your margin with 0dte to the tune of 6x leverage but not for 1dte? Is it because u can rely on stoplosses or u don’t think that the market will go from a low vix to high vix within the day? Or is there some other criteria u look at before u decide to sell some 0dte for the day?
Thanks!
1: overnight risk (gap risk)
2: the IB exposure fee
Karsten, were you aware if what happened on Tuesday 14 October?
There were strikes deep otm that had all liquidity pulled at 338pm EST when trump tweeted about cooking oil tariffs.
Anyone with a STP in was at risk, the first people out indeed exited around their stop levels but then there were cases of people paying 100 (ie $10000 per contract) on STP MKT orders for these strikes even though SPX only dropped 25 points.
As far as I can see the only way to have avoid this would have been to set STP LMT orders.
I’m not aware of that situation. Do you have news articles and the exact strikes and that were impacted?
No news articles that I can see but check out time and sales on 6535P for example at 15:38 est on 14 October
My advice: stay away from the low-liquidity contracts that end in 15,35,45,55,65,85,95.
I watched a bidless 0DTE put end up with a 120 ask for a few moments as liquidity dissapeared accross the entire chain. I thought the world was ending until I clicked over and SPX was only down a few dozen points.
You can see a bit of what happened here:
https://0dtespx.com/?date=2025-10-14
Expected move went from 9 to 45 over essentially a nothing tweet, market makers pulled liquidity and the spread selling strategies got slaughtered, erasing months of gains in a single minute before it stabilized.
But if I hadn’t glanced over at that screen that minute I might have only barely noticed later.
Yes and it made me think that if you are running a strategy such as this that is reliant on stops it may well be better to use STP LMT (with a generous limit) rather than STP MKT.
Of course that gives the risk of a skippped stop in a true gap event but I guess one has to pick their poison…!
I’m considering exactly that as well — maybe a $5 limit on the $2 SL? Just not sure what to do.
Yes, I will use the STP LMT like that from now on.
The expected move curve in the chart is based on ATM 0DTE puts, it seems? I wasn’t in the market on that day, but a much lower delta put (<1 delta) may have repriced to a much lower extent, right?
Very scary. I will reassess my STP strategy. I will use STP-LMT orders for now, i.e., set the STP as before, plus a LMT a little higher. That way you can still hedge the occasional market move in the wrong way, but you avoid the crazy fills everyone heard about.
Seems like IBKR requires the LMT and STP to be fairly close – I tested this quickly and with a $1.50 STP the LMT max is $1.75, $1.80 is rejected. $2.25 STP seems to max out around $2.60 LMT.
To hedge against a market buy that is 100x+ put premium in a gapping market, I suppose a $1.50/$1.75 STP LMT helps protect against this, but the new risk is holding a dangerously high priced put with no exit plan. Could this threaten account liquidation at extremes? I think maintenance margin calculation is based on simulating SPX and VIX movements, and if the extreme pricing is primarily liquidity-driven and not due to huge SPX/VIX moves, maybe not? In either case, is the hope that the brief extreme pricing doesn’t get *too extreme* and instead normalizes so that your LMT executes at $1.75?
In a slower moving market, STP LMT sort-of replicates STP MKT, but I’d prefer a larger gap between STP and LMT allowed. 1.75 is only 17% above 1.50, and I’ve had STP MKT orders fill at 20% higher. Does anyone have a better way?
Separately, did anyone actually experience extraordinary losses on 10/14? Aren’t there enough active market participants here (hi, Karsten!) that there should be firsthand accounts? I had a rare day off from markets, and I’m trying to wrap my head around the prospect of a far OTM 0DTE blowing up to 100x premium at 3:38 PM with 20 min left in the day.
If I’m overcomplicating or getting anything wrong, please let me know, thanks
I think this would make a good topic for a post from the big guy on this if possible. I think there is something like a ten second window where CBOE is suppose to bust the trade, and in theory that might be the best answer – not sure.
I am a member of a trading community where plenty of people (myself included) saw crazy fills on 10/14.
Luckily in small size so a scratch in the grand scheme of things following strategies different to this one, but e.g. I had one position over 200 points OTM that was trading at 0 / 0.05 fill at 17.00!
Obviously I asked for a trade bust, but something clearly went wrong that day and CBOE blanket refused all busts – so yes this is definitely a real risk.
Personally I don’t think STP MKT is a viable stop method any more as a result of this.
Yes, a 17.00 fill is likely not worth a trade bust.
I think a STP LMT does exactly what it’s designed: It hedges against a normal, garden-variety move against your positions. If there’s a crazy move it doesn’t fill. So, never set your leverage to where you couldn’t stomach moves we saw in, say 2020 without STPs.
Given some of the rumors I heard, relying on the CBOE is likely not sufficient. They apparently busted the puts that filled at a price of $5,555 but that was so outrageously wrong. But I wouldn’t trust CBOE to fix your issues with a $30.00 fill on a $2.00 STP.
Correct. You can’t set an unreasonably high LMT price.
My advice: Don’t enter trades you couldn’t stomach in the absence of STPs. I’ve been running this strategy form 2011 to 2022 without STPs and always scaled my trades that even a crazy blowup didn’t threaten my principal.
I had no issues on 10/14. None on 10/22 either.
Why do u think you had no issues on those dates? Was it the strike price you chose?
Also, am I right to say that people with no stoploss would have gone thru those days unscathed because in reality, the market did not drop much.
Several reasons:
1: I try to avoid odd strikes like 6335 or 6295, where you have less liquidity. Exceptions would be strikes ending in 25 and 75.
2: I target very little premium, usually only 0.10-0.15 for 0DTE. I set the STPs also very low. My STPs are executed relatively early. If you set your STP at 10 or more, it’s usually too late and liquidity has dried up.
I didn’t encounter any issues on those dates, but something unusual happened to me. On October 14, I sold the 6195 put for 0.15 – it was about 6% OTM at the time, and I placed the trade at market open.
I set up a stop market order to buy back at 2.5. Around 15:30 ET, when the whole saga occurred, the 6195P briefly spiked to 4.1 USD, but the stop wasn’t triggered. While I’m relieved it didn’t execute, I’m a bit concerned about why it didn’t trigger.
Any thoughts on what might have caused this?
Lucky you! Not sure what provoked that. It’s possible that either the broker or the exchange pause the STP if the crap hits the fan.
I’ve also had one STP that was executed but a lower price than my STP, i.e., the market just barely nicked the STP price but then settled down. Strange things happen sometimes.
Hi Karsten,
On 10 Oct, Trump tweeted about China at about 11 am. This caused the entire market to drop about 2%. At that time, I already opened my puts for the day with stoplosses. I sold 1dte puts at 5%otm.
All my 1dte (previous day and current day) and 0dte puts had their stoplosses triggered. In fact, premiums went up to higher than 20x even though the market only dropped 2%. So I would still have lost money with 20x stoploss. I removed my stoplosses briefly but in the end decided to close my puts at a loss.
May I ask if this is normal? If so, does it mean that my stoploss is too tight?
The market went nowhere near my strike price but the premiums shot up so high that the stoplosses triggered. I am still trying to recover from this loss. And I am trying to reflect on what I could have done to handle the situation at that time.
I also stopped both my 0DTE and 1DTE puts on that Friday. That was the only time that I’ve been running it all year that the puts stopped. They eventually went all the way over $5, so it was a decent stop. It was a very rare day indeed! I earned it all back by last Friday, so it was just a single week drawdown.
I am no longer negative as well. But I was pretty jittery for the whole week.
So was Apr 7 not as bad as Oct 10? The vix was pretty high on Apr 7. Oct 10 vix wasn’t that high.
My trading log has a gap between April 4 and 14th. Vix 50 got hands. I skipped both those events. The first one maybe a higher SL would have helped, but the second you absolutly needed to stop.
Ooh I see. So u sit out high volatility periods?
Glad you recovered the money. Oct 10 was a crazy day. But the premiums for Monday made up for that.
My STPs for the 0DTE were triggered. The overnight, 1DTE STPs were not. So that day was a pretty garden-variety loss for me.
There was some slippage, i.e., my STP was at $1.00 but filled at 1.20. That happens. I have not personally experienced these crazy STP fills, but I have heard rumblings about them. It’s something on mind. I am working on ways to address that. One solution would be a STP LMT order where you submit only a LMT order when the STP is crossed.
Karsten – I’m terrified of the low-liquidity situation above from Oct 14 so I don’t use STP order. However, there are also problems with STP LMT. I had a 15x STP @ 16x LMT order on Oct 10, and the price after the China Tweet blew through my LMT. Luckily I happened to be watching the markets at the time, and I had to manually close my positions at 21x. If I hadn’t been able to respond within seconds, I’d have suffered over a 100x premium loss at expiration.
This event combined with Oct 14 has me worried more about edge cases with this strategy. STP orders are obviously a no-go as they expose us to huge tail event liquidy issues, but STP LMT orders still don’t solve the problem, unless we put a much bigger LMT. Since retail investors can’t really do high-frequency trading, how do we protect ourselves against sudden and nearly-instantaneous drops in the market beyond our LMT orders?
for what its worth, when I do STP LMT, my STP is around 13X and LMT is around 18X, so I have a bigger gap between the two than what you did
Noted. I do the LMT about 1.2 to 1.25x the STP price.
Noted. But the STP LMT still does its job. If the flash crash is so rapid that liquidity disappears, you may just sit it out and wait for the situation to settle down. Most of the time, the STP is just a false alarm.
In a flash crash (such as 2010) where stop limits were jumped over due to low liquidity, margin requirements may go up a lot on the short puts, leading to margin calls. The puts may even end up in the money leading to a big portfolio loss end of day (if something like the 1987 crash happened).
So it seems like the protection options are
1) no stops, but a 1987 one day crash would be very very painful
2) stop limits, but a risk that with low liquidity your limit gets jumped over and you’re back to the mercy of no stops
3) stops, but with a risk that liquidity dries up and you end up with a really, really bad fill like what happened on Oct 14th at $100+ per put
Among those 3 options, what would you prefer to do?
I suppose a 4th option is just sell as a spread, which will eat into profits, but least your max loss is significantly less. Maybe constrain max loss per day to a fixed % of your total assets like 15%?
That’s why you always keep plenty of margin cushion.
I am currently using STP LMT. An event like 1987 would not have jumped over the STP LMTs because that decline was very gradual.
A fifth option is to use a Delta hedge with ES futures if your STP LMTs jumped over and there is no liquidity to get out of your short puts.
Karsten – How do you know when you’re going to get hit with IBKR’s exposure fee? I’ve been selling 0DTE and 1DTE puts for a year now, far out of the money at a .01 delta or less, and still get hit with the charges.
IBKR oh so helpfully sends an email the next day stating you got charged the exposure fee, not before you sell and trigger it.
If I could figure out when that gets triggered, I’d sell one or two less puts, since it’s effectively eating the premium of one or two, with all the downside risk.
And the cost of the exposure fee seems to vary by 4x sometimes! Is there any rough way to calculate it? Sometimes it’s on down days, and other times it’s on days that go up!
Thanks again for your blog. I’ve learned a ton here!
They calculate a theoretical 30% drop in all equity and equity-like assets. If you have a shortfall you get hit. It wouldn’t take into account the 0DTE, Only the 1DTE or longer.
Go to your account click Reporting, click Other Reports, under “Risk” click “Stress Test” and simulate if you’d have a shortfall after a 30% event.
Amazing!
Hi Karsten,
Do you have any opinion on selling commodity put options. If one gets exercised the underlying is a futures contract which you immediately sell. I am thinking mainly about Gold futures options but it may be Equity index options. Thank you
I’ve not tried the commodity route. But there should be a sizable vol premium there as well. Please let us know how the gold vol selling works for you!
Hello, thanks a lot for sharing your strategy.
I’m still going through all the posts and comments, but couple things that arent’ completely clear for 2024.
– could you advise, when you say the returns were 100k or roughly 5%, are these related to your whole portfolio ($2m in this case), or only the 40% “non-stocks” portion?
– what was the leverage used to generate these returns in 2024? And is the leverage multiplier also related only to “non-stocks” portion or overall portfolio?
$100k or 5% of the taxable portfolio at IB only.
I have other assets, too, that are not counted in the denominator because they are in other accounts and other account types: 401k, IRA, Roth IRA, real estate investments.
The leverage is calculated as roughly 1 short 1DTE put per $160k of taxable account value.
Hi Karsten, does $160k/short 1dte put imply leverage of ~4x? I get: ~6850 SPX * 100 * 0.95 / 4 = ~$163k, assuming selling at 5% OTM. 3.3x leverage would translate to ~$200k, which is closer to where I’ve been.
I suppose it follows that if SPX grows faster than underlying portfolio over time, capacity for short puts would decrease. And vice versa, if you beat the market you also get to sell more puts.
Thanks in advance for any thoughts
Yes, all correct. Recently I’ve been squeezed out a little bit because the index grew faster than my IB portfolio, which has only about 35% US equities.