How To Trade Options Efficiently: Part 3
In this post we’ll take a look at trade occurrences, capital efficiency and the Micro e-Mini Futures product.
Looking for the basics? Check out Part 1. Part 2 focuses on the utilization of leverage, capital allocation, overstatement of delta premium, overstatement of duration premium and tax-efficient scaling.
Let’s get started!
Contents
Trade Occurrences
Concept
Consider the historical performance of trading 1SD strangles on SPY 45 DTE:
Should we expect to make $67 and win 83% of our next few trades?
Maybe.
When it comes to probabilities we need to consider statistical significance. In essence, the more data points in a calculation the closer we get to realizing the stated probability.
Flipping a fair coin expects a 50 / 50 outcome of heads and tails. If we evaluate after just one flip and it’s heads, it will appear as though the odds are 100 / 0. As the number of flips increases, we begin to realize the actual probability.
So it is with options trading. We make our first 16 delta strangle trade. Volatility (also known as variance, luck, risk, etc.) is unfavorable and we’re unprofitable out of the gate despite expecting to win 83% of the time. However, if we continue placing trades, we will begin to realize the expected outcomes.
Mechanics
Ok, so we know we need to have a high number of occurrences to realize the stated statistics. How do we go about doing this in practice?
Suppose we intend to open 20 positions. The way we open those positions could be any of the following:
If we place a single order of 20 contracts, the options will all be at the same delta, strike and have the same IV. It’s essentially one big trade. This is not what we want.
The objective is to diversify across time, deltas and IVs.
Instead of deploying all our capital in a single trade, perhaps sell one or two options over the span of a day or week. It could mean selling in the morning then selling again in the afternoon if the market moved. If we’re targeting 16 delta puts, consider trading at 15.5 or 16.2.
I’m not suggesting we take all trades and make them one-lots just for the sake of increasing occurrences. The idea is to create a number of diverse trades for the purposes of realizing probabilities.
Winner: trade often and across time, deltas and IVs
Capital Efficiency
Taking profits is all about increasing capital efficiency, reducing risk and locking in gains.
Concept
As we learned in part 1 of this series, 16 delta put options reach 50% max profit, on average, about 61% sooner than expected (14 days vs 23 days). We also observed the average duration to reach respective profit targets.
Let’s graph this to see if there are any trends we can easily identify.
Options reach the smaller profit targets in an accelerated, nonlinear fashion. The trend levels off significantly after reaching 50% max profit.
Is this something we can use to our advantage?
You betcha!
Mechanics
Suppose we sell a 16 delta 270 PUT on SPY with 45 DTE and collect $1 in premium – an actual trade I could have made while writing this post. Assuming the trade is profitable, holding till expiration will give us the equivalent of a 3% annual return on capital (ROC).
Calculation: [(premium / strike) / (days in trade / days in 1 year) * 100 to convert to percent]
Calculation: [(1 / 270) / (45 / 365) * 100]
Managing at different profit targets and durations yields different annualized ROC values. We can see the outcomes in the table below:
green cells are where ROC is improved relative to holding till expiration
boxed cells highlight the average duration to reach the respective profit target
This can also be depicted as the increase or decrease relative to the 3% baseline of holding till expiration
ROC improvement relative to holding till expiration
By managing the position at 10% max profit after 2 days, we have improved our ROC by a factor of 2.25x – from 3% to 6.8%, a 125% improvement!
Also keep in mind, the average performance is to achieve 20% max profit after 2 days. Who knew more than doubling our ROC could be considered an underperformance!
Having earned the equivalent of a 6.8% annualized return over two days, our capital is now free to either standby risk free or be deployed in another trade.
Winner: managing earlier [than even 50%] can improve capital efficiency
Micro E-mini Futures
There’s a new kid on the block. Actually, there are several new kids but we’ll be focusing on just one: the Micro e-Mini Futures.
What is the Micro E-mini Futures product? It’s a solution fully-fungible with the E-mini futures at one tenth the size, launching early May 2019.
The initial offering will not have options available but I anticipate options will be added once product stability has been achieved and liquidity has ramped up.
Why do we care about a product 1/10th the size of /ES? Glad you asked.
First, this allows smaller accounts to participate in naked put writing without incurring outsized risk. Consider a 270 SPY PUT:
Second, this allows smaller accounts to realize the tax benefits of section 1256 – any capital gains and losses are treated 60% long term and 40% short term, regardless of holding duration. Check out my post on the Ultimate Guide to Taxes in Early Retirement to learn why having a portion of the gains classified as long term is beneficial.
Winner: consider selling MES options if/when the product becomes available and liquid
Summary
Increasing number of occurrences allows us to realize stated probabilities. Occurrences should be diversified across time, deltas and IV.
Efficiency of capital can be increased by managing trades early. Consider using 50% max profit as the ceiling for any trades placed, being open to manage trades at profit targets less than 50%.
MES is the ticker for a new futures product equal to /ES but 1/10th the size. If/when options become available and liquid for this product, smaller accounts will be able to write naked puts without incurring outsized risk and can take advantage of section 1256 tax rules. These rules classify capital gains as 60% long term 40% short term regardless of holding duration.
How often and large do you trade? Do you take profits early or hold till expiration? Share your approach in the comments below.
April 14, 2019 @ 10:34 am
I really enjoyed this article, it focuses on the yearly return efficiency, which is what really matters when you have Bills to pay. The new MES product is exciting too, it sounds like a good vehicle for those who are just starting out, or anyone looking to take options on a test drive.
April 14, 2019 @ 12:15 pm
Thanks for stopping by – glad you enjoyed it, Erick! Both are great points.
April 21, 2019 @ 10:57 am
Great series of articles! Everything in one place. Already printed it out for a re-read later on 🙂 Thank you very much and keep up the good work. Best regards from Slovenia 🙂
April 21, 2019 @ 10:38 pm
Thanks for visiting and the feedback, Tomaz! Glad to hear you’ve found the content helpful. More to come 🙂
May 8, 2019 @ 3:10 pm
Once again, great series of articles! Here are a few more questions:
1. Why diversify across deltas?
2. In the “duration improvement across profit targets” graph, does the trend level off more after 50% than it does after 40% max profit?
3. How did you compute what numbers to box in the ROC table? According to your data, would you agree that no edge to capital efficiency exists whether you manage after 1 day, after 2 days, or after 3 days?
May 9, 2019 @ 12:11 am
Thanks Mark!
1. It’s another form of diversification. I’m looking to shake up the greeks as much as possible to make each trade as unique as possible while stying within the guidelines. If I have software running these trades, there would be target delta with a +/- 5% variance, weighted based on VIX.
2. Yes. It levels off quite a bit more. I tend to manage well before 50%.
3. I used the numbers from the TT table above. I would have preferred median days, but that data wasn’t made available. As for edge, I would disagree about there being no difference in capital efficiency between 1, 2 and 3 days. Here’s why: reaching 10% max profit in 1 day provides a higher ROC than after 2 days, both of which outperform a hold-till-expiration approach from a ROC, and consequently, average daily P/L perspective to the tune of about 4.5x and 2.25x, respectively. That’s definitely tangible.
There’s also intra-day movement which has not been addressed. I’ve frequently closed positions after an intra-day movement only to have the end-of-day price be largely unchanged or down. I haven’t looked into optimizations based on intra-day movements – movements a profit-taker order will capture but are largely unaccounted for in backtesting tools and studies.
May 9, 2019 @ 9:58 am
1. Diversifying over time makes sense to me because you’ll sell when the market is higher/lower. You’ll also diversify across IV this way. If you use IV as an independent variable for diversification then you may go long periods of time without trading. Diversifying across delta doesn’t make much sense to me unless you think it’s possible a particular strike may become horrendously over/undervalued, but arbitrage should keep these in line for liquid issues. Is there data to support diversifying over delta?
2. It looks to me like the trend levels off comparatively from 40% to 50% as it does from 50% to 60%. If it’s about “trend leveling off,” then I would seriously consider 40%.
3. Doesn’t the table show 10% of max profit after 1 day, 20% of max profit after 2 days, and 30% of max profit after 3 days (not 10% max profit in 1 or 2 or 3 days), which is consistent with equal annualized ROC numbers (13.5%)?
I think the advantageous exits based on intraday movement are offset by the disadvantageous entries we might experience by trying to time trades or even taking at same time every day. In the former case, we may get an unable, which lowers annualized return. In the latter case, maybe we could have set a day limit order for a better price and gotten filled at some point. Given a large sample size, would you think these to offset?
Good stuff!
May 10, 2019 @ 12:36 am
1. Agree on time and IV points. As for across deltas, it’s a tactic to help decrease the amount of positions that cross a profit target. I just wrote a response to your question in part 1 which explains this in a little more detail.
2. I’m thinking the same thing! I rarely get out of the 30% range, as seen in the trade log in parts 2 and 4. I’m also quick to take profits when VIX is low. If it’s higher like it has been this week, I may be inclined to keep them on longer in anticipation of VIX contraction.
3. It does, but those are the average performances. Example: I put a position on this AM and the market closed with it at 31%. I was busy doing other things today and wasn’t looking at it, otherwise I would have taken it off. This represents a 13.5x improvement on ROC. Is this an average performance? No way. I’ve realized nearly a third of all the available profit in ~2% of the time. I now have a position that offers a third less profit for nearly the same trade duration. Not worth the risk. If VIX remains high I’ll redeploy capital. If it contracts significantly I’ll decrease capital allocation.
Good point. Yes, I’d agree they would offset.
May 10, 2019 @ 9:39 am
1. I’m not familiar with the flapping analogy, but you wrote “if a 15.5 delta position reaches a profit target but a 16.2 delta doesn’t, the problem has been mitigated without significantly changing the risk profile of the strategy.”
I’m not convinced. No matter how delta diversified you get, the strike you selected may always not hit a profit target when a farther OTM strike would have. The only way I see this matters is if a large segment of the market is always watching a particular delta strike such that supply/demand dynamics get thrown off making a particular strike option more/less likely to hit a profit target. Between different skew models and brokerage platforms, though, who knows what delta is the “right” delta or most common? I certainly have no idea what the MM (computers) are using (likely different ones).
On top of this, by laddering into positions you get temporal diversification across strikes due to market movement and changing DTE. This naturally diversifies delta. If I’ve traded a position every day then after a couple weeks I’ll look back and see I have A contracts at the X-delta strike, B contracts at the Y-delta strike, and C contracts at the Z-delta strike. Even from one moment to the next, what strike is X delta may vary simply from market movement so I would not assume the people with whom I am trading were monitoring my particular strikes even if I do use the same X for every single trade.
3. I was asking to clarify what you meant by capital efficiency. You wrote, “managing earlier [than even 50%] can improve capital efficiency.” This is not always the case if 10%, 20%, and 30% of max profit can be realized on average in 1, 2, and 3 days respectively because annualized ROC is equal. This linearity slows down starting with 40%.
May 11, 2019 @ 8:20 pm
1. Agree. I’m just saying using slightly variable deltas can help preserve temporal diversification. In agreement with your point, if the market moves such that a 15.5 position reaches a profit target but is just shy of the 16.2 profit target, that’s one less position that gets closed thus preserving some temporal diversification. The 16.2 position may never hit a defined profit target, but it can still be taken off manually, earlier, for [less] profit (loss).
3. Agree. That’s why I said “managing even earlier can improve” as opposed to “managing earlier improves.” 🙂
April 11, 2020 @ 3:10 pm
Hi! In your series you make it very clear that managing winners early is ideal. But when I look at some of your backtest studies the best strategies for some of them say it’s best to “hold till expiration”. Why the discrepancy?
April 11, 2020 @ 6:29 pm
Hi Coop! Great question. The short answer is that every underlying has a unique behavior and thus some underlying perform better with early management vs holding till expiration.
As for the series, it’s based on the SPY underlying. The majority of SPY short put strategies do outperform when managed early vs holding till expiration.
April 11, 2020 @ 6:44 pm
Thanks for clarifying!
Does anything about your system change during crazy times like now? I know ERN has been going with much lower deltas. Do you change anything?
April 11, 2020 @ 7:09 pm
Welcome!
Nope! That’s the great thing about it. It can be performed in any environment.
For me personally, I stopped selling options on SPY around fall last year since the data suggests I’m better off buying and holding in nearly all scenarios. There are a few exceptions, such as on T or EEM, but my portfolio doesn’t call for specific exposure to those underlying so I don’t trade them.
If the intent is to continue selling options on SPY, lowering delta targets is a definitely a way to lower the risk, as is converting from naked puts to vertical puts (spreads).
April 11, 2020 @ 9:06 pm
Thanks very much for your answers!
Am I correct in thinking that if I’m looking for monthly semi-passive income then selling options (though it generally underperforms buy and hold) is the best route?
April 12, 2020 @ 9:46 am
It depends on what your definition of “best” is. There are lots of ways to generate monthly income. What I will say is that options certainly isn’t a tool to guarantee monthly income.
Even 5D SPY options lost a significant amount of money since mid Feb. While each option sold does bring in some income, the losses accrued will take the better part of a year to break even.
From the lens of dependable monthly income, selling options isn’t a great tool. Can option selling be profitable? Sure. But it won’t always be consistent. In fact, the times in which selling short puts incurs losses are often the times when monthly income is needed the most.
April 12, 2020 @ 10:05 am
Understood, thanks Spintwig 🙂
What other forms of income do you recommend looking into? I’d like to research and learn about new methods. For example, I just stumbled upon your preferred shares post…that was new to me, had never heard of those 🙂
April 13, 2020 @ 12:41 pm
There’s the primary approach that I use – selling shares of index funds to generate your own dividend. I recommend that as it’s the most tax efficient for US-domiciled investors / traders.
Preferred shares are an option but I traded that asset class for their capital appreciation characteristics (shares were priced well below par / liquidation value). Holding them long term to collect the dividend wasn’t of interest since I can get comparable returns from an index with less risk and better tax implications. This doesn’t deny that they can still be a high income vehicle.
High-yield savings accounts are currently paying more than even the 30yr bond and are considered risk free, so that’s an opportunity – sell appreciated bond index positions and park the cash in savings. Vio Bank has been holding strong at 1.75% (no fees) since the cut. If you have liquid assets in a business entity, First Internet Bank has been holding strong at 1.61% (4k min to avoid fees).
If/when the US short-term treasury rates dip below 0% gain (they did this 2-3 weeks ago) one can purchase treasuries direct from the weekly auction (current regulation prohibits auction rates that imply a negative yield) then immediately sell them on the open market for a premium. Worst case, you get your capital back. Best case, you make the spread. Granted, a lot of capital is needed for this but it’s a possible income strategy with low risk.
The carry trade is alive and well. If you have an Interactive Brokers account that was opened using a promotional link, the margin rate for the first 1MM borrowed is benchmark + 1%. The benchmark has been around 5bps since the rate cut. For a time I was borrowing on margin then parking the funds in a high-yield savings account and earning the spread. On a 100k, that’s 700/yr assuming Vio Bank is used. When rates change, simply transfer the funds back to the brokerage to pay off the margin loan. Why don’t the big players do this themselves? It’s limited to the FDIC limit of 250k and most high-yield accounts are limited to natural persons (non-businesses/entities) only.
Hopefully these ideas help!
May 15, 2020 @ 12:19 am
Hey spintwig, you mentioned you stopped selling options on spy last fall. Some of Your leveraged spy puts data however seems to beat a buy and hold strategy. Why not do that?
May 17, 2020 @ 12:53 am
Great question!
Two reasons:
1) the leveraged SPY strategy is still underwater vs buy/hold after considering tax drag for my particular scenario. Option premiums are short-term capital gains whereas buy/hold allows me to take advantage of long-term capital gains tax rates. Also, I’m able to better control my AGI using buy/hold. This allows me to qualify for ACA subsidies (or keep from going over the subsidy cliff) in years in which it’s optimal to realize a “low income.” Selectively realizing capital gains is married with Roth IRA conversions as part of my overall tax strategy.
2) Between the time I stopped selling options on SPY and publishing the SPY leveraged backtest my fiscal goals changed. I no longer have the need or willingness to take the risk associated with leverage (aside from a modest carry trade using an FDIC savings account as the investment vehicle). Specifically, I’m not willing to accept the risk that the future may be significantly different from the past that’s captured in the backtest.
May 22, 2020 @ 6:04 pm
Hey spintwig, so are you saying selling options is a loss making proposition in the long run and you’ve stopped it as a result?
What about buying and holding SPY and using 30% of margin on it to sell options?
May 23, 2020 @ 4:40 pm
Hey RN. It’s not necessarily a loss-making proposition (at least when it comes to short puts on SPY) but it’s less lucrative than other forms of leverage on SPY.
If your margin collateral is SPY and you’re also selling SPY puts using 30% of your available margin, that’s the same as leveraging 2.5x on SPY: 1x for the SPY collateral + 1.5x on your options (~20% margin utilization is 1x).
For 2.5x leverage on SPY I could make far more $ using an instrument other than options.
May 26, 2020 @ 4:07 am
The 20% margin utilisation is 1x kinda beat me. Can you please explain it further? Or, let’s assume a $100k portfolio o make things easy. What other financial instruments can we use to make better returns than short puts? What do you recommend that I look into?
May 27, 2020 @ 12:33 am
A short, uncovered put option will require ~20% of the notional amount as margin collateral. If SPY is $300/share:
Notional: 1 contract = 100 shares * $300 = $30k
Margin utilization = ~20% of notional = $30k * .2 = $6k
If SPY hypothetically drops to $0 the option seller is still on the hook for 100 shares of SPY at $300 each. Whether the option seller gets assigned or buys the option back, the 30k loss will be realized.
Using a 100k portfolio [held in cash or BIL] with SPY @ $300, 3 SPY put options will require ~18k of margin (18% utilization) and will represent 90k worth of SPY downside exposure. Adding an additional contract will require 24k of margin (24% utilization) and represent 120k of downside exposure.
If the portfolio is held in SPY or similar and 4 SPY put options are sold, that’s 100k in downside exposure from the 100k in SPY + 120k in exposure from the 4 put options = 220k of exposure or 2.2x leverage.
As for financial instruments, it depends on what your risk profile is. If you’re comfortable with 2.2x leverage on SPY then /ES or MES futures contracts can facilitate this and will provide returns superior to selling puts while having lower notional exposure.
May 27, 2020 @ 1:48 am
Noted. Thank you so much. I’m trying to dabble in futures a bit. The drawdowns and MTMs worry me. Hence the apprehension. Will def paper trade for a bit and see how it goes.
May 30, 2020 @ 4:22 pm
Hello spingtwig, thanks so much for the very informative site!
Quick Q: In your backtests, do you always compare the returns of an options strategy to 100 shares of the underlying? If not, what is the number of shares you’re comparing to?
Thanks!
May 31, 2020 @ 3:56 pm
Hi Nadav – thanks for visiting; glad to hear the content is helpful.
All backtests compare a fixed amount of capital as opposed to a specific number of shares.
For example, the SPY 45 DTE cash-secured study compares 100k invested in buy/hold SPY vs 100k invested in hypothetical ETFs that execute the respective option strategies behind the scenes.
The leveraged studies have different mechanics but are presented similarly.
Hope this helps!
August 5, 2022 @ 5:06 am
Great article.
The question I have is if you use the mechanics of entering at 16 delta and closing at 50% a few days in of a 45 DTE position and only trading one underlying (SPX), how to you get in another trade when SPX is running up and up and away from support? True we take profits early, doesn’t risk increase when we continue to sell puts as the SPX continues to run up and more likely to revert back to the mean for a pullback? I understand getting out early would make sense if we were trading multiple stocks and there are plenty to choose from, but a little unsure when only trading one instrument. My thinking is that I would have to wait for a pullback before I re-enter into a new position.
Can you clarify?
Thanks in advance
August 6, 2022 @ 9:50 pm
Thanks Michael —
The idea of support/resistance (or any technical analysis for that matter) is completely ignored. Another position is simply opened.
There is no such thing as “reversion to the mean” for equities. They are only worth what someone is willing to pay for them.
Suppose someone asked: what is the SPX mean that we should be reverting to?” What would be the answer? The price as seen in 1999? Perhaps 2001 or 2020? What is this mean based on? Why should prices mean revert in the first place?
Other asset classes, such as VIX futures and preferred shares, have different characteristics which would lend themselves to structural price behaviors like mean reversion or price anchoring.
Risk associated with similarly-configured short put positions on SPX is the same each day regardless of what SPX did yesterday, the day before yesterday, etc.
Hope this helps!
August 7, 2022 @ 8:01 pm
Thank you so much for taking the time to respond. In this article https://earlyretirementnow.com/2020/06/10/passive-income-through-option-writing-part-4/ , Karsten talks about only using a 3DTE as opposed to your 45 DTE to cater for big downturn moves such as the March 2020 crash. From what I see both have merits and not sure which is better risk wise. Going further out time (45 DTE) enables you to go down further in strike which is “safer”. However, closing out your positions early (3 DTE) enables you less exposure which is also “safer”. The fact that he also thrived during the crash is also impressive.
Would love to hear your thoughts on that!
Cheers,
Michael
August 14, 2022 @ 4:34 am
You’re welcome!
Yes, there is less “time for things to go bad” when trading shorter-duration strategies.
Consider a hypothetical 5-delta SPX short put with VIX at 20. The 3 DTE contract might be 2% OTM whereas the 45 DTE contract may be 12% OTM. The distance OTM accounts for the additional time in the market / increased opportunity for “things to potentially go wrong”.
At the end of the day, these factors net out to roughly equal. No free lunch here.
The longer-duration strategies have increased exposure to vega (price movements due to changes in VIX) relative to shorter-duration strategies. Meanwhile, shorter-duration strategies have increased exposure to gamma (price movements due to changes in the price of the underlying) relative to longer-duration strategies.
In order to survive with either strategy, proper position sizing would be needed.
August 14, 2022 @ 5:33 am
I take it you would still favour the 45 dte?
So to get the best of both worlds, would it be an idea to sell a 45 dte and close at 50% or 1 day later, which ever comes first?
February 9, 2023 @ 1:14 am
Thanks Spintwig for your 3-part series. Very instructive!
I like charts with colors (eg heat map). It really helps to quickly visualize sweet spots and compare values across a chart. I am a non-US domiciled investor, and taxes are treated equally for SPY, SPX, XSP, ES, MES.
a) I am interested in lower DTE analysis to limit my exposure. Have you done a similar analysis on 0, 1 or 2 DTE?
b) CBOE has launched Nanos which is even smaller than XSP, but almost no liquidity. Would shorting more XSP contracts for a similar capital of 1 SPX will increase the number of trade occurrences (with variable strikes and deltas) and realize the expected statistics?
c) Have you considered using a bot to automate a strategy and/or generate more contracts?
d) You mentioned that you now prefer buy/hold versus put writing. Couldn’t you do both? Let’s say that I hold a portfolio of $1m (75% stock, 15% bond and 10% cash). I will likely leave the portfolio as is, and not trade. What is preventing me from writing puts on SPY to generate extra income on top of my B/H strategy for the stocks and bonds?
Thanks again!
February 14, 2023 @ 12:55 pm
Thanks!
a) Yes. 0-3 DTE backtests can be found at at https://spintwig.com/tag/0-dte/ (see methodology section regarding tolerances on DTE targets. 0 DTE is 0-3, closest to 0).
b) Ignoring the wider spreads, lower liquidity and increased commission costs, yes, as long as the contracts are spread out over time. If it’s single position of 10 XSP contracts vs 1 SPX, then no.
c) I have and decided against it. Having worked with automations extensively, if there’s a possibility of fiscal loss or deletion of data, such transactions are best left to human intervention for execution. Automation could queue the position for review, but I’d still need to be the one clicking the button.
d) You could absolutely do both. I still trade vol from time to time – whether an SPX option or /VX position – but it’s infrequent. Preferred shares (https://spintwig.com/mreit-preferred-share-dashboard/) took the stage last year from a risk/reward standpoint.