Theo Casey is CEO of EquityQuant.dev, co-founder of MarketColor, and a contributor to veteran FT Alphaville.
At one point, “It’s only worth the news because Kathy Wood is doing it,” the bandwagon runs out of road. But today is not that day.
One of Wood’s recent ventures includes his favorite downside protection derivatives strategy among the so-called buffers, which are ETF hipsters. ARK Invest was applied to the SEC in July and created four defined result buffered ETFs that reference one innovation ETF for each quarter, spurring the opportunity to code them in some way.
Potentially, say hello to the Ark-defined Innovation ETF family: Arki, Arkj, Arkl, Arkkm. Buffer Arks are listed in sequence simultaneously in a one-year series from January to January, April to April, July to July, and October to October. Although each ETF has a 12-month outcome period and no funding structure for funds, weighting four series in a portfolio means that one can be simulated. Each series expires and resets on their first birthday.
This structure includes 50% downside caps provided by Moneyput, funded by Money-at-In-Call. For upward participation, the rest of the call premium will buy a cheaper 5% money call. Like other buffers, B-Arkk uses the non-standard contract, Flex option, with customized terms sold at the counter to avoid headaches associated with allocations.
This is the high-betta variant of buffer we’ve seen so far. This relies on profits of over 5% over a 12-month period, which would have been safe for ARKK in recent years.
Why is Buffer Ark on sale now? Perhaps the funds are warmly received as defined results, with approximately $70 billion in net worth invested in downside protection exposures to S&P, Russell, Nasdaq and others. Cboe announced this week the scheduled first quarter of 2026 launch of Flex Options in Europe. There is anecdotal evidence suggesting that investors will buy and hold buffers longer than they will make funds without downside protection.
Curiosity is the gap between Buffer Ark’s shortcomings protection and upside participation. What happens during a 12-month period when the portfolio bounces between 0% and 5%? Very few. Investors do not have any losses, but no returns either.
In the dataset starting in January 2016, ARKK traded 175 of 1,745 in the 0-5% range. At the time, investors would be better off keeping ARKK than buffer arcs.
That said, we have modeled the virtual performance of Buffer-Arkk over ARKK. And that’s not bad.
Below is a visualization of Sigbox that simulates the realised historical return distribution. The passage of time from the beginning of January 2020 at the bottom of the box has risen until the end of June 2025. The parameters are the arithmetic mean returns and the volatility achieved. (I call it sigbox because it looks like a box signature.)
If the performance line is blue, ARKK should outperform the buffer arc. If the performance line is red, the buffer arc should surpass ARKK. The ghostly grey walls (static versions of black) represent the equivalent curve in which funds perform with each other on a risk-adjusted basis.
Next, follow a note to your fellow Quant. Normies can skip this bit.
I took some freedom to visualize what I could understand about Arkk vs Buffer-Arkk, and its potential future. Sigbox is designed to provide insightful analysis of ETFs, potential future ETFs, and their derivative holdings. This example uses a dynamic model to capture the realized distribution of underlying returns from historical data. Next, we calculate an equivalent curve to distinguish where ARKK and its comparators (i.e., a composite of hypothetically defined innovation ARKK ETF series) have the same expected rate of return. Relative outperformance from a risk-return perspective can be observed over time. Simulation is not a comparison backtest using historical data. And of course, the start date is always biased. The bespoke model excludes some extreme features, such as the Hearst statistics, which do not last and may skew long-term plots.
Inferring your intended customer is what makes Ark’s regular N-1A filing so interesting. Who was the drawdown protection rapper asking for a high beta fund? Again, the buffer arc limits the loss to 50%. If Arkk drops to 72% (as happened between 2021 and 2022 on each of the next chart), the hypothetical buffer arc would be around 36%. That’s a neat trick, but referring to the AQR discussion on buffers, why not buy a low beta fund?
Perhaps it is another manifestation of the Ark’s insatiable brand equity. Buffer Ark is probably Ark’s Full Fat original diet cola. Customers who are probably going to have a glass of water are happier when they put their faith in the brand. They want Ark’s full-fledged potential, but they may prefer a version of the left tail with less gapdown in its upside down potential, less circuit breakdown, and less risky version. And perhaps investors will be willing to give up 5% every 12 months to achieve such defined outcomes.
Some content could not be loaded. Please check your internet connection or browser settings.
When a New York magazine announced the rise of Cathy Wood in 2022 (“Listen to her was to feel your mind liquefy in a flood like the clickbait of the dopamine-inducing buzzword”), it may have been difficult for many to take her technobubble seriously. However, Wood has always been more of an aura and presentation than a fund manager, and her flock doesn’t get enough ARKK. With 8 active ETFs, 4 crypto ETFs, 1 venture fund and 2 private funds, ARK is happy to offer.
ARKK led an active ETF revolution, and today it places and calls a deal with ARKK as a reference asset. You may soon add a buffer fund that consists of Flex-Option equivalents of these American-style listed options in ARKK to your list of ARKK funds.
What’s coming next? Leverage, inverse, revenue arc? No – because it already exists courtesy of other ETF publishers. TRADR ETFS offers double length and reverse ARKK ETFs, with leverage shares of three times longer and reverse ARKK ETP. AldeMax has a premium seller revenue ARKK ETF.
Before the conceptual presence grows very large, the industry can wrap the same custom basket inventory and create its own gravity pull…or push. Some may believe we are already there, even before the buffer is added to the ranks.
