Four lines. No filing number. No docket reference. No leverage cap, no funding interval, no index methodology, no named compliance officer.
That is the entire payload of the item that crossed my desk this week: a New Zealand-registered venue operating under the name O.G. Com says it has submitted an application to the United States Commodity Futures Trading Commission to list perpetual futures contracts written on individual stocks, and that the stated purpose is to bring leveraged equity trading into DeFi.
I read it three times. The fourth pass was only to confirm I hadn't missed the part that actually determines whether a perpetual contract functions — the margin schedule, the settlement currency, the treatment of dividends. It isn't there. Four information points, zero primary sources, no document to verify against.
I have traded perpetuals since 2020. I have watched a funding rate quietly drain a position while the price chart sat perfectly flat. A perpetual contract is not a product description. It is a set of rules about who pays whom, when, and in what asset — and none of those rules were disclosed here. In the sprint, hesitation is the only real cost. But so is chasing a headline whose settlement mechanics do not exist yet.
Context: Why an Equity Perp Is Not a Crypto Perp With a Ticker Change
Perpetual futures are a crypto-native invention. A perp has no expiry. It never converges to spot through physical delivery. Instead it holds itself in place with a funding rate — a periodic payment exchanged between longs and shorts, sized by the spread between the contract's mark price and an index price. If the perp trades rich to the index, longs pay shorts. If it trades cheap, shorts pay longs. Hold it long enough and the carry forces the contract back toward the reference.
That mechanism works in crypto for one boring reason: BTC and ETH trade 24 hours a day across dozens of venues, with deep, continuously observable spot books. Index construction is a solved, unglamorous problem. Nobody argues about the ETH index at 3 a.m. because there is always a print.
Equities are the opposite animal. A cash equity trades roughly 6.5 hours a day on a primary listing venue. It halts on news. It goes ex-dividend. It splits. It gets acquired for cash, for stock, or for some blend of the two. It gets delisted and it ceases to exist. Every one of those events changes the fair value of a derivative written on it, and every one of them must be encoded in a written contract specification before the first trade prints. Not after. Before.
America has been down this road. Under the Shad-Johnson Accord of 1982, single-stock futures were banned outright in the United States — the SEC and the CFTC carved the world between them and equities derivatives stayed on the securities side. The Commodity Futures Modernization Act of 2000 lifted the ban. OneChicago launched single-stock futures in 2002 with genuine institutional backing and an exchange structure built for it. The volume never arrived. The venue wound down in 2020.
That is the base rate the current narrative ignores. The last time America legalized single-stock futures, the product died of indifference — not of prohibition. The binding constraint was never permission. It was market-making economics, demand, and the fact that a listed SFP carries an expiry date, which gives you a convergence event, while a perpetual carries none.
One more piece of context that matters for the rest of this piece: single-stock futures in the US sit under joint CFTC and SEC jurisdiction. That is not a footnote. It is the load-bearing wall.
Now the bear market frame. In a drawdown, capital does not evaporate; it pools. It parks in stablecoins and waits for a story that justifies movement. That makes a bear market the most narratively efficient market on earth — every four-line wire item gets repriced as a sector thesis inside a day. So the question is not whether this thing is bullish. The question is whether it is real, and if it is real, whether it is real for someone who has to still be solvent in six months.
That is why four lines deserve this much space.
Core: Six Things the Filing Doesn't Say, and Why Each One Is the Whole Product
The funding rate has no anchor to catch
A crypto perp's funding rate points at a spot index that refreshes every second. An equity perp's funding rate has to point at something that, for roughly seventeen and a half hours a day, does not exist as a live print.
So what does it point at? Options: the official closing auction print, the last NBBO midpoint before the bell, a synthetic fair value constructed from ADRs, index futures, and overnight futures on the same name, or some weighted blend. Each of these has failure modes. The closing print is a single auction and gets gamed. The NBBO midpoint at 15:59:59 is stale by the time Asian hours open. The synthetic build requires an oracle making a judgment call about corporate news that has not yet been priced.
Then there is the weekend. A stock closes Friday. Your perp keeps trading. If the company pre-announces on Saturday, the perp prints a new level while the index is frozen on Friday's close, and funding goes haywire for two days. Whoever is on the wrong side of that gap gets bled through a mechanism they cannot see.
For an equity perp, the index is the product. Everything else is plumbing. The item does not mention an index at all.
Corporate actions turn a smart contract into an operations desk
Take a stock yielding three percent, paying quarterly. A long perp holder must receive the economic equivalent of that dividend, or the perp trades permanently below spot and shorts eat the difference with no way to hedge it. With no expiry, there is no delivery to true it up. So the adjustment has to run through the funding stream — which is a cash transfer, triggered by a corporate calendar, executed by the venue.
That is an operational function. Not a smart contract, unless somebody has built an oracle that reads dividend declarations and special distributions and return-of-capital events reliably. That oracle does not exist at production quality, and the reason is that corporate actions are genuinely messy: special dividends, spin-offs, rights issues, merger consideration in mixed cash and stock.
Splits are worse. A split resets the contract multiplier on a live contract with open interest. You either adjust every position or you halve everyone's notional overnight. Mergers either settle the contract early at a defined value or roll it into the acquirer — both need a written rule. Delistings need a written rule. And halts: what happens to your open interest when the underlying is LULD-halted on a Friday afternoon and your perp is still matching orders at 2 a.m. in Singapore?
I audited the EigenLayer withdrawal queue logic back in late 2023. The flaws that actually threatened capital were never cryptographic. They were accounting assumptions about who owed what to whom, and under what conditions the queue could be gamed. Same lens here. The dangerous part of an equity perp is not the matching engine. It is the corporate action clause nobody has published.
The leverage ceiling kills the pitch
The most direct US precedent for retail leverage caps is the CFTC's retail forex framework — 50:1 on major pairs, 20:1 on minors. Read the political temperature on a US-retail 10x stock perp and the answer is not a design choice, it is a vote. It will be capped low on the first pass, if it passes at all.
Now sit with what that does to the product. If leverage lands at 2x, the entire stated use case — leveraged equity trading — evaporates. You can already get 2x on any margin account at any broker. You get real dividends, tight spreads, real corporate action handling, and a clearing structure that has been stress-tested for decades. The perp adds nothing except a 24-hour clock and a funding leak.
"DeFi" is a claim about control, not a sticker
Strip it down. What makes a derivative DeFi: permissionless access, non-custodial collateral, on-chain settlement, composability with other protocols, and no gatekeeper deciding who is allowed to trade.
Now describe a CFTC-regulated product for US customers: KYC, AML, an introducing broker, a futures commission merchant, a clearing house, trade surveillance, position reporting. Those two sentences cannot both be true. They are architecturally exclusive.
So the item is describing one of three things, and doesn't tell you which:
A CeFi product wearing a DeFi label.
A genuinely on-chain product that cannot be CFTC compliant.
A hybrid — a regulated wrapper fronting a permissioned chain, which is "DeFi" in name only.
That single missing fact is worth more than every other sentence in the filing.
And note the direction of the effect. Regulators in 2026 are not naive about decentralization marketing. Four years of enforcement have taught every division what a "decentralized" label is worth when you pull the admin keys. The DeFi framing does not soften approval friction. It raises it.
Four gates, not one
Filed. Accepted. Published for comment or self-certified. Live.
Those are four separate states and the wire item collapses them into one verb. Worse: most US-listed derivatives never get "approved" at all. Designated Contract Markets self-certify under the CFTC's Part 40 framework, filing terms and conditions and either certifying compliance or requesting approval only for genuinely novel products. Foreign venues take the Foreign Board of Trade route, which requires comparable home-country regulation and a registration process that takes years.
So "asked the CFTC for approval" is either a simplification for the headline or a misunderstanding of the regime. Both are tells.
Pair that with jurisdiction: single-stock futures are CFTC and SEC jointly under CFMA. A source naming only the CFTC has either compressed the story or does not know the structure. Either way, downgrade it.
And the entity is New Zealand-registered, serving US clients. That means DCM registration — capital, compliance, books and records, surveillance — or FBOT with comparable NZ FMA oversight on the home side. Cross-border coordination, multi-year, eight figures before the first tick prints. On a venue whose brand nobody in my group chat could place.
What is actually scarce, and who is actually threatened
Offshore venues already list tokenized equities and equity-linked perps. They work precisely because they do not have to care about Part 40. So if O.G. Com is chasing the compliant version, it is chasing a license, not a product. The moat being sold here is a permission slip.
Who gets hurt if it works? Not Coinbase. The offshore equity-perp books, marginally. The perp DEXs whose growth story depends on staying ahead of regulated competitors — GMX, dYdX, the Hyperliquid-class venues — see a narrative competitor, not a real one, because the user bases barely overlap. US-compliant retail and permissionless on-chain leverage are two different populations with two different risk tolerances.
Traditional derivatives exchanges — CME, Cboe — face a small, long-dated entrant. Tokenized-equity and RWA baskets get a mild narrative lift. That is the entire transmission map, and it is thin.
What my own book says about this
January 2024, ahead of the spot Bitcoin ETF approval. We built an arbitrage bot in Python on AWS to capture the spread between ETF NAV and the Coinbase spot print. Fifty thousand dollars deployed, twelve percent in two weeks, near-zero directional exposure. The most profitable two weeks of that quarter.
Here is the part people get wrong when they cite that trade. We did not trade the approval headline. We traded the flow, which was a mechanical consequence of a structure already fully specified: creation and redemption mechanics, cash versus in-kind, custody, trading hours, settlement cycle. The specification existed before we wrote a line of code. We modeled it, sized it, and executed.
Without the spec, there is no trade. Without a spec, there is nothing to hedge. The only expression left is directional — buy some adjacent asset and hope the headline spreads. That is not a trade. That is a lottery ticket with a story stapled to it.
Same lesson from last March, when I ran autonomous agents on the Berachain testnet against other AI-driven funds. Five thousand-plus micro-transactions, Sharpe of 3.2. The edge was never the model. It was the kill switch — the human-set risk parameters that stopped the agents from over-leveraging into a flash crash. An equity perp with undisclosed corporate-action handling is an agent with no kill switch. It looks brilliant until the first split, and then it is unpriceable.
Contrarian: The Trade Nobody Is Pricing Is the One That Never Happens
Applications that get withdrawn, returned for deficiency, or shelved indefinitely do not generate a second headline. The wire reports the submission and never reports the abandonment. That asymmetry is the entire information market here.
Run the numbers. Probability of a live, US-retail-accessible, levered single-stock perpetual in the next three years: call it ten to fifteen percent, and that is being generous to the optimists. Probability that this narrative moves some adjacent asset in the next two weeks: high. Multiply those together and what you are holding is not a product. It is a cheap option on regulatory attention, sold as a product announcement.
The deeper irony is that the two audiences being addressed are structurally incompatible. DeFi users want no gatekeeper. The CFTC requires a gatekeeper. The label makes the compliance desk's job harder while winning applause from people who will never be permitted to use the thing. Two mutually exclusive crowds, one set of four lines.
And the property that should worry anyone modeling this: with no public product specification, there is no way to falsify the claim. Unfalsifiable claims are the most durable narratives in a bear market, because nothing ever arrives to kill them. They just quietly stop being mentioned.
Takeaway: Three Gates, In Order
Does a submission exist in CFTC public records or the Federal Register — a self-certification, an approval request, a docket number? Until that appears, the item is a headline, not a market structure.
Which agency leads? Single-stock futures are CFTC and SEC jointly. A CFTC-only story gets downgraded on contact.
Any specification at all — funding interval, index methodology, leverage cap, corporate action policy. That is the gate that converts a narrative into something a desk can price.
Track those three. Size nothing. The scarcity being sold is a license. The thing being purchased is an idea about a license. In the sprint, hesitation is the only real cost — but so is buying settlement rules that have not been written yet.