On October 16, in the U.S. District Court for the District of Columbia, the Commodity Futures Trading Commission filed a reply brief that argues against CME Group using CME's own monthly volume reports. Twelve crypto futures products. Total contracts: 2,660,960 in May, 3,282,530 in September. A net gain of 621,570 contracts, or 23.4%, across four months.
The brief draws one conclusion from that table: "cryptocurrency perpetual futures have had no adverse impact on its crypto-futures suite."
Most standing briefs argue doctrine. This one argues about denominators. An anomaly is just a story waiting to be read, and the anomaly here is that two parties are examining an identical dataset and reaching opposite conclusions about injury. That is not a legal disagreement. It is a disagreement about what a net aggregate can prove.
Context: what is actually being litigated
The procedural spine is short. In May, the CFTC cleared perpetual futures for U.S. exchanges. In June, CME sued the agency and Chairman Michael Selig, arguing that perpetuals are swaps under the Commodity Exchange Act and therefore fall outside the relief granted. In September, the agency moved to dismiss. The October 16 filing supports that motion. The court has not ruled.
The doctrinal hinge is Article III standing. CME must allege a concrete injury, not a speculative one. Competitor standing is a real doctrine โ the CFTC concedes as much, writing that it "is not some kind of Article III unicorn" โ but it requires an imminent competitive threat, and imminence is where the agency attacks.
Its weapon is CME's July 22 second-quarter earnings script. The CEO said the company has "the full technical and operational capabilities to launch perpetual futures" but has "not heard demand from our customers." He also said that "[t]hrough the first half of 2026, 94% of our volume originated from institutional customers."
The agency's line: "CME cannot have it one way with the Court and another way with the investing public on earnings calls."
Two secondary points matter for the data. First, the CFTC argues that reclassifying perps as swaps would not remove them from the market, because designated contract markets can list both โ Kalshi has held that status since 2020. Second, on October 5, CFTC staff issued no-action relief permitting exchanges to convert perpetual-style security index futures into true perpetuals. That is the regulator building a parallel track while the courtroom argument continues.
Core: decomposing the 621,570
Here is where I stop reading the brief as a legal document and start reading it as a data release. In January 2024 I built a dashboard tracking daily net inflows across IBIT, FBTC, and GBTC, correlating them against Coinbase and Binance order book depth. The lesson from that build has held ever since: a net figure is a subtraction, and subtractions conceal composition.
The 621,570 net gain is not distributed evenly across twelve products. The figures surfaced from that table identify only a fraction of the twelve line items, but those disclosed components account for roughly two-thirds of the increase.
- Micro Bitcoin futures: +240,369 contracts (38.7% of the net gain)
- Micro Ether futures: +176,496 contracts (28.4%)
- Link futures: โ340 contracts (the only product below its May level)
Micro Bitcoin and Micro Ether together contributed 416,865 contracts โ 67.1% of the total increase. Both are the smallest-denomination products in the suite. Both are the retail-adjacent instruments.
That composition is the most important number in the filing, and neither party has addressed it directly. The CFTC uses the aggregate to argue absence of harm. CME calls the same figures "at best an incomplete picture, and at worst intentionally misleading," and points to "increased competition for retail business."
Both are arguing past the decomposition. If two-thirds of your growth comes from micro contracts, your suite is not demonstrating immunity to retail competition. It is demonstrating that your retail-adjacent products are the ones still growing. Those are different claims, and only one of them is being briefed.
Now apply the CEO's own 94/6 split. If 94% of volume is institutional, the retail segment CME says it will lose is, at maximum, 6% of its book. That is the arithmetic ceiling on the injury CME has described. A 6% exposure, on a product line that grew 23.4% in four months, is a difficult thing to call imminent and concrete. Every transaction leaves a scar; I map the wound โ and this wound has a measurable maximum size.
The Link futures line is the weak point in the CFTC's exhibit, and it deserves flagging. It is the only declining product at โ340 contracts. But the brief does not disclose the May base for Link futures. Without a denominator, โ340 is uninterpretable: it could be a rounding artifact on a thin book, or a material contraction. A change without a base is not evidence. It is a number looking for a context. Any analyst citing that line in either direction is overreading the filing.
One methodological note on the tape. The market is sideways. That matters for inference. In a range-bound market, aggregate volume growth is less easily explained by price beta โ which strengthens the CFTC's inference that perps did not cannibalize the suite. In a trending market, the same +23.4% would prove almost nothing. The regulator received a favorable control condition, whether or not it noticed.
What would a defensible test look like? It would need a counterfactual: per-product volumes in a jurisdiction where perps are licensed and domestic, benchmarked against one where they are not, over the same window. The brief offers a before-and-after on a single venue. Before-and-after without a control cannot distinguish "no adverse impact" from "adverse impact smaller than beta." That is not a criticism of the CFTC's conclusion. It is a statement about the resolution of its instrument.
Contrarian: the quote the brief did not finish
The CFTC's strongest paragraph is also its most fragile, and the fragility sits inside its own quotation.
The CEO said he had "not heard demand from our customers." The brief transposes that onto CME's retail-competition claim. But the same script states that 94% of volume came from institutional customers. Read together, "our customers" means institutions. The CFTC converted a statement about institutional demand into a rebuttal of a claim about retail competition. That is a category substitution, not a contradiction โ and a court that reads the transcript rather than the excerpt may notice.
The counter-argument is that CME's own filing characterizes the harm as retail competition, so CME chose the frame. Fair. But frame selection is not evidentiary nullity, and the "one way with the Court, another with the public" line is rhetorically stronger than it is analytically.
There is also a causation problem the brief does not engage. Volume growth after May is consistent with three hypotheses: no harm, harm offset by market-wide beta, or harm deferred because perps have not yet scaled domestically. A four-month window on twelve products cannot separate them. Correlation is not causation in either direction โ and the CFTC is using correlation here to prove a negative.
Takeaway
I do not predict the future; I trace the past. The trace says the standing motion will turn on imminence, and the data will not settle it. What will settle it, faster, is the October 5 no-action relief: if designated contract markets begin converting perpetual-style security index futures into true perpetuals, the question of whether CME faces imminent competition becomes observable rather than argued.
The signal to watch is not the aggregate. It is per-product month-over-month contract counts for the two micros and for Link futures, published in CME's next monthly report. If Micro Bitcoin keeps absorbing a third of net growth while the institutional book stays flat, the CFTC's table keeps working. If the micros roll over, the brief's central exhibit becomes a four-month artifact. The pattern emerges only after the dust settles โ and the dust has another month to fall.