Why the CFTC’s Trader Classification Still Distorts the Oil Market#

“For now we see through a glass, darkly; but then face to face.” The Apostle Paul was talking about the limits of what humans can know in this life versus the clarity waiting on the other side. He might as well have been writing about the Commodity Futures Trading Commission’s Commitments of Traders report.

Every Tuesday, the CFTC releases what’s widely treated as the definitive window into the structure of commodity futures markets. The Commitments of Traders report — the CoT, in trader shorthand — breaks down the open interest on every major futures exchange by category of trader. Thousands of market participants, academics, and journalists download it, parse it, chart it, and cite it every week. It is the single most important dataset for anyone trying to answer the question: who is trading oil futures, and which way are they leaning?

It is also, as this chapter will argue, a fundamentally misleading instrument — a mirror that warps as much as it reflects.

The Binary Machine#

At its core, the CoT report runs on a binary sorting engine. Every reportable trader gets dropped into one of two buckets: commercial or non-commercial.

A commercial trader is one who uses futures to hedge a position in the physical market. An oil producer selling futures to lock in a price for next quarter’s output is commercial. A refiner buying crude futures to secure feedstock is commercial. A trading house managing physical inventory is commercial.

A non-commercial trader is everyone else — hedge funds, managed money, proprietary trading desks, anyone whose futures position isn’t tethered to a physical commodity operation.

The logic feels intuitive enough. Hedgers on one side, speculators on the other. The CoT report tells us how much open interest each group holds, and from that, analysts draw conclusions about the degree of speculative activity and whether it might be moving prices.

The problem is that this binary classification was designed in the 1980s, for a market that no longer exists.

The Swap Dealer Problem#

In the 1980s, the commercial category was populated by producers, refiners, and physical trading firms — entities whose futures activity had a clear, traceable link to their physical oil operations. A swap dealer, if one existed at all, was a bit player.

By 2006, the swap dealer had become the single largest participant in the commercial category.

How? The mechanism is simple enough to follow. A swap dealer — Goldman Sachs, Morgan Stanley, JPMorgan — strikes an over-the-counter swap with a pension fund, giving the pension fund synthetic long exposure to the S&P GSCI commodity index. To offset its own risk, the swap dealer goes long crude oil futures on Nymex. The CFTC looks at the swap dealer’s exchange position and stamps it “commercial” — because the dealer is hedging a legitimate business risk (its OTC obligation to the pension fund).

But trace the economic reality one step further. The pension fund’s OTC position is a speculative bet on commodity prices. It has no physical oil to hedge, no refinery to feed, no tanker to fill. The swap dealer is just a conduit — a pipe through which speculative capital flows into the futures market. The exchange position gets labeled commercial hedging. The economic substance is speculative demand.

By the mid-2000s, swap dealer positions had ballooned so large that they dwarfed the combined positions of every traditional commercial participant — producers, refiners, and physical traders all put together. The “commercial” category, supposedly representing the physical oil industry’s hedging needs, had been taken over by financial intermediaries whose real job was funneling speculative capital from OTC clients into exchange-traded futures.

The label said “commercial.” The reality was anything but.

How Eighty-Five Becomes Thirty-Five#

The fallout from this misclassification isn’t cosmetic. It’s structural.

Take a hypothetical snapshot of the oil futures market. Total open interest: 100 units. Fifty units belong to swap dealers (classified as commercial). Fifteen belong to traditional commercial hedgers (also classified as commercial). Thirty-five belong to hedge funds and managed money (classified as non-commercial).

The CoT report would read: Commercial = 65. Non-commercial = 35. A casual reader would conclude that speculators hold just 35 percent of the market. The physical industry — hedgers dealing in real barrels — controls the majority. The market looks fundamentally sound.

Now reclassify the swap dealers as what they actually are — conduits for speculative OTC money — and the picture flips. True hedging = 15. Speculative activity (direct plus intermediated) = 85. The market is overwhelmingly driven by financial flows, not physical hedging.

The gap between 35 percent and 85 percent isn’t a rounding error. It’s the difference between “speculation is a sideshow” and “speculation is running the show.” The CoT report’s binary sorting doesn’t just blur this distinction. It systematically produces the wrong answer.

In May 2026, Bloomberg reported that the CFTC’s trader classification system was under renewed scrutiny as part of the agency’s regulatory reform push, with critics charging that the binary framework “fails to capture the blurred roles of modern market participants.” The observation wasn’t new. It was nearly two decades old. The classification system remained unchanged.

The Self-Examination Paradox#

The misclassification problem spawns a second-order failure that may be even more damaging than the first.

The CFTC doesn’t just publish CoT data. It also uses that data to run its own research into whether speculation moves commodity prices. In a string of staff reports and interim analyses — most notably during the 2008 price spike — the agency examined the link between non-commercial positions and oil price movements. Its conclusion, stated again and again, was that the evidence didn’t support the claim that speculation was driving prices.

The conclusion was framed as rigorous, data-driven analysis. In a narrow technical sense, it was. The regressions were competently run. The statistical methods were standard.

But the data was wrong.

If the commercial category is systematically absorbing speculative positions — if swap dealers channeling pension fund index bets get counted as hedgers — then the “non-commercial” variable in the CFTC’s regressions is missing most of the speculative activity it’s supposed to capture. Running a correlation between an incomplete measure of speculation and price movements, then finding no significant relationship, doesn’t prove speculation is harmless. It proves the yardstick is broken.

This is the self-examination paradox: the CFTC used its own flawed data to investigate whether its data was capturing reality, and concluded — based on that same flawed data — that everything was fine.

It’s like a hospital testing its thermometers by checking the temperature of a room with a broken thermostat, noting that the readings match the thermostat, and declaring the thermometers accurate.

The Distant-Month Migration#

There’s one more dimension to the data problem worth unpacking, because it reveals not just the scale of the misclassification but its direction.

When the CFTC finally updated its co-integration study in 2008, it included data on how positions were distributed across different contract maturities — near-month, mid-month, and distant-month futures. The findings were striking.

Between 2000 and 2008, swap dealer positions in distant-month contracts — those expiring six months, twelve months, or further out — surged as a share of total open interest. Traditional commercial hedgers, meanwhile, stayed clustered in the near-month contracts, where their physical operations actually needed price protection.

This migration toward the back end of the curve is exactly what you’d expect from index fund money flowing through swap dealers. Index funds hold long positions across the entire futures curve, not just the front month. As their capital swelled from $15 billion to $260 billion, the swap dealers hedging their OTC exposure had to buy futures across the full maturity spectrum. The distant-month buildup was the fingerprint of index fund infiltration — visible even in the CFTC’s own data, if anyone bothered to look.

Few did. The data was there, but the classification system rendered it invisible to standard analysis. Swap dealer positions in distant-month contracts were filed under “commercial.” The infiltration pipeline’s footprint was hiding in plain sight, camouflaged by a label that had stopped meaning what it said.

The Broken Mirror#

This chapter borrows its title from Paul’s epistle, but the metaphor stretches well beyond theology. We see the oil market through a glass, darkly — not because the data doesn’t exist, but because the lens we look through was built for a simpler era and never upgraded to match the market’s transformation.

The CoT report’s binary classification worked well enough when commercial meant physical and non-commercial meant speculative. It broke down the moment swap dealers — financial middlemen with no physical oil of their own — became the biggest players in the commercial category. At that point, the mirror started to distort. And the distortion wasn’t random. It ran in one consistent direction: understating the speculative share of the market and overstating the hedging share.

Every study built on uncorrected CoT data was peering at the market through the same warped glass. Every Congressional testimony that leaned on the CFTC’s own research was relying on conclusions drawn from data the agency itself knew — or should have known — was structurally incomplete.

The volatility that erupted after the UAE’s departure from OPEC in early 2026 dragged these transparency gaps back into the spotlight. Analysts pointed out that CoT reports offered only a partial picture of true market positioning, with significant activity concealed in OTC channels. The observation was correct. It was also, by this point, a cliché — a criticism so well-worn it had lost its power to spark action.

We still see through a glass, darkly. The question is whether we’ll ever choose to replace the glass.