The COT Report Explained: What the CFTC Data Really Shows

The Commitments of Traders report is one of the few genuinely public datasets in a market where almost everything else is private. That scarcity gives it authority, and the authority is often misplaced. This is also the only published positioning data available for currencies, which matters when assessing claims made about institutional order flow that are drawn from price charts alone.

Most guides teach one version of the report, describe three categories of trader, and suggest that when the speculative category reaches an extreme a reversal is near. Very little of that survives contact with what the data actually is.

What follows describes the instrument itself: who publishes it, what it counts, which of the four reports a currency trader should be reading, and which conclusions the data cannot support no matter how it is charted.

What the Commitments of Traders Report Actually Is

The Commitments of Traders report is published by the United States Commodity Futures Trading Commission, the federal regulator of the American derivatives markets. It is a regulatory disclosure, not a market-analysis product.

The CFTC describes the reports as providing a breakdown of each Tuesday’s open interest for futures and options on futures markets in which 20 or more traders hold positions equal to or above the reporting levels established by the CFTC.

Every clause in that sentence constrains what the data can tell you. It is a snapshot rather than a flow. It counts open interest, meaning contracts still outstanding, not volume traded. And it covers only those markets that clear the twenty-trader condition.

That last point is the one most often missed. Inclusion is conditional. A contract market that falls below twenty reportable traders stops appearing, and its disappearance says something about market structure rather than about sentiment.

Futures and Options, Not the Spot Market

This is the single most important thing to understand before drawing any conclusion about a currency pair, and it is the point the popular guides pass over fastest.

The report covers futures and options on futures. Spot foreign exchange is a decentralised over-the-counter market with no central exchange, no consolidated tape and no regulator collecting a public position census. There is no Commitments of Traders report for spot.

So when an analysis says that speculators are heavily long the euro, what has actually been observed is positioning in a euro futures contract listed on a US exchange. That is a real instrument with real participants, and it is not the same instrument as the pair on your platform.

Currency futures and spot rates are closely related, because arbitrage links them. But the futures contract has a fixed size, a delivery date, exchange margin and a different participant mix, dominated by institutions rather than by the global bank flow that sets the spot rate.

The honest formulation is that the report tells you about positioning in a related instrument, from which something may be inferred about the broader currency. That is a weaker claim than most sources make, and it is the accurate one.

The contracts behind those positions are dated and standardised by the exchange. Our page on currency futures sets out how they are sized and when each one stops trading.

The Four Reports and Why the Choice Matters

Most explanations of the Commitments of Traders data behave as though there is one report. The CFTC publishes four main reports, and they do not use the same trader categories.

The four are Legacy, Supplemental, Disaggregated and Traders in Financial Futures. Legacy reports are broken down by exchange, and the Supplemental report covers thirteen selected agricultural commodity contracts.

The Disaggregated reports cover agriculture, petroleum and products, natural gas and products, electricity, metals and other physical contracts. Traders in Financial Futures covers financial contracts.

The familiar three-way division into commercial, non-commercial and non-reportable belongs to the Legacy report. It is the oldest presentation and the one that populates most third-party charts and tutorials.

A reader who learns only that division, and who trades currencies, has been taught the categories of a report that is not the most appropriate one for the contracts they care about. Nothing in the typical guide flags this, because the report being used is usually never named.

Which Report to Read for Currencies

Currency contracts sit in the Traders in Financial Futures report, which the CFTC describes as covering financial contracts. It divides reportable open interest into four categories rather than the Legacy three.

Those categories are Dealer/Intermediary, Asset Manager/Institutional, Leveraged Funds and Other Reportables. The CFTC describes the dealer and intermediary group as sell-side participants that earn commissions selling financial products, capture bid and offer spreads and accommodate clients, and that tend to run matched books or offset risk across markets and clients.

The asset manager and institutional category comprises institutional investors including pension funds, endowments, insurance companies, mutual funds and portfolio or investment managers. Leveraged funds and other reportables account for the remaining reportable positions.

There is a structural difference in this report worth knowing about, and it changes how the categories behave. The CFTC states that in the Traders in Financial Futures report traders are classified in the same category for all commodities.

In the other reports, by contrast, traders are able to report business purpose by commodity, and can therefore carry different classifications in different markets.

The practical consequence is that a category label here describes the firm, not the intent behind a particular position. An institution classified as an asset manager is an asset manager everywhere in the report, whatever any individual trade is for.

Reading the Categories Without Over-Reading Them

The common interpretation runs that one category is the smart money, another is the trend-following crowd, and that when the crowd reaches an extreme the move is nearly over. The categories will not carry that weight.

A category is a classification of who holds a position, based on the business the firm is in. It is not a statement of conviction, of expected holding period, or of whether the position is a directional bet at all.

A large short in one category may be a hedge against an exposure held somewhere else entirely, and the offsetting exposure is invisible because it sits outside the futures market. The report shows one leg of what may be a two-leg arrangement.

This is why positioning extremes are so much easier to identify after the fact than in advance. An extreme is only visible as an extreme relative to a history that you choose, and nothing in the data tells you that the previous high was the ceiling.

The defensible use is narrower and duller. The report describes how the composition of open interest has changed over recent weeks, which is context. Treating it as a timing signal asks a positioning census to do a job it was never built for, in the same way that leading and lagging indicators are routinely asked to forecast rather than describe.

Why Non-Reportable Is Not Retail Sentiment

Every version of the report carries a non-reportable line, and it is very widely described as the small trader or the retail crowd. That description mistakes an arithmetic remainder for a measurement.

Non-reportable positions are derived rather than collected. Total reportable long and short positions are subtracted from total open interest, and what is left is reported as non-reportable. Nobody surveys those traders, and no data is gathered from them directly.

What that residual contains is everyone below the reporting level. That includes small speculators, but it also includes small commercial participants and any position that simply falls under the threshold, held by a firm that may be very large.

So the line moves for reasons that have nothing to do with small-trader conviction. If a reportable trader trims a position below the reporting level, the residual grows without a single new participant entering the market.

Reading it as a contrarian retail gauge stacks two assumptions that the data does not support: that the residual is retail, and that retail is systematically wrong. Neither is established anywhere in the report.

The Tuesday-to-Friday Lag and What It Rules Out

The CFTC states that the report is generally published each Friday at 3:30 pm Eastern Time, using data from the immediately preceding Tuesday of that week.

The arithmetic is worth doing explicitly, because it settles a whole class of proposed uses. At the moment of release the snapshot is three days old. By the following Tuesday, when a fresh snapshot is taken but not yet published, the newest figure a trader can see is a week old.

Three sessions is a long time in a currency market. Any position that was going to be reduced in reaction to a mid-week event has already been reduced by the time the file appears, and the file will not show it for another week.

That rules out any use that depends on knowing where participants are positioned now. It does not rule out slower work, such as observing how the composition of open interest has drifted over a quarter, where a three-day lag is immaterial.

The lag also interacts with revisions elsewhere in the fundamental picture. As with the figures in an economic calendar, the version of history available today is not necessarily the version that was available when the market moved.

Who Should Not Use This Data

Some tools are wrong for a job rather than merely difficult, and this is one of them for several common approaches.

Anyone trading intraday should ignore it entirely. A weekly snapshot published with a three-day delay has no bearing on a position measured in hours, and building a session-level rule around it means acting on a picture that has already expired.

Anyone trading a currency pair with no listed futures contract should also set it aside. If the contract is not in the report, or the market has fallen below the twenty-trader condition, there is nothing to read.

Anyone looking for an entry signal is asking the wrong question of it. The report has no price, no level and no timing information in it. It describes composition, and composition is context rather than instruction.

It is most defensible as one slow input among several, for a trader holding positions for weeks or longer, who already has a view formed elsewhere and wants to know whether the structure of open interest is consistent with it. Used that way it sits naturally alongside risk management rules that do not depend on it, and alongside hedging decisions that have their own logic.

If a strategy stops working when this dataset is removed, the dataset was carrying more weight than a three-day-old weekly census of a related instrument can bear. That is the test worth applying, and it is more useful than any threshold.

Frequently Asked Questions

What is the Commitments of Traders report?

It is a weekly report from the United States Commodity Futures Trading Commission that breaks down each Tuesday open interest for futures and options on futures markets in which 20 or more traders hold positions at or above CFTC reporting levels. It is a regulatory disclosure rather than an analysis product.

Does the COT report cover the spot forex market?

No. It covers futures and options on futures. The spot foreign exchange market is decentralised and publishes no equivalent positioning report, so any conclusion drawn about a spot pair from this data is an inference from a related but separate instrument that has its own contract size, delivery date and participant mix.

Which COT report should a currency trader read?

The Traders in Financial Futures report, which covers financial contracts including currencies. It divides reportable positions into Dealer/Intermediary, Asset Manager/Institutional, Leveraged Funds and Other Reportables, rather than the three-way commercial and non-commercial split of the Legacy report that most tutorials describe without naming.

How old is the data in a COT report when it is published?

Three days at the moment of release. The CFTC states the report is generally published each Friday at 3:30 pm Eastern Time using data from the immediately preceding Tuesday. By the following Tuesday the most recent published figure is a week old, which rules out any use that depends on knowing current positioning.

Does the non-reportable category show what retail traders are doing?

Not reliably. Non-reportable positions are a residual, calculated by subtracting reportable positions from total open interest, so the line contains everyone below the reporting level rather than a surveyed group of small traders. It can move simply because a large trader reduced a position below the threshold.