Commitment of Traders report: what the COT data shows, and what it does not

The Commitment of Traders report, officially the Commitments of Traders (COT) report, is the weekly release in which the CFTC, the US futures regulator, breaks down the open interest of each futures market between broad categories of traders. It says who held what on Tuesday, not what price will do next. This guide goes through the CFTC's own definitions, the four report formats, one real line read step by step, and the limits the regulator itself writes down.

• A Tuesday snapshot: positions are taken as of Tuesday and generally released on Friday at 3:30 p.m. Eastern Time.
• Four formats: Legacy, Supplemental, Disaggregated and Traders in Financial Futures, each with its own categories.
• Traders are classified, not trades: the category comes from what the trader reports, not from the reason for each position.
• Not a signal: the CFTC does not analyze the data, and the studies read here find little forecasting power in it.

The COT has a flattering reputation: it is said to show what the 'smart money' is doing. The document itself is more modest. It is a table of aggregated positions, three days old when it comes out, whose categories rest on a self-reported statement.

This guide tells you neither what to buy nor when. It sticks to what the CFTC publishes and explains, and to what two published studies measured, so that you can read the report without making it say more than it does.

What is the Commitment of Traders report? The CFTC's definition

The CFTC (Commodity Futures Trading Commission) oversees the US markets for futures contracts and options on futures. The CFTC Glossary defines the Commitments of Traders Report as a weekly report from the CFTC providing a breakdown of each Tuesday's open interest for markets in which 20 or more traders hold positions equal to or above the reporting levels established by the CFTC.

Two words in that definition need explaining. Open interest, according to the COT report's Explanatory Notes, is the total of all futures and/or option contracts entered into and not yet offset by a transaction, by delivery or by exercise; the aggregate of all long open interest is equal to the aggregate of all short open interest. The reporting level is the position size at or above which a trader has to be reported to the regulator every day.

The CFTC also says what the report is for and what it does not do with it. On the Commitments of Traders reports page, it explains that it publishes the reports to help the public understand market dynamics. Further down, asked how to use the data, its staff answer plainly: the CFTC does not analyze the data nor make recommendations on it.

Where the numbers come from

Traders do not send in their positions themselves. According to the Explanatory Notes, clearing members, futures commission merchants and foreign brokers, collectively called reporting firms, file daily reports with the Commission. Those reports show the positions of traders above the reporting level. As soon as a trader reaches that level in any single futures month, the trader's entire position in that commodity is reported, regardless of size in the other months.

The same notes give an order of magnitude: the aggregate of all traders' positions reported to the Commission usually represents 70 to 90 percent of the total open interest in any given market. The rest appears in the 'Nonreportable Positions' column, derived by simple subtraction. For those positions, the CFTC specifies, the number of traders involved and the classification of each trader are unknown. They are often called 'small traders', but the report does not say who they are.

A market only appears in the report if it has enough large participants. When the number of reportable large traders drops below 20, the reports page states, the market no longer appears in the COT, and it is added again once it has reached 20 or more.

When the COT report is released

The general rule fits in one sentence from the CFTC: the report is published each Friday at 3:30 p.m. Eastern Time, using the data from the immediately preceding Tuesday. Positions are calculated as of the close of business on Tuesday, and it takes three days to process the data. The CFTC's release schedule specifies that federal holidays may delay release by one or two days.

That pace is recent on the report's own timescale. The CFTC's 'About the COT Reports' page traces its antecedents back to 1924, with a first annual report by the US Department of Agriculture on hedging and speculation. The data became monthly as of June 30, 1962, for 13 agricultural commodities. Publication moved to mid-month and month-end in 1990, to every two weeks in 1992, and to weekly in 2000. The report has been freely available on the CFTC's website since 1995.

The practical consequence: when you read the report on Friday evening, you are looking at Tuesday's positions. Everything traded on Wednesday, Thursday and Friday is not in it. The CFTC also states that, because of resource constraints, it is only able to release this report once a week, and that historical data is not updated once published.

The four report formats and their categories

The CFTC publishes four main reports from the same position data. They do not differ in the underlying numbers but in the way traders are sorted. Category names are given as they appear in the report.

The 'futures and options combined' version converts options into futures equivalents using delta factors supplied by the exchanges. The Explanatory Notes give the example of a trader holding a long put position of 500 contracts with a delta factor of 0.50: that trader is considered to be holding a short futures-equivalent position of 250 contracts.

Commercial and non-commercial: what the categories cover

In the Legacy format, a reportable trader is classified either as 'commercial' or 'non-commercial'. According to the Explanatory Notes, all of a trader's reported positions in a commodity are classified as commercial if the trader uses futures contracts in that particular commodity for hedging as defined in CFTC regulation. A trader generally gets that status by filing a statement, on CFTC Form 40, that it is engaged in business activities hedged by the use of the futures or option markets. A trader may be commercial in some commodities and non-commercial in others, but not both in the same commodity.

The Disaggregated format, published since September 4, 2009, splits each of those two families in two. The Disaggregated Explanatory Notes define a 'producer/merchant/processor/user' as an entity that predominantly engages in the production, processing, packing or handling of a physical commodity and uses the futures markets to manage or hedge risks associated with those activities. A 'swap dealer' deals primarily in swaps for a commodity and uses the futures markets to manage or hedge that risk. A 'money manager' is a registered commodity trading advisor (CTA), a registered commodity pool operator (CPO) or an unregistered fund identified by the CFTC, engaged in managing and conducting organized futures trading on behalf of clients. Every other reportable trader is placed in 'other reportables'.

For financial contracts, the Traders in Financial Futures report, announced on July 22, 2010, sorts traders differently. The TFF Explanatory Notes distinguish 'dealers/intermediaries', the sell side of the market, typically large banks and dealers in securities, swaps and other derivatives; 'asset managers/institutional', institutional investors such as pension funds, insurance companies and mutual funds; 'leveraged funds', typically hedge funds and various types of money managers; and 'other reportables', which mostly use markets to hedge business risk. Those notes warn that the TFF is not a disaggregation of the Legacy format: a trader in one of these four categories may be drawn from either the commercial or the non-commercial category.

The Supplemental format adds, in selected agricultural markets, an 'Index Traders' category drawn from the other two. The Explanatory Notes describe these traders as generally replicating a commodity index by establishing long futures positions and then rolling those positions forward from future to future using a fixed methodology.

How to read a line of the report: wheat on September 29, 2026

Let us take a real line, copied as it stands. In the Chicago Board of Trade 'futures only' Legacy report dated September 29, 2026, soft red winter wheat (WHEAT-SRW, code 001602) shows an open interest of 483,142 contracts of 5,000 bushels each.

Check that the line adds up. Long side: 115,160 + 156,942 + 176,178 + 34,862 = 483,142. Short side: 131,622 + 156,942 + 162,727 + 31,851 = 483,142. Both totals equal open interest, as its definition requires: every contract has a buyer and a seller. The 'spreads' column counts on both sides.

Spreads. According to the Explanatory Notes, this column measures the extent to which a non-commercial trader holds equal long and short positions. The CFTC's example: a trader with 2,000 long contracts and 1,500 short contracts appears for 500 in 'long' and for 1,500 in 'spreading'. Here, 156,942 contracts, or 32.5% of open interest, are therefore positions that offset each other within the same trader.

Net positions. The report does not print them; they are calculated. Non-commercials: 115,160 − 131,622 = −16,462 contracts, hence a net short position. Commercials: 176,178 − 162,727 = +13,451. Nonreportable: 34,862 − 31,851 = +3,011. The three add up to zero, by construction: one group is only net short because the others are net long by the same amount.

The change over one week. The 'changes' line compares with the report of September 22, 2026. Non-commercials hold 4,967 fewer long contracts and 4,135 more short contracts: their net position goes from −7,360 to −16,462 contracts, while total open interest barely moves (−137). The report says this shift took place between two Tuesdays. It says neither on which day, nor at what price, nor why.

The number of traders. The report counts 424 reportable traders in total, yet adding the five columns gives 143 + 114 + 137 + 96 + 122 = 612. That is not an error: the Explanatory Notes specify that a trader is counted in each category in which the trader holds a position, so the sum will often exceed the total. The long formats add the percent of open interest held by the largest four and eight traders.

The limits of the report, as written by the CFTC

The most useful caveats do not come from commentators; they are in the regulator's own documents.

• Classification is self-reported. A trader's category is based on the predominant business purpose self-reported on Form 40, subject to review by CFTC staff for reasonableness.
• The trader is classified, not the position. The CFTC writes that its staff does not know specific reasons for traders' positions. A producer classified as 'producer/merchant/processor/user' has all of its positions in that commodity filed there, regardless of whether the position is for hedging or speculation.
• A category mixes activities. Some traders classified as 'swap dealers' also engage in commercial activities in the physical commodity, and the reverse exists. A hedge fund filed under 'leveraged funds' might be using foreign exchange futures to hedge another position.
• The same participant can change boxes. A trader can have different classifications for different commodities, an organization may have several of its trading entities classified separately, and a classification may change over time.
• History is reconstructed. The CFTC does not maintain a history of large-trader classifications: older data was classified using recent categories, which, it says, diminishes the data's accuracy as it goes further back in time.
• Names stay confidential. The CFTC does not publish information on how individual traders are classified, for legal reasons.
• A three-day lag. Positions are Tuesday's, publication is on Friday, and nothing is published in between.

Does the COT report predict prices? What two studies say

It is the use most often attributed to the report, and the one research supports least. The study by Sanders, Irwin and Merrin, published in 2009 in the Journal of Agricultural and Resource Economics, covers ten agricultural futures markets. According to its abstract, Granger causality tests show very little evidence that traders' positions are useful in forecasting returns. However, there is substantial evidence that traders respond to price changes, and non-commercial traders display a tendency for trend following. The authors conclude that their results generally do not support the use of the COT data in predicting price movements in those markets.

An earlier piece of work is more nuanced without being more encouraging. Changyun Wang's article published in 2003 in the Journal of Futures Markets examines speculators and hedgers in fifteen US futures markets. Its abstract states that trades of speculators are positively correlated with subsequent abnormal returns, and those of hedgers negatively. But it adds at once that it does not appear that speculators possess superior forecasting power: the author attributes the relation to hedging pressure effects instead.

These two studies do not close the subject. They cover specific markets and periods, and other work exists. They are nonetheless enough to set one idea aside: neither validates a mechanical rule in which one category's position would point to the direction of the next move. The fact that non-commercials follow the trend rather suggests the opposite: their positions partly reflect what price has already done.

When the report does not come out

The COT depends on the US administration operating and on the reporting chain. Both have already broken down.

• Government shutdown, 2018-2019. According to the CFTC press release of January 30, 2019, publication was suspended during the lapse in appropriations. The last report dated from December 21, 2018, publication was expected to resume on February 1, 2019, then at a rate of two reports a week, in chronological order, until the normal schedule was restored.
• Incident at a service provider, 2023. The CFTC press release of February 16, 2023 reports that a cyber-related incident at ION Cleared Derivatives, a third-party service provider of cleared derivatives processing, was preventing reporting firms from submitting timely and accurate data. The report due on February 17 was postponed, resumption was announced for February 24 with the report originally scheduled for February 3, and the backlog was expected to be cleared by mid-March.
• Lapse in federal appropriations, 2025. The CFTC's special announcements on the COT report state that the processing and publication of the data were interrupted from October 1 to November 12, 2025. A CFTC press release on the backlogged reports then announced accelerated publication, still in chronological order, to eliminate the backlog by December 29, 2025.

During those weeks, positions kept moving and nobody could read them. A method that relies on a weekly data point has to plan for the case where it is missing. And these interruptions rarely fall at random: a government shutdown or a breakdown in the reporting chain belongs to those moments when the market leaves the ordinary, like the ones described in the guide to the black swan event.

What you can do with it without telling yourself a story

The COT remains a public, free and rare data point: few markets say every week who holds the positions. It is useful for describing a market. Knowing whether funds are heavily present, whether open interest is growing or shrinking, whether the structure changes from one week to the next, is context.

If you want to do more with it, the only honest route is to test it like any other idea. Write the rule down before looking at the result, and respect the lag: a report dated Tuesday is only known on Friday after 3:30 p.m. Eastern Time, so a test that uses it from Tuesday cheats without meaning to. That is one of the traps described in the guide on how to backtest a trading strategy. Until the measurement has been made on your own trades, you do not have a trading edge, you have a hypothesis.

Also keep in mind what the report does not contain: no price, no time, no reason. A net short position of funds in wheat does not say whether they sold on Monday or the week before, nor whether they already bought back on Wednesday.

The COT in your trading journal

Tradoshi does not publish the COT report and does not interpret it: the source is the CFTC's website, and it is free. A journal serves another purpose. If you look at the COT before trading, note it in the trade concerned, along with what you concluded from it. After a few dozen trades, your own statistics will say whether that reading changes anything in your results, which no article can tell you in your place. For keeping a journal on these markets, see the guide to the futures trading journal.

Frequently asked questions