Key Takeaways
- A Commitments of Traders report measures positions held on a Tuesday and is published days later. The explorer shows the observation date, the publication date and the retrieval time separately.
- Net position is long minus short. Net as a share of open interest makes weeks with different market sizes comparable.
- The percentile ranks the latest reading against weekly history up to the as-of date only. It is withheld below 52 weekly observations.
- Participant categories are CFTC reporting classifications. They describe who reports an exposure, not why they hold it.
- The built-in sample is fictional and live feeds are not connected. Your pasted rows stay in your browser.
What Is COT Positioning?
COT positioning is the reported long, short and spreading futures exposure of groups of traders, taken from the weekly Commitments of Traders program run by the Commodity Futures Trading Commission. The Traders in Financial Futures report covers equity index, Treasury, currency and volatility contracts, which is why it appears in the market sentiment framework on this site.
The useful questions are narrow ones. Who holds long, short and spreading positions in this market? How unusual is the current reading against its own history? How fast did it change this week? How old is the report? The explorer answers those four and nothing else. It does not say what any group intends, and it does not say what prices will do next.
For the conceptual background, read the Commitments of Traders guide. For why a report dated Tuesday cannot be treated as a Friday observation, read the guide to data latency and vintages.
COT Positioning Explorer
The default data is an illustrative sample with fictional values and fictional dates ending March 31, 2026. Live feeds are not connected. To analyze real positioning, choose "Paste rows" and supply weekly Traders in Financial Futures figures downloaded from the CFTC Commitments of Traders page.
All processing runs in your browser. The selected category, metric, window and as-of date are kept in the page address so a view can be shared. Pasted rows are never placed in the address.
The Five Participant Categories
The Traders in Financial Futures report sorts reporting traders into these groups. They are reporting classifications. A group can hold a position for hedging, relative-value or directional reasons, and the report does not say which. See the CFTC explanatory notes for the official definitions.
| Category | Plain-language description |
|---|---|
| Dealer/Intermediary | Large banks and securities dealers that typically act as intermediaries between other market participants. |
| Asset Manager/Institutional | Pension funds, endowments, insurance companies, mutual funds and similar institutional investors, plus the managers of those assets. |
| Leveraged Funds | Money managers and hedge-fund-style participants that use leverage and trading strategies such as arbitrage, macro and relative value. |
| Other Reportables | Reportable traders that do not fit the other three categories, such as certain corporate and proprietary accounts. |
| Nonreportable Positions | The remainder of open interest held by traders below the CFTC reporting thresholds. It is derived by subtraction, not reported trader by trader. |
How the Numbers Are Computed
Every figure comes from one calculation module that the page and the machine-readable tool share, so the two cannot disagree.
- Net position: long positions minus short positions, in contracts.
- Net as % of open interest: (long minus short) divided by open interest, times 100. When open interest is missing or zero the result is N/A, never a number.
- Weekly change: the value at a report date minus the value at the previous report date in the data. If a release is missing the change spans more than one week, and the explorer says so. Changes in long and short are shown separately, so "more long" is distinguishable from "less short".
- Gross exposure: long plus short. It can rise while net stays flat.
- Percentile: the rank of the latest net % of open interest among the weekly readings in the lookback window, counting ties as half. 0 is the most net short reading and 100 the most net long. Only readings dated on or before the as-of date, and already published by it, are used.
- Z-score: distance from the window mean in sample standard deviations. It is a normalization aid, not a probability, because positioning distributions are often skewed. It is not defined when every reading in the window is identical.
- Minimum history: percentile and z-score are withheld below 52 weekly observations, and the observation count is always shown.
Three dates stay separate. The observation date is when positions were measured. The publication date is when the report became public. The retrieval time is when the data was obtained. Freshness is judged against a weekly release cadence: a report is current until its next release is due, then pending for a short grace period, then stale. Holidays that move a release can be supplied by the caller. Positioning is one family in the sentiment composite framework, where Leveraged Funds and Asset Manager net positions are the two published components, and the family has a fixed weight in the published methodology version 1.0.0.
Worked Example: Reading One Week
The numbers below are hypothetical and chosen for arithmetic clarity. They describe no real market.
- A category reports 240,000 long and 300,000 short contracts, with 2,000,000 contracts of open interest.
- Net is 240,000 minus 300,000, which is -60,000 contracts. As a share of open interest that is -3.0%.
- Gross is 540,000 contracts, 27.0% of open interest.
- The previous report showed 250,000 long and 290,000 short, so net changed by -20,000. Long fell by 10,000 and short rose by 10,000, so both legs pushed net lower.
- If -3.0% ranks above 20% of the 156 weekly readings in a 3-year window, the percentile is 20, a fairly net short reading for that history.
That last line is a statement about history only. It does not say what happens next.
What This Tool Does Not Do
- It does not connect to the CFTC or to any live feed. The real, current reports are on the CFTC Commitments of Traders page.
- It does not verify pasted rows. Check them against the source before relying on any figure.
- It does not say what a category intends, and it never labels a group as informed or uninformed money.
- It does not predict price direction or suggest an action. A percentile is a rank in history.
- Positioning is one evidence family. Read it beside positioning, sentiment and breadth and cross-asset risk appetite rather than alone.
This page is educational and is not personalized investment advice.
Frequently Asked Questions
What does the Commitments of Traders report show?
The Commitments of Traders report is a weekly CFTC publication that breaks down open interest in futures markets by groups of reporting traders. The Traders in Financial Futures version covers financial contracts and splits each market into Dealer/Intermediary, Asset Manager/Institutional, Leveraged Funds, Other Reportables and Nonreportable Positions, with long, short and spreading positions for each. It records positions held on a Tuesday and is published days later, so it describes the past, not the present.
Why does this tool show three different dates?
Each date answers a different question. The observation date is the Tuesday the positions were measured. The publication date is when the CFTC made the report public. The retrieval time is when you obtained the data. Collapsing them into one date would make a report that is already several days old look current, so the explorer keeps all three visible and measures data age from the observation date.
How is the percentile calculated and why is it sometimes withheld?
The percentile ranks the latest net position, expressed as a share of open interest, against the weekly readings inside the selected window, using only readings dated on or before the as-of date. A reading of 0 is the most net short value in the window and 100 is the most net long. The percentile and z-score are withheld when the window holds fewer than 52 weekly observations, because a rank built on a few points looks more precise than it is.
Does a large net short position by Leveraged Funds mean prices will fall?
No. A net position is a reported exposure, not a forecast. Leveraged Funds can be short for hedging, basis, relative-value or arbitrage reasons, and the report does not reveal which. The percentile only shows how unusual the reading is compared with the selected history. The explorer therefore never labels a category as informed money and never turns a reading into a buy or sell signal.
Can I trust the sample data on this page?
The built-in sample is fictional. Its values and dates were generated to show how the tool behaves, and they describe no real market. Live feeds are not connected to this page. To analyze real positioning, download the weekly Traders in Financial Futures data from the CFTC, paste the rows in the documented column format, and check the results against the source.
Related Reading
- Commitments of Traders: the full guide to what the report measures and how to read it.
- Data latency and vintages: why observation, publication and retrieval are different moments.
- Positioning versus sentiment versus breadth: where positioning fits among related evidence.
- Cross-asset risk appetite: reading positioning beside credit, volatility and rates.
- Market sentiment analysis: the hub for this evidence framework.
References
Category names and the report structure follow the CFTC's own program documentation. No statistic on this page comes from an external dataset: the sample is fictional and the worked example is hypothetical, and both are labelled as such.
- CFTC: Commitments of Traders: the primary source for the weekly reports and their historical files.
- CFTC: Commitments of Traders Explanatory Notes: the official definitions of the report types, participant categories and futures-equivalent methodology.
Jurisdiction: United States. Last reviewed by the Swoopr Editorial Team in October 2026. This page is educational and is not personalized investment advice.