COT: Commitment of Traders Report
Also known as: COT report, CFTC COT, commitments of traders, COT positioning data
What is it?
The Commitment of Traders report is a weekly breakdown, published by the US Commodity Futures Trading Commission, of who holds the open futures positions in a given market. It splits open interest into commercials, who are hedging a real business exposure, non-commercials or large speculators, who are trading for profit, and non-reportable small traders below the reporting threshold. The timing is the first thing to internalise: the data is a snapshot of Tuesday's close and is not released until Friday at 15:30 New York time, so what you are reading is already three days old before you see it.
- Crowded short
- Middle of the range
- Crowded long
That makes it context, not a trigger. What traders actually extract from it is positioning extremes. When large speculators are historically stretched to one side — say net long the euro at the largest position in three years — the pool of buyers left to push price further is thinning, and the market becomes vulnerable to a sharp unwind if sentiment turns.
Commercials typically sit on the opposite side and are read as the slower, better-informed hedgers. The classic use is contrarian and slow: an extreme reading suggests a trend is crowded and its risk profile has changed, over weeks to months. It never tells you the turn has arrived, because positioning can stay extreme far longer than an account can sit through the drawdown, and it covers only exchange-traded futures — spot forex, which is decentralised, is only visible through its currency futures proxy.
Why it matters: It shows when speculative positioning has become crowded to one side, flagging a trend whose risk of a sharp unwind has quietly risen.
Net position = long contracts - short contracts, per trader category
It reframes the risk of an existing trend over weeks, but is far too lagging to time any entry.
Real-world example
Large speculators reach a three-year record net-long position in euro futures; the crowded side unwinds over the following weeks as the trend stalls.
How SignalBots handles it
SignalBots signals are generated from live price and volatility conditions, so weekly positioning data belongs in your broader risk view rather than in any individual entry. See /risk-warning.
Pro tip
Read net positioning against its own multi-year range rather than its raw contract count — an absolute number means nothing without knowing what extreme looks like for that market.
Common pitfalls
Treating an extreme reading as a reversal signal and fading a strong trend, when positioning routinely stays stretched for months.
Frequently asked questions
When exactly is the COT report released?
Every Friday at 15:30 New York time, covering positions as of the prior Tuesday's close. That three-day lag is built into the report and is the main reason it cannot be used for timing.
What is the difference between commercials and non-commercials?
Commercials hedge a genuine business exposure, such as a producer locking in a price, so their positioning is driven by need rather than a view. Non-commercials are large speculators trading for profit, and their crowding is what most analysts watch.
Does COT data cover spot forex?
Not directly. Spot forex is decentralised and has no central reporting. Traders use the CME currency futures as a proxy, which correlates reasonably well but represents only a slice of total currency activity.
Can a bot trade the COT report?
It can ingest the weekly figures and compute positioning percentiles as a slow filter on trend exposure. Using it as a direct entry trigger is impractical, because the signal resolves over weeks and the data is already stale on arrival.
Is an extreme reading a reliable reversal signal?
No. Extremes identify crowded positioning, not turning points, and markets can stay crowded for months while a fade bleeds. Treat it as context that changes position sizing, and remember your capital is at risk. See /risk-warning.
Trading involves substantial risk of loss. Historical and backtested results do not guarantee future performance. Read the full risk warning.
