COT Data Reversals: Reading Crowded Positioning

COT data reversals cannot be timed from the report, but the data shows how much speculative money is already committed to a trend, and a trend with little new money left to join is fragile. This guide is about that judgement: when the CFTC’s positioning data says a trend is crowded, and what else to look for before you read a turn into it.

It assumes you know which report to read; for currencies that is the Traders in Financial Futures report and its leveraged funds (COT Report Explained). Every currency’s weeks are on COT report for forex.

The crowded trade

A trend lasts as long as new money keeps entering. At some point most of the speculative money that wanted in is in. With fewer new buyers left, a surprising release or a change in a central bank’s tone can set off a fast unwind as positions close.

That logic has limits. Stretched positioning raises the risk of a sharp move the other way; it does not promise one. Positioning can stay extreme for months while a strong fundamental trend runs on.

• A healthy trend: new positions every week, net positioning growing steadily, price backed by fresh money.
• A vulnerable trend: positioning near an extreme of its own history, weekly additions shrinking or reversing, price still moving with fewer new positions behind it.

Measuring an extreme

A raw contract count is hard to judge on its own. Three ways put it in proportion:

• Net as a share of open interest. The net position divided by all the contracts open in that market. It compares across years as the market grows or shrinks (What Is Open Interest?).
• A percentile. Where this week ranks against the market’s own weekly history. At the 90th percentile, funds are more net long than in 90% of the weeks in the window.
• A z-score. How many standard deviations this week sits from its average (What Is a Z-Score?).

If positioning were normally distributed, a z-score beyond ±2 would turn up about 4.6% of the time, about 2.3% on each side (NIST normal table). It is not: positioning trends and clusters. Read ±2 as unusual for that market, not as an exact probability.

The window matters too. A z-score against the last 52 weeks and one against the last ten years can disagree about the same week.

Four warning signs

An extreme tells you when to pay attention. These four patterns are evidence that an unwind may be under way. None is reliable alone, and each has given false alarms.

• Price and positioning diverge. Price keeps making new highs or lows while the funds’ net position starts to shrink. The move has less fund money behind it.
• The net position crosses zero. Funds go from net long to net short, or back. Their positioning has changed direction, not just size. It is clearer than a divergence, and it arrives later, often after much of the price move.
• A fast unwind. The position falls sharply without crossing zero. Funds that go from net long 120,000 contracts to 45,000 in three weeks have cut almost two thirds of it, and the trend has lost much of their support.
• An extreme meets a change in the fundamentals. Funds heavily long the dollar, then inflation falls unexpectedly and gives the Fed room to cut. Now the crowd is fragile and has a reason to leave.

Each currency’s page, such as the Euro COT report, prints the net position, its change on the week and its share of open interest for every week, and whether the latest week is the most long or the most short of the weeks shown.

An example, with invented numbers

The numbers below are invented to show the method; they are not a historical episode. Each Friday report describes positions on the Tuesday before.

• Week 1, Friday: leveraged funds are net long 82,000 euro contracts, the most long of the past year. EUR/USD has already traded three more days that the report does not cover.
• Week 2, Friday: EUR/USD made a new high on Wednesday. The report, from Tuesday, shows the net long down to 71,000. Price higher, positioning lower: a divergence, measured before the latest high.
• Week 3, Wednesday: US inflation comes in hotter than expected and markets price fewer Fed cuts. EUR/USD falls. Friday’s report is from Tuesday, the day before, so it shows nothing of the reaction.
• Week 4, Friday: the net long is down to 30,000. The unwind is in the data, about a week after it began in price.

The report flagged a crowded trade, then a weakening one, and confirmed the unwind after the fact. At no point did it say when to act.

Common mistakes

• Fading a trend too early. An extreme is a risk warning, not an entry.
• Ignoring the fundamentals. Funds long the dollar while the Fed is hiking have a reason to be there. An extreme means more when the driver behind it is changing.
• Reading the wrong report. Leveraged funds in TFF for currencies, managed money in the Disaggregated report for commodities.
• Treating COT as a timing tool. The Tuesday-to-Friday lag, and the fact that the report shows who is positioned rather than when they will leave, suit it to horizons of weeks, not days.