How do I know traders are actually trapped?

You do not. It is an inference from a failed break plus a volume asymmetry: little business done beyond the level, much more coming back through it. Treat it as a hypothesis the next few minutes confirm or refuse, with an invalidation written before entry.

Is this the same as a stop hunt?

They are different claims. Trapped traders describe a mechanical consequence of where stops sit. A stop hunt asserts intent — that someone pushed price specifically to trigger them. Intent is not observable in the data, and the trade is identical either way, so the assertion adds a way to be wrong without adding information. [Spoofing](/glossary/spoofing/), by contrast, is a specific and legally defined behaviour.

Does the idea apply outside futures?

The mechanism needs stop orders and a visible reference level, which exist in any centrally traded market. What changes is measurability: across fragmented venues, crypto or equities, the volume asymmetry you rely on is split across books you may not be aggregating.

Glossary

Trapped traders

Also called: trapped longs · trapped shorts · failed breakout fuel

Trapped traders are participants caught on the wrong side of a failed move, usually a breakout that reversed. Their stop orders sit above or below the failure point, and when price returns, those stops execute as market orders in the opposite direction. The trap supplies the fuel for the move against them.

Why a stop is fuel

A protective stop is an instruction to send a market order. When a long's stop triggers, it sells at market. A cluster of longs entered above a level is therefore a pool of pending sell market orders waiting under that level.

That is the entire idea. When price comes back through the failure point, the exits are themselves aggressive orders in the new direction, and they arrive whether or not anyone changed their mind about value. The move accelerates because it is forced, not because it is believed.

Where they are, approximately

You cannot see positions, but you can see the prices at which entering was obvious. Session highs and lows, the edges of the initial balance, yesterday's extremes, a poor high, the far side of a single print — breakout orders concentrate there, so the trapped end up there too.

The order book will not show you this. Stop orders resting at the exchange do not appear in the displayed depth, and many stops are not orders at all: they sit in a platform, a broker's system, or a trader's head. What you can observe is the aftermath — a break that fails, then a burst of volume back through the level out of proportion to anything that was announced.

The numbers that circulate

Retest rates for failed breakouts are quoted across order flow teaching, and no version of the figure arrives with a public dataset, a stated definition of failed or of retest, or a replication anyone can point to. It is repetition, not evidence.

Treat it as a hypothesis with a missing experiment. Define a break as N ticks beyond the level, define failure as a close back inside within M minutes, then count both outcomes over a period you fixed in advance. You will end up with a number that is yours. It will probably look nothing like the one you were told, and that is the point of measuring it.

A worked example

In a synthetic ES sequence, the session high is 5 341.25. Price trades three ticks above it on 480 contracts with a delta of +310 — a small amount of business for a new session high. Four minutes later it is back below 5 341.25.

Over the following eleven minutes, 6 200 contracts trade with a cumulative delta of −2 400 and price covers eleven points to 5 330.25, where volume normalises. The asymmetry is the reading: 480 contracts established the breakout, 6 200 unwound it.

The trap

Assuming the fuel exists. You inferred a crowd of trapped longs from a price pattern; you never observed one. If the participants above that level were sized so the break was noise to them, or were hedging something else, or ran no stop at all, there is nothing to force out and price simply chops.

The second error is scale. This is a description of the first minutes after a failure, not of the day. It explains why the initial move is fast. It says nothing about where the move ends, and sizing up on the story converts a short, mechanical effect into a directional bet the mechanism never justified.

Frequently asked

How do I know traders are actually trapped?: You do not. It is an inference from a failed break plus a volume asymmetry: little business done beyond the level, much more coming back through it. Treat it as a hypothesis the next few minutes confirm or refuse, with an invalidation written before entry.
Is this the same as a stop hunt?: They are different claims. Trapped traders describe a mechanical consequence of where stops sit. A stop hunt asserts intent — that someone pushed price specifically to trigger them. Intent is not observable in the data, and the trade is identical either way, so the assertion adds a way to be wrong without adding information. Spoofing, by contrast, is a specific and legally defined behaviour.
Does the idea apply outside futures?: The mechanism needs stop orders and a visible reference level, which exist in any centrally traded market. What changes is measurability: across fragmented venues, crypto or equities, the volume asymmetry you rely on is split across books you may not be aggregating.