What the ICT Power of 3 actually is
ICT Power of 3, also called AMD, breaks a trading day or week into three phases, accumulation, manipulation and distribution, and ICT traders use it as the backbone sequence a session tends to follow. It is the simplest map in the methodology, three words that describe what price does before, during and after the real move, and the Market Maker Model is the fuller version that explains why each phase happens (The Inner Circle Trader, ICT methodology).
The three phases are accumulation, where price builds a quiet range and liquidity pools form on both sides, manipulation, the false move that sweeps one of those pools and traps early entrants, and distribution, the real directional move that follows. I read them as a sequence the day often follows, not a law it must obey.
The honesty framing carries through from the rest of the cluster. ICT traders posit that the manipulation phase exists to let institutional flow fill a position against trapped retail money, and that is the narrative layered on the observation rather than a mechanism confirmed by the data.
The three phases mapped to the trading sessions
The session mapping is what makes Power of 3 tradable, because it ties each phase to a window a trader can actually watch. The common read is that the Asian session forms the accumulation range, the London open delivers the manipulation, and the New York session completes the distribution, and the killzones page covers the exact session windows (The Inner Circle Trader, ICT methodology).
Accumulation is the quiet range that sets the highs and lows the rest of the day trades against, and the liquidity above and below it is the fuel for what comes next. Manipulation is the false move through one side of that range, and distribution is the displacement that follows once the trapped money has been cleared.
I treat the mapping as a default, not a guarantee. Some days the manipulation comes at the New York open instead of London, some days the Asian range never holds, and the value of the framework is that it gives you a template to compare the live session against rather than a rule to force on it.
The Judas swing, Power of 3 named
The manipulation phase has a name, and it is the one most associated with ICT. The Judas swing is the false move that opens the distribution leg, a push through obvious liquidity that reverses into the real direction, and it is the moment the cycle either confirms itself or fails.
The killzones page covers the mechanics in depth, so I will only frame it here.
I read the Judas swing as the test of the whole framework. When the false move sweeps liquidity and reverses into a clean displacement leg, the cycle is running, and when the false move keeps going and never reverses, you are looking at a real trend, not a manipulation.
The reason the Judas swing matters so much to ICT traders is that the highest-quality entries sit directly behind it. A trader who reads the false move correctly enters the distribution leg at its start, and a trader who reads it as the real move gives the session away.
Power of 3 versus Wyckoff, the contested lineage
This is the section I want to get right, because the SERP either forces the comparison or denies it, and the honest answer is more awkward than either. Richard Wyckoff's early-twentieth-century method documents a real accumulation and distribution cycle in serious detail, with defined phases and a composite-operator concept, and it is the genuine historical antecedent to anything that describes price as accumulating, manipulating and distributing.
What complicates the lineage is that ICT publicly rejects it. The methodology's author states there is nothing like the Wyckoff theory in his work, so the honest position is that Power of 3 overlaps with Wyckoff in concept while differing in structure, and the connection is real in the eyes of outside analysts and contested by the methodology itself.
I do not try to resolve the dispute. The practical use of the Wyckoff anchor is that it gives the Power of 3 a documented intellectual history that the ICT-specific version does not have on its own, and a trader can study both without needing to declare one the source of the other.
Why the manipulation leg works, sometimes
The manipulation phase is the part of Power of 3 with the closest thing to external evidence, and it is worth separating from the doctrine. Carol Osler's work on currency stop-loss and take-profit clustering, published as a New York Federal Reserve staff report and in the Journal of Finance, documents that stop orders cluster at round numbers and prior highs and lows, and that price tends to accelerate once those clusters are taken out.
That is a real mechanism for what ICT calls the Judas swing, because a move through an obvious level is mechanically a move through a cluster of resting stops, and the acceleration that follows is the cascade those stops trigger. What Osler's work does not do is validate the full Power of 3 cycle, and it is forex data, not the intraday session mapping ICT teaches.
I anchor the manipulation leg to that evidence and leave the rest as doctrine. The honest claim is that the sweep has a defensible mechanism and the three-phase sequence is a useful organising frame, and neither of those is the same as a peer-validated trading edge.
Power of 3 versus the Market Maker Model
The cleanest way to hold Power of 3 and the Market Maker Model in your head at once is the framing the methodology community has settled on, that Power of 3 tells you the sequence and the Market Maker Model tells you why each leg is there. The two describe the same cycle at different resolution, and they are built to be read together rather than chosen between.
Power of 3 gives you the three-word skeleton, accumulation, manipulation, distribution, and the Market Maker Model breaks the distribution leg into the expansion move, the retracement to the entry object, and the delivery to the target liquidity. The first tells you where in the day you are, and the second tells you where in the move to act.
I reach for Power of 3 when I want the overview of the session and the Market Maker Model when I want the entry. Most traders who use one end up using both, because the sequence without the entry is observation, and the entry without the sequence is a guess at timing.
| Framework | Phases | Best read as |
|---|---|---|
| Power of 3 (ICT) | 3, accumulation, manipulation, distribution | The sequence of the day |
| Market Maker Model (ICT) | 5, accumulation, manipulation, expansion, retracement, reversal | Why each leg happens and where to enter |
| Wyckoff Method | 4, accumulation, markup, distribution, markdown | The documented historical lineage ICT rejects |
Identifying which phase you are in right now
The practical question is always where in the cycle the current price sits, and the read starts with the range. If price is consolidating inside a tight band with liquidity above and below, the day is in accumulation, and the job is to mark the pools and wait.
A change of character out of that range is the signal that manipulation has begun.
Once a side is swept and price reverses with displacement, the day has moved into distribution, and the question becomes whether to enter the leg or wait for the retrace inside it. The framework gives you the phase, and the entry objects inside the methodology give you the level.
I keep the read probabilistic. A clean accumulation range that sweeps one side and reverses is a high-read Power of 3 day, and a session that never forms a range or sweeps both sides is a day the framework does not apply to, and recognising the second is as important as trading the first.
When Power of 3 fails, double Judas and no-manipulation days
The framework breaks in two specific ways, and the pages that skip this are the ones that cost traders money. The first is the double Judas, where price sweeps one side of the range, reverses to suggest the distribution leg has started, and then reverses again through the other side, which is two manipulation legs in a row and a trap for anyone who entered the first reversal.
The second is the no-manipulation day, where the accumulation range is swept and price simply continues in that direction without reversing, which is a trend day rather than a Power of 3 day. The framework assumes the sweep is false, and on a trend day the sweep is real, so forcing the read hands you a loss against a move that never comes back.
The defence is the same one I use across the methodology. Size for the case where the read is wrong, treat the sweep-and-reverse as a hypothesis the next candle can invalidate, and accept that some days the framework does not apply and the right trade is no trade.
Power of 3 on weekly and monthly frames
The three-phase read is fractal, and ICT traders apply it beyond the daily session to the weekly and monthly charts, where accumulation, manipulation and distribution play out over days or weeks rather than hours. A weekly accumulation range swept on a Monday open and reversed into a Friday distribution close is the same cycle on a slower clock.
I find the higher-timeframe read most useful for bias rather than entry. A market in weekly accumulation tells me to expect a manipulation leg and a distribution direction, and the daily Power of 3 then frames where inside that week the trade sets up, which keeps the higher and lower timeframes aligned.
The honesty caveat scales with the timeframe. A weekly read is harder to verify and easier to fit to a narrative in hindsight, so I weight it as context rather than confirmation, and the entry always comes from a lower-timeframe phase, not from the weekly label.
