A deliberate extension of price beyond a structural level — prior high or low, Asian range boundary, OB extreme — that triggers the stops and breakout orders clustered there, fills the institutional position from the opposing side of that flow, and reverses. Chart evidence: a wick beyond the level, a close back inside the range (the wick is the sweep, not the close), and a reversal candle of equal or greater size within 1–5 candles. Every Phase 2 event — the Judas, the 09:30 Venom sweep — is this mechanism executing at a pre-planned boundary.
A large-body candle (or consecutive series) that drives price aggressively in one direction, leaving unfilled order book gaps — FVGs — in its wake. The FVG between C1 high and C3 low is the direct evidence: the algorithm moved price so fast that the order book in that zone was never filled. The MSS displacement candle at 02:33 EST is this mechanism announcing Phase 3 — and the gap it creates is the re-entry zone for the orders that did not fill at the initial velocity.
After displacement, the algorithm returns price to the FVG zone and OB level to fill the resting orders the displacement left behind — what ICT traders call the retracement to the FVG CE or OB mitigation. Not a random pullback: the completion of an interrupted fill. The CE — the gap midpoint — is the densest concentration of resting orders, which is why the FVG CE reaction is precise rather than approximate. This is the mechanism that fills the ICT trader’s limit order.
After the sweep (collection), displacement (MSS), and rebalancing (fill), the algorithm delivers price to the pre-planned External Range Liquidity target — the PDH, PWH, PMH, or IPDA data range level. This is the directional movement session model entries are positioned to capture. It is not a range extension; it is the completion of a pre-planned delivery to the target liquidity level.
| What the chart shows | Retail interpretation | ICT algorithmic interpretation |
|---|---|---|
| Price drops 25 pips below the prior day’s low on a bullish AMD day | Bearish breakout — support broken. Enter short. | SSL sweep — the pool below the PDL collected; institutional longs filled from retail sell flow. Phase 3 bullish delivery begins. Enter long from the BPR/FVG after the MSS. |
| A large bullish 5M candle (C2) leaves a gap between C1 high and C3 low | Momentum candle — wait for a pullback to “support” at C2’s open. | Displacement — the gap is the institutional order book void. The retracement returns to the FVG CE to complete the fill. Place the limit at the CE. |
| EUR/USD trades at the same level as 3 days ago after much movement | Consolidation — wait for a breakout in either direction. | Phase 1 accumulation within a daily/weekly AMD. Mark the boundaries as future Judas targets; identify which side the macro direction favours. |
| Price reaches the prior week’s high and reverses sharply | Resistance — strong level. Enter short at the level. | BSL collection at the PWH — the delivery completed, the buy-side swept. Take profit at or just before the PWH; the reversal is a new AMD cycle beginning, not a rejection. |
| RSI above 70 during a strong bullish move | Overbought — sell signal, or at least avoid buying. | Phase 3 distribution to the ERL. Phase 3 always reads “overbought” — sustained delivery produces sustained RSI elevation. Expected, not a reversal signal. |
| A large bearish candle sweeps the Asian low, then an equal bullish candle reverses | High volatility — choppy market, stay out. | Phase 2 Judas (SSL collection) followed by Phase 3 MSS (displacement above the prior swing high). The “choppy” two-candle sequence is the highest-probability ICT entry pattern. |
The algorithm does not reverse at arbitrary technical levels — it reverses after collecting the stops and breakout entries clustered at structural levels. The Judas is not an accidental overshoot; it is the planned collection event preceding every major AMD reversal. This is why patience during the sweep is structurally rational, not just psychologically demanding.
Displacement creates a structural obligation to return: the gap represents orders that were never filled at velocity. The FVG CE retracement is the fill — the structural basis of the CE limit entry.
Kill zones are not merely high-activity periods; they are the windows in which the delivery sequences are programmed to run. The 02:33 macro, the 09:50 Silver Bullet, the 09:30 Venom sweep — time-specific algorithmic events. A structurally valid FVG at 04:45 EST lacks the delivery backing. Time is part of the specification, not a filter.
Nearest IRL (PDH/PDL) first, then the ERL (PWH/PWL/IPDA). The partial profit at the first IRL is profit-taking at the algorithm’s nearest planned target — where it may pause, consolidate, or retrace before continuing.
Every AMD transition is announced by one: the Judas candle (Phase 1→2), the MSS candle (Phase 2→3), the ERL delivery candle (Phase 3 completion). The 02:33 MSS is the algorithm announcing Phase 3 has begun — which is why the entry signal is that candle, not an arbitrary level.
A move in the true delivery direction designed to trigger premature entries that the Judas will stop out. Bullish inducement before a bearish Judas builds the BSL above the prior high that the sweep then collects. Inducement looks like Phase 3 but is Phase 1 or pre-Phase 2 positioning — the reason false breakouts precede the sweep.
No delivery runs indefinitely: every Phase 3 eventually reaches an ERL that reverses the cycle or a premium/discount extreme where distribution turns to re-accumulation. AMD is a cycle framework, not a trend-following one — every ERL delivery sets up the next cycle.
When NQ makes a new low but ES holds its own: NQ is the sweep vehicle (SSL collection) and ES’s hold is the directional confirmation. SMT divergence is the algorithm operating on two correlated instruments simultaneously — the non-sweeping one reveals the true direction.
Lio has traded ICT and Smart Money Concepts on forex majors and US indices since 2021 and built LiquiditySweeps.com to teach the framework the way it should be learned: in sequence, on real charts, with free live tools instead of paid indicators. Every article on this site follows the same rule — nothing gets published that wouldn’t survive a trade review.
About All articles Mentorship