LiquidityScan

· ENTRY MODELS & TIMING · 12 MIN READ · UPDATED TODAY

Best ICT Model for Crypto and for Forex: Which Setups Carry Over and Which Don't

Best ICT Model for Crypto and for Forex: Which Setups Carry Over and Which Don't

The ICT models that travel best to crypto are built on price events, not on a session clock: sweep, structure shift, retrace. Here is which setups carry over from forex unchanged, which need adjusting, and which lose their logic in a market that never closes.

The short answer: the best ICT model for crypto is one that does not need a clock

There is no single best ICT model for crypto, but there is a clear pattern in which ones hold up. Models built from price events carry over almost unchanged: a liquidity sweep (a wick that runs the stops beyond an old high or low and comes back), then a structure shift (a close through the last opposing swing point), then a retrace into a fair value gap (FVG, the three-candle imbalance left by a fast move) or an order block (the last opposite-coloured candle before that move). Models built on a session clock or an opening gap were designed around forex and index hours. On a market that never closes, the Silver Bullet, the Asian range raid and the weekly opening gap either need adjusting or lose their reason to exist.

For forex the emphasis changes, not the substance. The same sweep, shift and retrace skeleton works, and the session models sit on top of it as a native timing layer. The rest of this article goes model by model and gives the reason each time.

Three differences that decide which models transfer

The clock

Forex has a trading week, a daily rollover and three regional sessions that hand over in sequence. A kill zone (a window around a session open in which ICT models expect the day's liquidity raid) means something there because desks in that region are actually starting work. Crypto perpetual contracts trade around the clock, weekends included, so there is no open to raid. Activity still clusters at certain hours, but that is a tendency of the participants, not a structural event. The longer version of that argument is in whether kill zones apply to a market that never closes.

The gaps

Forex shuts on Friday and reopens on Sunday, often at a different price. ICT's opening-gap concepts depend on that break. A perpetual contract has no session break, so its fair value gaps come from one source only: displacement, meaning fast one-sided candles that leave the wicks of candle one and candle three without overlap.

Whose orders rest beyond the level

Above equal highs on a currency pair you mostly find stop orders and breakout entries. On a crypto perpetual you also find the liquidation prices of leveraged positions. When price reaches them, the exchange closes those positions at market, which pushes the wick further in the same direction before anything turns. The practical consequence is about stop placement: a stop tucked just beyond the swept level is more exposed on a perpetual than the same stop on a major pair. How stop hunts differ between forex and crypto covers this in detail.

Models that carry over cleanly

Sweep, structure shift, retrace

This is the skeleton of the ICT 2022 model, and none of its price steps refers to time. Picture a bullish case. Price trades below a prior swing low and takes the sell stops there. It then turns and closes above the level where the last run of down candles into that low began: a CISD, or change in state of delivery. A stricter trader waits for the close above the last lower high, the market structure shift (MSS). The move up leaves an FVG or an order block behind it, and the entry is the retrace into that zone. The idea is wrong if price closes back below the swept low.

On forex, ICT (Michael J. Huddleston) teaches this inside the London or New York session. On crypto you can drop the session condition or keep it as a filter you test yourself. The price logic is identical either way. The 2022 model entry checklist lists every condition in order.

Candle Range Theory

CRT is a relationship between candles: one candle pokes beyond the high or low of an earlier candle, fails to hold there and closes back inside its range, with the opposite side of the range as the objective. Because it is defined by candles alone, it works on a 4H or daily chart of any market. The catch is the candle boundary. A crypto daily candle closes at 00:00 UTC on most venues, while most forex brokers roll the day at 17:00 New York time. The pattern is the same, but the candles are not, so a CRT rule copied from a forex chart must be re-read on crypto's own daily candles. The full method is in how to trade CRT highs, lows and midpoint entries.

Order block and FVG continuation

After a break of structure (BOS, a close beyond the last swing in the direction of the trend), price pulls back into the order block or FVG that launched the break and continues. This carries over to both markets, with one filter. On thinly traded altcoins a three-candle gap can be nothing more than a quiet candle with few trades, which is not displacement. Exotic currency pairs have the same weakness: wide spreads and gappy prices produce zones that look clean and mean little. Keep the model to liquid markets. Order blocks in forex versus crypto goes through the differences zone by zone.

Models that need adjusting

Kill-zone entries and the Silver Bullet

The Silver Bullet is an FVG entry taken inside a fixed one-hour window, such as 10 to 11 AM New York time. On forex and index futures that window follows a real open. On crypto the price part still makes sense: a sweep, a gap formed by displacement, an entry at the gap. The window itself is borrowed. Some crypto traders keep New York hours on the argument that US participation is large. Treat that as a hypothesis to check in your own journal, not as a rule the market owes you.

Asian range into London

The forex version rests on one premise: European and US pairs are quiet during Asian hours, so they build a tight range that London raids on one side before the real move. Crypto has no session that is quiet by design, and Asian hours can be the most active part of a coin's day. Mark the range, then look at it. If it is not tight, a break of it is just price moving, not a raid. The marking rules are in how to mark the ICT Asian range.

Power of Three and daily bias

Power of Three reads a day as accumulation, a false move through the opening price, then distribution the other way. It needs an anchor open. Forex traders use midnight New York. On crypto you have to choose between that and the 00:00 UTC daily open, and then stay with the choice, because switching anchors after the fact lets you explain any day. Daily bias taken from how the previous candle closed works in both markets on the same condition: know which candle you are reading.

SMT divergence

SMT compares two markets that normally move together and flags the moment one makes a new extreme and the other does not. Correlated majors give forex a natural pair. On crypto many coins simply track bitcoin with extra noise, so a divergence can be noise too. The model works only when you can say why the two markets should agree.

Models that mostly stay in forex and indices

The new week and new day opening gaps are defined by a market that closes and reopens. A perpetual contract has neither event, so there is nothing to mark. Bitcoin futures on the CME do close for the weekend, and some crypto traders watch that chart's gap, but it is a different instrument from the perpetual most of them trade. If the gap concept interests you, the guide to NWOG and NDOG explains it on the markets it was built for. The same goes for anything tied to the 9:30 New York opening bell: it is an index model.

Model by model: crypto versus forex

ModelWhat it depends onOn cryptoOn forex
Sweep, structure shift, retracePrice events onlyCarries over; drop or test the session conditionNative; usually paired with London or New York
Candle Range TheoryOne candle's range and a later closeCarries over; mind the 00:00 UTC daily boundaryCarries over; mind the broker's daily rollover
Order block and FVG continuationA BOS made with displacementCarries over on liquid marketsCarries over on majors and crosses
Silver Bullet, kill-zone entriesA fixed time windowPrice part works; the window is an assumption to testNative
Asian range raidA quiet session before LondonOnly when the range really is tightNative on European and US pairs
Power of ThreeAn anchor opening priceWorks once you fix one anchorNative with the midnight New York open
SMT divergenceTwo markets that should agreeNeeds a genuinely linked pairNative on correlated majors
Weekly and daily opening gapsA market that closesNot on perpetualsNative

How to choose one model for your market

  1. Write out the model's conditions and cross out every one that names a time or an open. If what remains still gives you a location, a trigger and an invalidation, the model transfers. If it does not, the clock was doing the work.
  2. Check the instrument before the pattern. A sweep on a liquid pair and a wick on an illiquid one look the same on a chart and are not the same event.
  3. Fix your candle boundaries. Decide which daily open and which daily close you use on each market, and write it down.
  4. Decide how you treat the weekend. Forex gives you two days with no candles. Crypto gives you weekend candles that count in every pattern. Choose in advance whether weekend structure has equal weight for you.

What traders get wrong about this

  • Copying session times without checking them. A forex kill-zone table pasted onto a coin is an untested assumption, even if it sometimes lines up.
  • Thinking crypto needs different concepts. It does not. Sweeps, structure and imbalances are read the same way. What changes is timing, gap logic and how far a wick can travel.
  • Treating the model name as the edge. Nobody can honestly rank these models by success rate for you. Results depend on the market, the timeframe and the execution, and the only figure worth trusting is your own record over a large enough sample.
  • Tight stops behind crypto sweeps. The forced-liquidation mechanism described above is the reason a stop distance that suits a major pair can be too small on a perpetual.

Seeing the building blocks on LiquidityScan

The models above are assembled from a small set of price events, and those events are what LiquidityScan detects. The All Scanners page puts all 15 scanners on one page with live counts, and each card shows how its pattern works. The app has three modes, Crypto, Forex and TradFi, and all 15 scanners run in all three, so the same pattern definition is applied to a perpetual and to a currency pair. Pages inside the app need a free account.

How the pieces map to this article:

  • Sweep and shift. CISD, CRT and Bias are among the eight classic scanners, which are free to read. Liquidity Sweep and Market Structure are premium scanners. Market Structure flags BOS and CHoCH (change of character, the first break against the trend).
  • Retrace zones. Order Blocks (OB+, a strong order block confirmed by a liquidity sweep) and Fair Value Gaps (graded by multi-timeframe nesting) are premium as well.
  • The session model. Asia Range Sweep looks for an Asian-range sweep plus reversal, aligned to a daily bias. It runs in crypto mode too. Whether the range was tight enough to matter is still your call.
  • Time as context. Several classic scanners tag a signal KZ when its candle opened inside the 12:00 to 16:00 UTC kill zone, on 1H and 4H only.

On the Free plan the premium seven show a live count, not the setups. The Starter plan opens all 15 to browse. Most scanners work on 1H, 4H and 1D. Only Fair Value Gaps and CISD+ also have 5m and 15m, on the Base plan and above. So use it to find context and zones on 1H and higher, not as a scalping tool, and time a one-hour window model yourself.

Detection happens on closed candles. In forex mode exotic pairs sit on a separate tab and stay out of alerts unless you pick them. A detection is context for your own decision. The platform does not tell anyone to buy or sell.

Frequently Asked Questions

What is the best ICT model for crypto?

The ones defined by price alone carry over best: a liquidity sweep, a structure shift or CISD, then a retrace into an FVG or order block. Candle Range Theory and order block continuation also carry over. None of them is known to be best in a measurable sense. They simply rely on nothing that crypto lacks.

Does the Silver Bullet work on Bitcoin?

The price half of it applies: a sweep, displacement, an entry at the fair value gap. The fixed one-hour window was built around forex and index opens, and Bitcoin has no open. If you keep the window, treat it as something to test on your own trades.

Do ICT kill zones matter for crypto?

They are a tendency, not a structural event. Crypto trades all day and all weekend, so no session open forces a liquidity raid. Activity can still cluster around London and New York hours, which makes time a reasonable filter and a poor requirement.

Can I trade the same ICT model on crypto and forex?

Yes, if the model survives with its time conditions removed. Keep the price rules identical and adjust three things per market: the daily candle boundary, the stop distance behind a sweep, and whether a session filter is native or assumed.

Which ICT concepts do not apply to crypto perpetuals?

The opening-gap concepts, because a perpetual never closes and reopens. Anything tied to the 9:30 New York opening bell is also an index model. The Asian range raid applies only on days when the range is actually tight.

Hayk Muradian

Hayk Muradian

Founder & Lead Analyst at LiquidityScan · 12+ years ICT/SMC trading · Institutional order flow specialist

Hayk Muradian is the founder of LiquidityScan, a professional trading intelligence platform built for ICT (Inner Circle Trader) and Smart Money Concepts (SMC) traders. With over a decade of hands-on experience reading institutional order flow across crypto, forex, and futures markets, Hayk specializes in identifying liquidity events, order blocks, and CISD setups on closed candles.

He built LiquidityScan after years of frustration with retail charting tools that ignored the mechanics institutions actually use. The platform now scans 400+ markets in real-time, surfacing the same patterns floor traders watch — without the noise.

Hayk writes about the methodology behind ICT and SMC, with a focus on practical, data-driven analysis rather than hype. He is a vocal critic of "smart money" content that misrepresents institutional intent and a strong advocate for methodology-respectful education.

View all 395 articles by Hayk Muradian →

Not trading advice. LiquidityScan publishes educational content for informational purposes only. Trading involves substantial risk of loss.