What smart money concepts means when you apply it to crypto
Smart money concepts, or SMC, is a retail trading methodology built on the ICT framework of Michael Huddleston, the Inner Circle Trader, and applying it to crypto means adapting a method designed for forex sessions to a market that never closes. ICT traders read price action as if institutional flow were driving it, using patterns like order blocks, fair value gaps and liquidity sweeps, and the cluster home covers each of those mechanics in depth (The Inner Circle Trader, ICT methodology).
The methodology was systematised in the 2010s and is most often applied to forex, indices and crypto on intraday timeframes, and most of the doctrine was built before crypto perps existed at scale. I treat that history as the first thing to keep in mind, because a framework forged around the London open does not transfer to a 24 hour market without adjustment.
I keep the honesty line that runs through the whole cluster: ICT patterns are precisely defined, but they are doctrine within a community, not findings confirmed by order-book data, and that matters more on crypto where the order book you can see is usually not the one that matters.
How crypto and forex differ for SMC
The differences that matter fit in one table, and I keep it here so the rest of the page can expand on each row rather than restate it. The pattern vocabulary is identical across both markets, and what changes is the context the patterns live in.
| Aspect | Forex | Crypto |
|---|---|---|
| Market hours | Weekdays only, with a daily close | 24 hours, 7 days, no close |
| Session structure | Clean Asia, London and New York handoff | Partial and inherited, weaker than forex |
| Liquidity proxy | Stop clusters at round numbers | Funding rates and liquidation maps |
| Order book you can see | Retail sees no interbank book | Spot shows a venue book, perps show funding |
| Transparency edge | None | On-chain whale and exchange flow |
| Win-rate evidence | An FX stop-clustering study, no ICT backtest | No backtest at all |
I read that table as the contract for the rest of the page. Every section below takes one row and explains what it means for a live trade, so the comparison stays concrete instead of abstract.
The SMC stack, briefly
The stack is the same one the forex pages describe, so I will only name the pieces here and point you to the deeper guides. An order block is the last opposite candle before a move, and a fair value gap is the three-candle imbalance a fast move leaves behind.
A break of structure or change of character marks the trend shift, the premium and discount zones split the range into a buy half and a sell half, and the killzones are the session windows where ICT traders expect the cleanest moves. None of that changes on crypto, and what changes is the context around it.
I reach for the full SMC setup when I want to see how the pieces combine, then I swap the forex assumptions for the crypto ones in the rest of this page.
The structural break: crypto trades 24 hours a day with no session close
The first thing that breaks is the daily cycle. Forex has a clean structure built around session opens and a daily close, and ICT killzones lean on that rhythm, the Asia range that sets the high or low, London that sweeps it, and New York that expands away from it.
Crypto has no close, so the tidy session-to-session handoff that organises a forex day does not exist in the same form.
That does not mean crypto is structureless. A 2024 peer-reviewed study of intraday crypto behaviour found clear time-of-day patterns in returns, volatility, liquidity and volume across a wide set of pairs and exchanges, and the authors titled the paper on the observation that crypto activity clusters around a specific window, which tells you the session effect is real even without a market close.
My read is that crypto inherits a partial session structure because the fiat on and off ramps, the TradFi correlation, and the staffing of trading desks still flow through London and New York hours. The pattern is weaker and noisier than forex, and the clean killzone sequence is something you approximate, not something the market hands you.
Whether ICT killzones work on Bitcoin and Ethereum
The honest answer is that they work loosely, and no peer-reviewed study backtests ICT killzones on crypto. The intraday crypto study confirms that volume and volatility cluster by time of day, and Amberdata's 2025 analysis of more than 50,000 minutes of Bitcoin order-book data found predictable time-based patterns in when liquidity concentrates and disperses, which is the raw material a killzone read needs.
What neither source gives you is the specific ICT claim, that price sweeps liquidity in the Asia range and expands in the London or New York window, validated as a profitable rule on BTC. I have not found that study, and neither has any page I have seen cite one, so I treat the killzone timing as a reasonable filter rather than a proven edge.
I use the windows the way I use them in forex, as a time filter that concentrates my attention on the hours when moves are likelier, and I drop the expectation that the Asia range always sets the daily high or low. On crypto that sequence breaks often enough that trading it as gospel will cost you.
Funding rates and liquidations: the crypto-native liquidity layer
This is the part forex traders miss, and it is where crypto actually has something forex does not. On a spot crypto exchange you cannot see the real institutional order book the way the ICT narrative implies, because most retail venues show you their own internal book, not a consolidated interbank flow, so the footprint story is weaker than in forex.
What crypto does expose, transparently and in real time, is perpetual-futures funding rates and liquidation maps. A baseline funding rate of 0.01 percent per eight-hour interval costs the crowded side roughly 11 percent annualised, and in extreme markets funding has sustained above 0.3 percent per interval, which annualises to well over 100 percent, and those extremes flag a crowded side that often gets squeezed.
I treat that as the closest thing crypto has to a real order-flow signal, and the funding rates guide and the liquidations guide cover the mechanics in depth.
The practical translation is that when ICT traders talk about liquidity on crypto, the honest proxy is not a line on the spot chart, it is the funding skew and the cluster of liquidation levels sitting above obvious highs and below obvious lows. A draw on liquidity on BTC is, mechanically, often a move into a pocket of leveraged stops, and you can see the fuel for it on the funding and liquidation map before it happens.
On-chain data: the transparency layer forex does not have
Crypto has one genuine edge over forex for following large players, and it is the part I would not trade away. Every major transfer is on a public ledger, so whale-wallet movement, exchange inflows and outflows, and stablecoin mint and burn events are visible to anyone, through services like Glassnode, CryptoQuant and Nansen, and nothing equivalent exists in forex where the interbank flow is private.
I want to be careful about what that proves. On-chain data is a real transparency layer, and it can confirm that large holders are moving coins to exchange wallets before a sell-off or that stablecoins are being minted before a bid, but it does not validate ICT doctrine itself.
It is a separate, independently useful signal, and bolting it onto an order-block read does not make the order block a proven institutional footprint.
My honest use is to treat on-chain flow as context that forex SMC traders simply do not get, a way to see whether the big moves have real backing or are just leverage unwinding, and to keep it separate from the labelled ICT patterns rather than pretending it confirms them.
Order blocks and fair value gaps on crypto charts: same pattern, weaker claim
The patterns themselves print identically. A last opposite candle before a displacement move looks the same on BTCUSDT as it does on EURUSD, and a three-candle fair value gap is a three-candle gap on any chart, so the displacement and imbalance mechanics transfer directly.
What weakens is the interpretation. The ICT claim is that an order block is the candle where an institution loaded a position, and on crypto spot the order book behind that candle is usually hidden or internal to the venue, so you are reading the footprint without seeing the foot.
The pattern is real; the institutional story attached to it is the same unverified doctrine as in forex.
I trade the patterns the same way, as levels of interest where price may react, and I drop the certainty that a reaction there means smart money is involved. On crypto a perfectly clean order block can fail because a single large leveraged long got liquidated through it, which is mechanics, not methodology.
Stablecoin pairings and weekend illiquidity
Two more crypto specifics change how the zones read. Most BTC and ETH spot trades against a stablecoin, USDT or USDC, which means the quote currency is itself a crypto asset with its own depeg risk, and a premium and discount array is only as clean as the peg holding underneath it.
Weekend liquidity is the second. Saturday and Sunday volume is materially thinner than weekday volume, the same intraday study documents a clear intraweek pattern, and funding rates drift to extremes when the crowd is one-sided and the book is thin.
I am slower to take signals on a weekend because the same move on a thin book is more likely to be a trap than a confirmation.
The combination matters for the array read. A premium or discount level measured against a thin weekend book and a stretched stablecoin is a weaker level than the same level measured on a Tuesday at the London open, and I size accordingly.
A worked crypto SMC setup, end to end
I run the setup the same way I run it in forex, with the crypto substitutions layered in. Start with a higher-timeframe bias on the 4 hour or daily, wait for the London or New York window from the killzones, and watch for a sweep of obvious liquidity, which on crypto means the highs or lows where leveraged stops are clustered.
Wait for a break of structure or change of character on the 15 minute to confirm the turn, draw the entry to an order block or fair value gap on the lower timeframe, and check the funding rate before you size. If funding is extremely positive into a long, the crowd is already on your side and the squeeze risk is real, which is the crypto-specific filter forex does not require.
I treat this as an application of the full SMC strategy, not a new system, and the only steps that differ are the funding check and the on-chain context. Everything else is the same stack described on the capstone page.
What is actually proven, and the win-rate claim nobody can source
The honest evidence sits one level above ICT, and it is thinner on crypto than on forex. Carol Osler's paper in the Journal of Finance documents that currency stop-loss and take-profit orders cluster at round numbers, and that trends reverse at those levels and accelerate once they break, which is the closest peer-reviewed anchor for the ICT idea that liquidity sits at obvious levels and gets swept.
The data is from forex, not crypto, and it says nothing about order blocks or fair value gaps specifically.
No peer-reviewed or tier-1 study backtests ICT patterns on Bitcoin or any crypto asset, and the 80 to 90 percent win-rate figures you read on educator blogs trace back to unsourced claims repeated until they read like data. I have not found a verified backtest, and neither has any page I have seen quote one with a citation, so I will not give you a number.
What I can say is that the crypto-native layers, funding rates, liquidations and on-chain flow, are real and independently useful, and they give a crypto SMC trader information a forex trader does not get. That is a genuine edge, and it is not the same thing as proof that ICT methodology works.
Where the crypto SMC setup fails
A false sweep is the most common failure, and crypto makes it worse. Price wicks through an obvious high, the stops fire, and instead of reversing the move keeps going, because on a thin book there was no large player behind the sweep at all, just cascading leverage.
A funding-driven liquidation cascade looks exactly like smart money until you check the funding tab. Price displaces through a level with a gap and a break of structure, the textbook ICT confirmation, but the fuel was a crowded side being liquidated, not an institution positioning, and the move reverses the moment the cascade exhausts.
A stablecoin depeg can read as a break of structure on the paired chart, because the quote currency moved, not the asset. I have watched a wick on BTCUSDT that was entirely a USDT wick, and trading it as a character change would have been a mistake.
The discipline is the same as in forex, size so a failed pattern costs only what you planned, with the added crypto check that the move you are reading is in the asset and not in the stablecoin underneath it.
