Smart money concepts (SMC) is a way of reading price charts that assumes large traders leave footprints: the swing points they break, the gaps they leave and the stops they run. Keel turns each concept into a rule you can backtest and run on Hyperliquid.
Part of our guide to ICT trading.
Smart money is the name traders give to banks, funds and other large players. Smart money concepts start from one idea: a large order cannot be filled all at once without moving price, so big players need someone on the other side, and the place to find them is where other traders have left orders resting. On a chart, that leaves footprints you can learn to read.
Most of the vocabulary comes from ICT, a trader who taught it in free video lessons, and some from older supply and demand trading. Smart money concepts, or SMC, is the name a later wave of educators gave the same material. The two are close enough that our ICT trading guide covers the same patterns with ICT’s names and its rules about time of day.
Keel treats each concept as a rule a computer can follow, written from the definitions the most used indicators run, with every disputed choice as a setting. That is what lets you backtest it.
Everything in smart money concepts starts with market structure: the run of swing highs and swing lows. An uptrend makes higher highs and higher lows; a downtrend makes lower lows and lower highs. A swing high is a candle with lower highs on both sides of it, and it only counts once the candles after it have closed.
Two events move the structure along. A close past the last swing in the direction of the trend is a break of structure: the trend carries on. The first close past a swing against the trend is a change of character: the trend may be turning. Every other concept is read against that backdrop, which is why traders mark structure first.
Each concept below links to its own guide, with an example from real candles, the exact rule our engine runs, and live examples from the most traded Hyperliquid perps.
A close past the last swing point, with the trend.
Rule: Close beyond the last swing high or low
The first close past a swing point against the trend.
Rule: The first break against the trend
The price range a fast three-candle move skipped.
Rule: Third candle's low above the first's high
A fair value gap price closed through, now read from the other side.
Rule: A candle closes through a fair value gap
The candle a strong move started from, treated as a zone.
Rule: The candle a structure-breaking move began
A failed order block whose role flips once structure breaks.
Rule: A failed block after a sweep, retested
A run past a high or low that closes back inside.
Rule: Wick past a swing level, close back inside
A failed breakout of an older high or low, faded.
Rule: Failed break of the 20-candle high or low
The concepts are meant to be used together. A typical smart money concept trading setup reads like this:
Every step is a choice: which timeframe, how big a swing, which candle is the block, how long to wait. Written as exact rules, each choice can be backtested, and small changes in them can change the result a lot.
We ran a plain rule for four of these concepts on the top 10 Hyperliquid perps, net of fees, slippage and funding, and published the results whether they held up or not. Each page shows the rule, the numbers by coin and timeframe, and a version you can fork.
The best known SMC indicator is Smart Money Concepts [LuxAlgo], an open-source TradingView script, which draws structure breaks, order blocks, fair value gaps and equal highs and lows on a chart. There is also smart-money-concepts, an open-source Python library that marks the same patterns in code. The two disagree in places, such as how a swing is found and which candle is the order block.
Keel’s blocks follow the same definitions and make each of those disputed choices a setting, with the more common reading as the default. The difference is what comes next: the same rule runs in a backtest with fees and funding charged, then in paper trading, then live on Hyperliquid.
Smart money concepts grew up on forex charts, where the market closes at weekends and the day has clear sessions. The patterns read the same on crypto, with two differences. Crypto perps trade around the clock, so the hours ICT cares about need checking rather than assuming; our ICT kill zones guide does that with our own data. And on a perpetual futures exchange the stop runs are visible: a liquidation is a forced close the venue sends to the public order book.
The core patterns are the same: fair value gaps, order blocks, liquidity and market structure. ICT keeps its own names, such as market structure shift for a change of character, and adds rules about time of day and Fibonacci entries. Smart money concepts educators lean more on supply and demand zones, which order blocks largely rename, and use a few terms of their own, such as inducement. If you learned one, you can read the other.
No peer-reviewed study tests smart money concepts as a method. Parts of it sit on documented effects. Stop orders do cluster past old highs and lows and at round numbers, and when they trigger they push price further, in waves (Osler, Currency Orders and Exchange Rate Dynamics (Journal of Finance, 2003); Osler, Stop-loss orders and price cascades in currency markets (Journal of International Money and Finance, 2005)). Rules that buy the break of a prior high have shown predictive power over long stock index histories, though costs absorb much of it (Brock, Lakonishok and LeBaron, Simple Technical Trading Rules and the Stochastic Properties of Stock Returns (Journal of Finance, 1992)). The claim that order blocks attract price back has no test behind it at all.
That is why we publish backtests rather than win rates: a rule either holds up across coins and timeframes after costs, or it does not.
Live SMC events are updating. They appear after the next hourly data run.
Open your assistant with this request already written. It reads this page and builds the strategy in Keel, where you can backtest it.
Keel is a strategy builder for Hyperliquid. Describe an idea, backtest it with fees, slippage and funding, and run it live on Hyperliquid’s crypto perps. Stock and index perps are coming soon.
Start free. Connect a Hyperliquid wallet when you’re ready to trade.
Smart money concept (SMC) trading reads a chart for the footprints large traders are thought to leave: the trend's swing points, the gaps a fast move leaves, the zones a move starts from, and the stops resting past old highs and lows. Traders combine them into entries with a stop and a target.
A typical one reads the trend from market structure on a higher timeframe, waits for price to run the stops past a high or low, waits for a change of character the other way, then enters on a pullback into the fair value gap or order block the move left, with a stop beyond it.
No published study has tested smart money concepts as a method. Parts of it rest on documented effects, such as stop orders clustering past old highs and lows, while the claim that order blocks attract price has no test behind it. Backtest a rule with fees and funding before trusting it.
Yes. The most used is LuxAlgo's Smart Money Concepts indicator on TradingView, and there is an open-source Python library that marks the same patterns. Keel's blocks follow the same definitions, with each disputed choice as a setting, and run in backtests and live on Hyperliquid.
Smart money concepts is mostly the ICT material taught under a new name by a later wave of educators, with some ideas from supply and demand trading added. The core patterns are the same. ICT keeps more time-of-day rules, such as kill zones, and its own names, such as market structure shift.
The patterns read the same on any chart. Smart money concepts grew up on forex, where markets close at weekends; crypto perps trade around the clock, and their liquidations make stop runs visible, since a forced close is a fill the venue executes on the public book.
The same patterns with ICT's names, plus kill zones, OTE, the judas swing and inducement.
The hours ICT traders watch, in UTC, checked against our own BTC and ETH data.
The steps to test any of these rules honestly, with every cost charged.