Liquidity Sweeps, Stop Runs, and False Breakouts: What the Backtests Show
The SMC Liquidity Sweep signal posts a 71% win rate yet beats buy-and-hold in only 8% of backtests—here is why that gap exists and what to use instead.
What a Liquidity Sweep Actually Is
A liquidity sweep—also called a stop run or stop hunt—is a price move that briefly breaches a well-known support or resistance level, triggers the resting orders clustered just beyond it (stop-losses and pending entries), and then reverses. In market microstructure terms, the sweep is the mechanism by which a market temporarily exhausts one side of the order book before filling the opposing side at a more favorable price.
The academic literature treats this as a structural feature of limit-order markets, not a conspiracy. When stop orders cluster at a round number or a recent swing high, any sufficiently large market order—or a temporary imbalance in liquidity provision—will push price through that level. The move is evidence that liquidity existed there, not proof of deliberate manipulation. FX microstructure research confirms that price frequently overshoots technical levels before mean-reverting, which is the empirical basis for false-breakout trading.
Smart Money Concepts (SMC) and related retail frameworks have popularized the term, framing these moves as intentional hunts by institutional players. Whether or not that framing is accurate, the tradeable question is the same: does identifying these sweeps before they reverse produce a reliable edge out of sample?
What the Research Literature Says
Several market microstructure papers document that price does overshoot technical levels with above-random frequency, particularly in FX. The mechanism is well-supported: stop-loss clustering at obvious levels is documented, and the subsequent mean-reversion after the flush is a real statistical pattern in tick data.
The research consistently notes two caveats. First, the effect is strongest at the microstructure scale—very short-term, sub-minute data—where retail traders face the widest spreads and fastest execution requirements. Second, identifying a sweep in advance is much harder than labeling one in hindsight. Most papers reporting positive results either test at timeframes inaccessible to most retail participants, use hand-labeled data that introduces look-ahead bias, or report gross returns before accounting for the spread costs that arise precisely when you are trading against a fast-moving market.
These are not reasons to dismiss the concept entirely. They are reasons to test it rigorously, at realistic costs, across a broad universe of assets and timeframes, out of sample—which is what the IndicatorEdge methodology does.
What 660,005 Backtests Actually Found
Across 660,005 out-of-sample backtests covering 903 assets and four timeframes—1-Hour, 4-Hour, Daily, and Weekly—the SMC: Liquidity Sweep indicator produced a median win rate of 71.2%. That is one of the highest in the entire test universe. The other number is the one that matters: it beat buy-and-hold in only 8% of backtests.
It is not alone in this pattern. A cluster of indicators with win rates between 71% and 74% all fail to beat buy-and-hold in more than 11% of cases. Murrey Math Lines hits a 74.3% median win rate but beats buy-and-hold in just 11% of backtests. Holy Grail Confluence reaches 73.3% median wins at 8%. The win rate is being inflated by a systematic bias—these strategies generate many small wins and occasional large losses, or they hold positions infrequently enough that the drift of a trending market overwhelms them.
The broader result is consistent: across all 382 indicators tested, only 26% of indicator-asset combinations beat buy-and-hold. For the SMC family specifically, no SMC-based indicator beat buy-and-hold across any asset class in the test universe. The median best Sharpe ratio across the strategies that did clear the bar is 0.62—not spectacular, but positive risk-adjusted return after costs.
Why Win Rate Is a Misleading Scorecard
A 71% win rate with net-negative alpha is not a paradox once you account for reward-to-risk. If losing trades are roughly twice the size of winning trades—which is common when you are fading a sweep and the market continues in the original direction—the system loses money even at 70% accuracy.
Liquidity sweep strategies face a structural problem on the timeframes we test. The 1-Hour, 4-Hour, Daily, and Weekly bars our backtests cover capture multi-hour to multi-day moves. A stop run that resolves in minutes appears as noise inside a single bar. The theoretical edge exists at a granularity these timeframes cannot reliably measure, and it is precisely at those shorter timeframes where transaction costs are most punishing.
The honest interpretation: the concept is grounded in real microstructure mechanics. The indicator implementations tested here, at the timeframes tested, do not produce returns that justify their use over passive exposure. A high win rate printed by a backtest engine is not the same thing as an edge.
What the Data Does Support
If you trade Forex and want a signal with an actual out-of-sample track record across this universe, the Fisher Transform led the Forex category across 17 assets. For equities, Fibonacci Pivots topped the Stocks class across 22 assets, followed by Projection Bands (16 assets) and Camarilla Pivots (16 assets). For crypto, MA Envelope led across 5 assets and Fibonacci Pivots across 4.
These are not recommendations—they are a ranked starting point. If you are going to build a rules-based system around a particular instrument class, the empirical record gives you a list of what worked out of sample, and that list does not include any SMC-derived signal.
The full rankings by asset, indicator, and timeframe are on the assets page. If a specific asset you trade is in the 903-asset universe, its top-ranked indicator and full backtest summary are there.
Questions, answered
What is a liquidity sweep in forex market microstructure?
A liquidity sweep is a price move that breaches a technically significant level—a recent high, low, or round number—where stop-loss and pending orders cluster, triggers those orders, then reverses. In microstructure terms it is the market clearing an order imbalance at that level before returning toward equilibrium. The term is neutral; it does not imply intentional manipulation by any single participant, though retail frameworks often frame it that way.
Does the SMC Liquidity Sweep indicator work in backtests?
In 660,005 out-of-sample backtests across 903 assets at 1-Hour, 4-Hour, Daily, and Weekly timeframes, the SMC: Liquidity Sweep signal produced a median win rate of 71.2% but beat buy-and-hold in only 8% of cases. No SMC-based indicator in the test universe beat buy-and-hold across any asset class.
Are there research papers on stop hunts and false breakouts?
Yes. Market microstructure literature documents stop-loss clustering at technical levels and the resulting mean-reversion after price breaches those levels. The consistent finding is that the statistical effect is strongest at very short timeframes—sub-minute tick data—where spread costs and execution speed make it difficult for most retail participants to exploit. At the 1-Hour to Weekly timeframes most retail strategies operate on, the edge is much harder to isolate after costs.
Are these real trading results or hypothetical?
These are hypothetical backtests run out of sample with realistic transaction costs applied. They show what a mechanical rule would have produced historically, not actual traded returns. Past backtest performance does not guarantee future results. Nothing here is financial advice or a recommendation to trade any particular instrument or strategy.
Every figure here comes from our own out-of-sample backtests, costs included — not a course or a guess. Educational information only — not investment advice. Hypothetical backtested results; past performance does not guarantee future results. Trading involves risk of loss.
Keep reading
Get the weekly edge report
The best-performing indicator per asset, what changed this week, and the honest caveats — straight to your inbox.