Does the SMC / ICT Strategy Actually Work? Backtest Evidence and Win Rate
660,005 out-of-sample backtests across 903 assets tested SMC's Liquidity Sweep signal — the headline win rate looks compelling, but the edge disappears on closer inspection.
What We Tested
IndicatorEdge ran 660,005 out-of-sample backtests across 903 assets and 382 indicators. Every backtest covers one of four timeframes — 1-Hour, 4-Hour, Daily, or Weekly — uses realistic transaction costs, and is measured against a passive buy-and-hold benchmark. That last part matters: beating buy-and-hold is the bar, not just making money.
Among the 382 indicators tested were several signals derived from Smart Money Concepts (SMC) / ICT methodology, including the SMC: Liquidity Sweep. The question is whether the signal produces a reproducible statistical edge, or whether it just looks good on a hand-picked chart.
The Win Rate Trap
SMC: Liquidity Sweep produced a median win rate of 71.2% across the assets where it was tested. More than seven in ten trades closed in profit. If you follow SMC content online, you have probably seen screenshots that look exactly like this.
Win rate alone means nothing without knowing the size of the average winner versus the average loser. If your losers are twice the size of your winners, a 71% win rate still produces a net loss over time. The metric that actually matters is whether the strategy outperforms passive buy-and-hold after costs. On that measure, SMC: Liquidity Sweep beat buy-and-hold in only 8% of assets tested.
This pattern — high win rate, low real edge — is one of the most reliable traps in retail trading. Holy Grail Confluence sits at a 73.3% median win rate with an 8% beat rate. Murrey Math Lines reaches 74.3% with 11%. RSI Mean-Reversion shows 71.7% with 10%. The Liquidity Sweep is firmly in that company.
Independent Verification: No SMC Signal Beat Buy-and-Hold
Across every SMC-derived indicator we could formalize into a mechanical rule, none beat buy-and-hold consistently across the 903-asset universe. That result held across all four tested timeframes: 1-Hour, 4-Hour, Daily, and Weekly.
This does not prove that every trade taken with SMC ideas is a loser — discretionary execution, position sizing, and asset selection all affect outcomes. What it does mean is that the signal itself, isolated and tested mechanically, carries no measurable statistical advantage over simply holding the asset.
For context: in 63% of assets we tested, at least one of the 382 indicators did beat buy-and-hold. Edge exists in markets — it just was not concentrated in the SMC family.
What Actually Works on the Assets ICT Traders Favor
If you trade forex — the primary market in most ICT/SMC content — our data shows Fisher Transform ranked as the top indicator across 17 forex assets. DMI Direction came next at 3 assets. For commodities such as crude oil, Keltner Mean-Reversion led on 3 assets, with Laguerre RSI and Hammer each appearing on 2. These are not cherry-picked; they are the indicators that survived out-of-sample testing across the full test period.
Across the full universe, only 26% of all indicator/asset/timeframe combinations beat buy-and-hold. Even the indicators that work only work on a subset of assets. Browse by asset to see which indicator topped each market, or check the full indicator rankings.
Hypothetical Backtests — Not Financial Advice
Every result on IndicatorEdge is from hypothetical backtesting on historical data. Past backtest performance does not guarantee future results. Backtests cannot fully account for execution slippage beyond modeled costs, real-world liquidity constraints, or market regime changes that have not yet occurred. Nothing here is financial advice, and nothing here constitutes a recommendation to buy or sell any security, futures contract, or currency pair. Your trading decisions are your own responsibility.
Questions, answered
Why does a 71% win rate still fail to beat buy-and-hold?
Win rate only measures how often a trade closes in profit, not how much it earns per trade. If your average loss is larger than your average gain, a high win rate still produces a net negative expectancy. Buy-and-hold also captures every up-move during the holding period with no friction, so an active strategy needs to clear a meaningful bar — not just win more often than it loses.
Did any SMC or ICT indicator produce an edge in your backtests?
No. Across every SMC-derived signal we could encode as a mechanical rule and test, none beat buy-and-hold consistently across the 903-asset, 660,005-backtest universe. That covers all four tested timeframes: 1-Hour, 4-Hour, Daily, and Weekly.
Which indicators actually work for forex and commodities?
For forex, Fisher Transform ranked first across 17 assets — the dominant result in that class by a wide margin. For commodities, Keltner Mean-Reversion led on 3 assets, with Laguerre RSI and Hammer each on 2. These results are out-of-sample, not curve-fit to the same data used to identify them.
Are these real trading results?
No. All results are hypothetical backtests on historical data with modeled transaction costs. They do not represent real trades, real accounts, or a promise of future performance. This is not financial advice.
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.