The ICT Turtle Soup is a reversal setup built on intentional liquidity trapping: price sweeps a key level to trigger stop-losses from breakout traders before reversing sharply in the opposite direction. Unlike a generic stop hunt, the Turtle Soup follows strict market structure conditions and must occur within specific institutional Kill Zones.
What is the Turtle Soup in ICT trading?
The Turtle Soup is not a random stop hunt. It represents a recognizable pattern of institutional behavior that appears on the chart when specific structural conditions are met. Understanding its origin and characteristics allows traders to distinguish it from the countless false signals that trap uninformed market participants.
Origin of the concept (Michael Huddleston)
The concept was introduced by Michael Huddleston, an American professional trader known as "ICT" (Inner Circle Trader). Huddleston built a comprehensive methodology around Smart Money behavior (institutions, central banks, proprietary trading desks) and how they exploit available liquidity to fill large positions at optimal prices.
The "Turtle Soup" name is a direct reference to the Turtle traders of Richard Dennis and William Eckhardt from the 1980s. These traders used breakout strategies targeting the 20- and 55-period highs and lows. Huddleston identified that institutions know exactly where these stops are clustered and use them as liquidity reserves to execute large orders in the opposite direction. The Turtle Soup is therefore a "turtle trap": price breaks the expected breakout level, then reverses after collecting the Turtles' liquidity.
Turtle Soup vs generic stop hunt
The confusion between a Turtle Soup and a generic stop hunt is common among traders new to ICT. Here are the key distinctions:
| Criterion | Generic Stop Hunt | ICT Turtle Soup |
|---|---|---|
| Target level | Any stop-loss zone | Precisely the 20-candle high/low |
| Required context | None | Kill Zone + HTF bearish/bullish context |
| Confirmation | Not required | BOS or FVG mandatory after the raid |
| Timing | Any session | London or New York session only |
| Precision | Subjective | Defined rules, backtestable |
The fundamental difference: the Turtle Soup is a structured setup with defined entry conditions that can be systematically backtested. A generic stop hunt is a subjective interpretation that cannot be reliably quantified.
How to identify a Turtle Soup setup
Identifying a valid Turtle Soup requires verifying multiple conditions simultaneously before any entry. Missing one of them means trading an invalid signal.
Market conditions (liquidity raid + BOS)
A valid Turtle Soup builds in two distinct phases:
Phase one: the liquidity raid
Price makes an incursion above an equal high or a clear swing high formed by the last 20 candles (or below an equal low/swing low for a bullish Turtle Soup). This raid exceeds the level by one or two pips (or ticks on indices) intentionally, just enough to trigger the stops and absorb the orders of breakout traders.
Phase two: Break of Structure (BOS)
Immediately after the raid, price forms a Break of Structure in the opposite direction. This BOS confirms that liquidity has been collected and the real institutional move is beginning. It is the confirmation signal that distinguishes a Turtle Soup from a simple false breakout that resumes in the original direction.
For a deeper understanding of BOS and its role in SMC structure, see our guide on Break of Structure BOS in SMC.
Kill Zone timing requirement
Timing is a non-negotiable condition of the Turtle Soup. Valid setups occur almost exclusively during ICT institutional Kill Zones:
- London Kill Zone: 02:00 to 05:00 EST (07:00 to 10:00 UTC)
- New York Kill Zone: 08:30 to 11:00 EST (13:30 to 16:00 UTC)
- London Close: 10:00 to 12:00 EST (15:00 to 17:00 UTC)
These windows correspond to the highest-volume institutional periods when large positions are initiated and liquidated. A Turtle Soup forming outside these hours (during the Asian session or in the middle of the night) has significantly lower probability of follow-through.
For a complete guide on Kill Zones, read our ICT Kill Zones trading hours strategy.
M5 or M15 for Turtle Soup?
Most ICT traders identify Turtle Soup setups on M5 and M15 charts during Kill Zones. The directional context (HTF bias) is established on H1, H4, and Daily. A Turtle Soup on M5 aligned with the H4 bias has substantially higher follow-through probability than a counter-bias setup.
Confirmation with FVG or Order Block
After the BOS, the optimal entry occurs in a retracement zone identifiable by one of these elements:
Fair Value Gap (FVG): a price imbalance created by the structure-breaking candle. Price often partially returns to fill this gap before resuming its direction. This is the preferred retracement zone for Turtle Soup entries.
Order Block (OB): the last bearish candle before the bullish move (or the last bullish candle before the bearish move) that preceded the BOS. This zone represents the initial institutional accumulation.
The presence of a FVG or Order Block in the post-BOS retracement zone significantly increases setup quality. For more on the Fair Value Gap, read our guide on Fair Value Gap trading strategy.
Entry, stop loss, and take profit on Turtle Soup
Optimal entry level (OTE Fibonacci)
ICT recommends using the Fibonacci retracement on the post-BOS impulse move to identify the Optimal Trade Entry (OTE). Key levels are the 61.8% to 78.6% retracement zone. This zone frequently coincides with the FVG and represents the point where institutions finish filling their position before the directional move resumes.
Entry can be placed:
- At market as price approaches the OTE/FVG zone
- Via a limit order positioned within the FVG at the 61.8%-78.6% retracement level
- On lower timeframe confirmation (M1 or M2 confirmation within the FVG)
For the full OTE and ICT Fibonacci methodology, read our ICT Optimal Trade Entry and Fibonacci guide.
Stop loss placement
The stop loss on a Turtle Soup is always placed beyond the extreme of the liquidity raid:
- For a bearish Turtle Soup: above the raid wick by a few pips/ticks
- For a bullish Turtle Soup: below the raid wick by a few pips/ticks
This placement is logical: if price returns beyond the raid point, the setup is invalid and the market has likely continued in the initial breakout direction. Invalidation is clear and measurable.
Stop loss too tight: the most frequent mistake
A stop loss placed too close to the entry point (within market noise) is the primary cause of failure on Turtle Soup trades. The stop must clear the raid wick, not just the entry point. A poorly positioned stop creates negative mathematical expectancy even on a statistically sound strategy.
Targets: next liquidity pool
The natural targets for a Turtle Soup are the next liquidity pools in the direction of the trade:
- First target: the next price equilibrium (inefficiency) or an old HTF support/resistance area
- Second target: the next liquidity pool (previous equal lows/highs or Previous Day High/Low)
- Final target: the HTF "draw on liquidity" (the major objective of the current market cycle)
The minimum recommended risk-to-reward ratio on a Turtle Soup is 1:2. The best setups, in the correct directional context during optimal Kill Zones, can achieve ratios of 1:4 to 1:8.
Backtesting the Turtle Soup strategy
Building the rules in no-code
Backtesting the Turtle Soup is made accessible by visual tools like Backtrex that allow traders to define conditions without writing a single line of code. The conditions to encode are:
Identify HTF bias
Define Kill Zone filter
Detect the liquidity raid
Confirm the BOS
Entry in the FVG/OTE
Stop and target
The advantage of no-code backtesting is the ability to test these rules on 5 to 10 years of historical data in minutes, without selection bias. The visual approach eliminates implementation errors that can skew results when coding strategies manually.
Mandatory anti-repainting rule
In any Turtle Soup backtest, only use the close of the previously confirmed candle (close[1]), never the current candle's close[0]. Using the current bar creates look-ahead bias: the backtest displays unrealistic performance because it "sees" future data. Backtrex enforces this rule by default for all setups.
Key metrics to monitor
When backtesting a Turtle Soup, the priority metrics are:
| Metric | Definition | Indicative Turtle Soup threshold |
|---|---|---|
| Win rate | % of winning trades | 40-55% (offset by RR ratio) |
| Profit factor | Gross gain / Gross loss | > 1.5 acceptable, > 2.0 solid |
| Max drawdown | Maximum capital decline | < 15% in backtest |
| Expectancy | Win rate x avg RR - loss rate | Positive and stable > 0.3R |
| Frequency | Valid setups per week | 3 to 8 on EUR/USD M5 |
According to ESMA (European Securities and Markets Authority), between 74% and 89% of retail client accounts lose money when trading CFDs. Rigorous backtesting is the primary differentiator between traders who compound sustainably and those who fall into this statistic.
Typical results on Forex and indices
Based on ICT community backtests and historical data available on major Forex pairs, the Turtle Soup shows the following characteristics under real conditions:
- EUR/USD M5 in London Kill Zone: valid setups (with BOS confirmation + FVG) typically show win rates between 42% and 55% depending on filter strictness
- NAS100 (Nasdaq) M5 in New York Kill Zone: higher frequency due to increased volatility, similar win rate but potentially higher RR ratio
- Less liquid pairs (GBP/JPY, AUD/USD): less frequent setups but spread and slippage impact is greater; requires specific backtesting per pair
These figures are indicative and vary significantly with market conditions (trending vs ranging, macro environment). A backtest on your own configuration is essential before any live deployment.
Common mistakes and quality filters
False Turtle Soup patterns
The main identification errors that produce "false Turtle Soup" signals:
1. Raid without confirmed BOS: price breaks the key level but does not break inverse structure within the next 3-5 candles. No trade taken. Do not anticipate.
2. Setup outside Kill Zone: a raid and BOS identified outside London/New York windows have significantly lower follow-through probability. These setups may look technically valid but fail structurally more often.
3. No HTF context: trading a bearish Turtle Soup within a clearly bullish H4 structure means going counter-bias. Even a technically perfect setup has low probability of follow-through if the HTF context is opposed.
4. Raid too deep: if price significantly exceeds the key level (multiple tens of pips on EUR/USD), this is no longer a liquidity raid but a potentially valid breakout. The Turtle Soup is characterized by a minimal incursion, just enough to trigger the stops.
Market contexts to avoid
Certain contexts make the Turtle Soup unreliable even with perfect execution:
- Major macro releases (NFP, CPI, FOMC): exogenous volatility destroys the normal structure of setups. Avoid setups within 30 minutes before and after any major release.
- Lateral consolidating markets across all timeframes: without a clear directional bias on H4 or Daily, the Turtle Soup has no targeted "draw on liquidity," reducing expectancy.
- Low liquidity periods (US/UK holidays, year-end): institutional behavior is disrupted during low-participation periods. Setups from these periods are not representative of performance under normal conditions.
According to AMF (Autorite des marches financiers) data, more than 70% of retail traders lose money on leveraged products. A significant portion of these losses come from executing setups in unfavorable market contexts, without a systematic filtering protocol.
For guidance on avoiding backtesting mistakes that distort strategy evaluation, read our guide on prop firm backtesting rules.
Important Risk Warning
Conclusion
The ICT Turtle Soup is one of the most precise reversal setups in the Smart Money Concepts methodology, provided all its activation conditions are respected: a liquidity raid on the 20-candle high/low, immediate BOS confirmation, entry within the FVG/OTE during a Kill Zone, and HTF alignment. Its distinctive value over other SMC approaches is its backtestability: conditions are sufficiently defined to be encoded and rigorously tested over years of historical data.
Integrating the Turtle Soup into a no-code backtesting workflow via Backtrex allows traders to statistically validate the optimal parameters (timeframes, Kill Zones, HTF filters) before any capital commitment, removing the subjectivity inherent to all discretionary approaches.
A Turtle Soup is an ICT reversal setup where price makes an incursion above or below the high/low of the last 20 candles to trigger stop-losses from breakout strategy traders (the "Turtles"), before reversing sharply in the opposite direction. This liquidity trap is followed by a confirmatory Break of Structure and a retracement into a FVG or Order Block zone for the optimal entry.
A stop hunt is a generic term for any market move that triggers stops before reversing. The ICT Turtle Soup is a specific setup with strict conditions: the raid must target exactly the 20-candle high/low, it must occur during a Kill Zone (London or New York), and it must be confirmed by a Break of Structure within the following candles. Without these conditions, it is not a valid Turtle Soup.
Visual backtesting tools like Backtrex allow traders to define Turtle Soup conditions (liquidity raid, BOS, FVG, Kill Zone) as logical blocks without writing any code. The backtest then runs automatically over multiple years of historical data, generating the win rate, profit factor, and drawdown metrics needed to validate the strategy before deploying real capital.
Most ICT traders operate the Turtle Soup on M5 or M15 for setup identification and entry. The directional context (bias) is established on H1, H4, and Daily. A Turtle Soup identified on M5 that aligns with the H4 bias and Daily structure has significantly higher follow-through probability than a counter-bias setup.
Yes, the Turtle Soup is particularly effective on NAS100 (Nasdaq) and ES (S&P500) during the New York Kill Zone (08:30-11:00 EST). The volatility of US indices generates well-defined liquidity raids and clean Breaks of Structure, making setup identification clearer than on some minor Forex pairs.
Backtests on EUR/USD M5 in the London Kill Zone generally show win rates between 42% and 55% depending on filter strictness (BOS confirmation, HTF alignment, no news). The Turtle Soup is not a high win rate strategy, but its positive mathematical expectancy comes from risk-to-reward ratios of 1:2 to 1:5 on quality setups.
No setup is 100% guaranteed. The main causes of failure on a technically valid Turtle Soup are: the presence of a major macro release (NFP, CPI) that disrupts normal structure, an ambiguous HTF context without a clear bias, and low liquidity during holiday periods. Systematic backtesting helps identify the conditions in which the failure rate is highest so they can be filtered out in advance.