Quick Summary

Way of the Turtle: The Secret Methods that Turned Ordinary People into Legendary Traders

by Curtis M. Faith (2007)

Extended Summary - PhD-level in-depth analysis (10-30 pages)

Way of the Turtle: The Secret Methods that Turned Ordinary People into Legendary Traders - Extended Summary

Author: Curtis M. Faith | Categories: Trading Systems, Trend Following, Risk Management, Trading Psychology


About This Summary

This is a PhD-level extended summary covering all key concepts from "Way of the Turtle," the definitive insider account of the most famous trading experiment in financial history. This summary distills the complete Turtle methodology - from the ATR-based position sizing system and Donchian channel breakout entries to the deeper philosophical framework of thinking in probabilities, managing risk systematically, and designing robust trading systems. For AMT/Bookmap daytraders, the principles here are foundational: edge identification, risk normalization, cognitive bias management, and the discipline to execute under pressure. Every systematic trader should internalize these concepts as operating principles that transcend any single market or timeframe.


Executive Overview

"Way of the Turtle" is not merely a memoir of the legendary Turtle trading experiment. It is a comprehensive treatise on what it means to trade systematically, written by the most successful participant of what remains the most celebrated trading experiment ever conducted. Curtis M. Faith was selected at age 19 to join Richard Dennis's group of novice traders in 1983, and over the next four years he traded the largest Turtle account, generating over $31 million in profits for Dennis. His book, published in 2007, goes far beyond revealing the specific Turtle rules - which Faith had already released publicly in 2003 - to articulate the complete philosophical and methodological framework that made the Turtle experiment succeed.

The book's central argument is both simple and profound: successful trading is not about prediction. It is about identifying a genuine statistical edge, sizing positions to survive the inevitable variance, and maintaining the psychological discipline to execute the system through extended periods of drawdown. This thesis directly challenges the prevailing retail trading culture that fixates on finding the "right" entry, predicting market direction, and achieving a high win rate. Faith demonstrates - through both narrative and rigorous analysis - that the Turtles' edge came not from superior market prediction but from a systematic framework that exploited persistent trends created by human cognitive biases, combined with position sizing rules that made survival through drawdowns mathematically inevitable rather than merely hopeful.

What makes this book indispensable for modern traders - including those using AMT and Bookmap - is its treatment of the meta-principles that underlie all successful systematic trading. The specific Turtle rules (20-day and 55-day Donchian channel breakouts with ATR-based stops) are historically interesting but may have diminished efficacy in today's algorithmic markets. The principles behind those rules - edge identification, risk normalization, robust system design, anti-fragile position management, and cognitive bias awareness - are timeless. They apply equally to a trend-following futures trader in 1984, a discretionary Bookmap scalper in 2026, and every systematic approach in between.

The book also serves as a powerful case study in the nature-versus-nurture debate applied to trading. Richard Dennis bet William Eckhardt that trading could be taught. Eckhardt believed that successful traders were born with innate qualities. The Turtle experiment was designed to settle this bet, and Faith's account provides nuanced evidence that both men were partially correct: the rules could be taught to anyone, but only those with certain psychological dispositions could follow them consistently under the extreme pressure of managing millions of dollars.


Part I: The Turtle Experiment - Origins, Selection, and Training

Chapter 1: Risk Junkies

Faith opens with the audacious origin story. Richard Dennis, who turned a borrowed $400 into over $200 million trading commodities, believed that his trading methodology was entirely learnable. His partner William Eckhardt disagreed, arguing that successful trading required innate psychological traits that could not be taught. To settle this intellectual dispute, they designed an experiment: recruit ordinary people through newspaper advertisements, teach them a complete trading system, give them real money to trade, and observe whether they could replicate Dennis's success.

The advertisement ran in the Wall Street Journal, Barron's, and the New York Times. Over 1,000 applicants responded. Dennis and Eckhardt selected 13 for the first class (1983) and 10 for the second class (1984). The selection process involved a true/false test and personal interviews. Faith was among the youngest selected - a 19-year-old with programming experience but no trading background.

The name "Turtles" came from Dennis's visit to a turtle farm in Singapore, where he observed that turtles could be "grown" systematically. He told Eckhardt, "We are going to grow traders just like they grow turtles in Singapore."

Key Insight for Modern Traders: The Turtle experiment's most important finding was not about any specific trading rule. It was the demonstration that a complete, rule-based trading system could be transmitted from expert to novice and executed profitably - provided the novice had the discipline to follow the rules. This has direct implications for anyone developing or following a systematic trading approach today.

Chapter 2: Taming the Turtle Mind

The two-week training program covered the complete system: entries, exits, position sizing, market selection, and risk management. But Faith emphasizes that the most important component was not the mechanical rules but the mental framework Dennis taught. The Turtles were trained to think in terms of probabilities, to detach from individual trade outcomes, and to focus entirely on process.

Dennis's training emphasized several psychological principles that Faith would later recognize as countermeasures to well-documented cognitive biases:

Cognitive BiasDennis's CountermeasureHow It Was Trained
Loss AversionAccept losses as the cost of doing businessTurtles were told that 60-70% of trades would be losers
Recency BiasFocus on long-term expectancy, not recent resultsTraining included historical drawdown analysis
AnchoringIgnore entry price when making exit decisionsExits were rule-based, not based on profit/loss from entry
Outcome BiasJudge decisions by process, not resultsDennis praised trades taken correctly even when they lost
Disposition EffectCut losers, ride winnersMechanical stops on losers; trailing exits on winners
Gambler's FallacyEach trade is independentPosition sizing was based on volatility, not recent results

Faith describes the psychological intensity of the training period. During the two weeks, the Turtles took live trades as practice. Faith's first trade - a long position in heating oil - made approximately $2 million. This early success created its own psychological challenge: the fear of losing what had been gained, which is a manifestation of the endowment effect that Dennis specifically warned against.

Chapter 3: The First Turtle Trades

This chapter covers the transition from training to live trading. Each Turtle was given an account sized according to Dennis's assessment of their abilities. Faith received one of the largest allocations, reflecting Dennis's confidence in him despite his youth. The chapter describes the brutal reality of executing a trend-following system in real time: long periods of small losses punctuated by occasional large winners.

Faith recounts the psychological difficulty of the first major drawdown period. After the initial success, the system entered a period of whipsaws where multiple breakout signals failed. Many Turtles struggled during this period, some reducing their position sizes below what the rules prescribed (a violation) or skipping trades entirely. Faith managed to continue following the rules, largely because he trusted the mathematical framework Dennis had taught.

Critical Observation: The divergence in Turtle performance was not caused by different rules - all Turtles traded the same system. It was caused by different levels of adherence to the rules. This finding has been replicated across virtually every study of systematic trading: the execution gap between knowing what to do and actually doing it is the primary determinant of results.


Part II: Thinking Like a Turtle - The Philosophical Framework

Chapter 4: Think Like a Turtle

This chapter articulates the core mental model that distinguishes successful systematic traders from everyone else. Faith identifies several key principles:

1. Trade in the Present The past is irrelevant to the current trade. Whether you made money or lost money on the last trade has no bearing on the probability of the next trade's outcome. Turtles were trained to treat each trade as an independent event in a long series of probabilistic outcomes.

2. Think in Probabilities No single trade matters. What matters is the expected value of the trading system over hundreds or thousands of trades. A system with a 40% win rate and a 2:1 reward-to-risk ratio has positive expectancy, and the law of large numbers ensures that expectancy will be realized over time.

3. Separate Process from Outcome A good trading decision can produce a loss, and a bad trading decision can produce a profit. Evaluating decisions based on their outcomes rather than their process is a fundamental cognitive error that leads traders to abandon winning systems during normal drawdown periods.

4. Avoid the Illusion of Control Markets are not controllable. Traders control only three things: what they trade, how much they trade, and when they enter and exit. Everything else - including whether a particular trade makes money - is outside their control.

Faith introduces what he calls the Turtle Mindset Matrix, which categorizes traders along two dimensions: their focus (process vs. outcome) and their timeframe (short-term vs. long-term).

Short-Term FocusLong-Term Focus
Process-OrientedGood execution on individual trades but may lack strategic visionThe Turtle ideal: disciplined execution within a long-term framework
Outcome-OrientedGambler mentality: chasing wins, avoiding lossesDangerous: may stay disciplined but optimizes for wrong metrics

Chapter 5: Trading with an Edge

This is one of the most important chapters in the book and arguably contains Faith's most valuable contribution to trading education. He defines "edge" with mathematical precision and explains how to identify whether a trading system possesses one.

Definition of Edge: A trading edge exists when a trading system's expected value per trade is positive. Mathematically:

Edge = (Probability of Win x Average Win) - (Probability of Loss x Average Loss)

But Faith goes deeper. He argues that merely having a positive expected value is insufficient. The edge must be large enough to survive the variance inherent in the system. A system with a tiny edge and high variance will require such a large number of trades to realize its expectancy that most traders will go bankrupt from drawdowns before the edge manifests.

Faith identifies several sources of edge that the Turtle system exploited:

Sources of Trend-Following Edge:

SourceMechanismPersistence
Cognitive BiasesLoss aversion, anchoring, and herding create trends that persist beyond what efficient market theory would predictHigh - human psychology is stable
Institutional ConstraintsFund mandates, benchmark tracking, and regulatory requirements force institutional behavior that creates trendsModerate - evolves with regulation
Information AsymmetryDifferent market participants receive and process information at different speedsDeclining - technology reduces this
Liquidity ProvisionTrend followers provide liquidity to hedgers who are willing to pay a premium for risk transferHigh - structural feature of futures markets

Quote: "Good trading is not about being right, it's about trading right. If you want to be right, you can achieve a 100 percent hit rate by never placing a trade. Being right is irrelevant; making money is what counts."

Faith also introduces the concept of edge decay - the phenomenon where a profitable trading system gradually loses its edge as more participants discover and exploit the same inefficiency. This concept is critical for understanding why the specific Turtle rules may have diminished effectiveness today while the underlying principles remain valid.

Chapter 6: Falling Off the Edge

This chapter examines the cognitive biases that cause traders to abandon systems that have a genuine edge. Faith draws extensively on behavioral finance research, particularly the work of Daniel Kahneman and Amos Tversky, to explain why humans are psychologically predisposed to make poor trading decisions.

The Six Deadly Biases of Trading:

  1. Loss Aversion: Losses feel approximately 2.5 times more painful than equivalent gains feel pleasurable. This causes traders to cut winning trades short and let losing trades run - the exact opposite of what trend-following systems require.

  2. Sunk Cost Effect: Once a trader has invested time, money, or emotional energy into a position, they become reluctant to exit even when the evidence suggests they should. This manifests as "hoping" a losing trade will recover.

  3. Recency Bias: Recent events are weighted more heavily than older events in decision-making. After a string of losses, traders overweight the probability of future losses and reduce position sizes or skip trades, thereby destroying the system's edge.

  4. Anchoring: Traders anchor to specific price levels (entry prices, round numbers, recent highs/lows) and make decisions based on these anchors rather than on the current market conditions. This prevents them from following mechanical exit signals.

  5. Bandwagon Effect (Herding): Paradoxically, the same herding behavior that creates trends (and thus the Turtle edge) also causes traders to abandon their systems when those systems are out of favor. During drawdown periods, traders hear about other approaches that are currently working and are tempted to switch.

  6. Outcome Bias: Traders evaluate decisions based on their outcomes rather than their process. A trader who follows the rules and takes a loss is doing the right thing, but outcome bias makes this feel like a mistake.

Faith argues that awareness of these biases is necessary but not sufficient to overcome them. The Turtle solution was to make the system entirely mechanical, removing discretion entirely so that cognitive biases had no opportunity to influence decisions. For discretionary traders - including AMT/Bookmap practitioners - the implication is that trading plans must be designed with explicit rules that constrain discretionary decision-making at moments of maximum psychological pressure.


Part III: The Turtle Trading System - Mechanics and Methodology

Chapter 7: By What Measure?

Before detailing the system rules, Faith addresses the critical question of how to measure trading system performance. He argues that most traders use inadequate metrics and that proper performance measurement is essential for both system design and ongoing monitoring.

The Turtle Performance Measurement Framework:

MetricFormula/DescriptionWhat It RevealsLimitation
CAGRCompound Annual Growth RateOverall return generationIgnores risk entirely
Maximum DrawdownLargest peak-to-trough declineWorst-case pain thresholdSingle historical event; future drawdowns may be larger
MAR RatioCAGR / Maximum DrawdownReturn per unit of drawdown painSensitive to single worst drawdown
Sharpe Ratio(Return - Risk-free Rate) / Standard DeviationRisk-adjusted returnPenalizes upside volatility equally with downside
RAR% (Regressed Annual Return)Annualized slope of log equity curveSmoothed return measureRequires regression calculation
R-CubedRAR% / Standard Error of RAR%Risk-adjusted, regression-based returnComplex; not widely used

Faith advocates strongly for R-Cubed (Robust Risk-Adjusted Return) as the superior metric for evaluating trading systems. His argument is that R-Cubed captures not just the level of return and the level of risk, but also the consistency of the equity curve's trajectory. A system that produces high returns with wild swings in equity will have a low R-Cubed, while a system that produces steady, consistent returns will have a high R-Cubed even if the absolute returns are lower.

For modern AMT/Bookmap traders, the measurement framework is directly applicable. Even discretionary traders should track their equity curves and calculate risk-adjusted metrics. The specific metric matters less than the discipline of measuring performance rigorously rather than relying on subjective impressions of "how trading is going."

Chapter 8: Risk and Money Management

This chapter contains what many readers consider the book's most valuable technical content: the complete Turtle position sizing and risk management system. Faith presents this system as the primary reason the Turtles survived and thrived while many other trend followers were destroyed during difficult market conditions.

The N System (ATR-Based Position Sizing):

The cornerstone of Turtle risk management was a concept they called "N," which is essentially the Average True Range (ATR) calculated over a 20-day exponential moving average. N represented the average daily price movement of a given market and was used to normalize position sizes across different markets and different volatility regimes.

The position sizing formula:

Unit Size = (1% of Account Equity) / (N x Dollar Value Per Point)

This formula ensured that each "unit" of a position in any market represented approximately the same dollar risk. A unit of soybeans would have the same expected daily dollar fluctuation as a unit of Treasury bonds or a unit of crude oil. This normalization was critical because it prevented any single market from dominating the portfolio's risk profile.

Risk Limits - The Layered Constraint System:

LevelConstraintPurpose
Single MarketMaximum 4 unitsPrevents excessive concentration in one instrument
Closely Correlated MarketsMaximum 6 unitsLimits exposure to correlated moves (e.g., gold and silver)
Loosely Correlated MarketsMaximum 10 unitsControls broader sector exposure
Single DirectionMaximum 12 units long or 12 units shortPrevents excessive directional bias

These layered constraints created a hierarchical risk management framework that operated at multiple levels simultaneously. Even if a trader was correct about a directional bias, the position limits prevented the kind of catastrophic concentration that destroys accounts.

Pyramiding Rules:

The Turtles added to winning positions (pyramided) at intervals of 0.5N from the previous entry price, up to the maximum 4 units per market. Each additional unit had its stop placed at 2N from its own entry price, and all previous units' stops were also moved up to the new unit's stop level. This created a structure where:

  • The first unit risked 2% of equity (2N)
  • After the fourth unit was added, the first three units were in profit and the total position risk was concentrated on the most recent entry
  • If the position reversed by 2N from the last entry, all four units were stopped out simultaneously

This pyramiding approach is a sophisticated form of the "add to winners" principle. It allowed the Turtles to build large positions in strong trends while limiting the maximum loss on the total position.

Quote: "The secret of trading is that there is no secret. There are no magic systems that can divine the future."

Chapters 9-10: Trading System Building Blocks and the Complete Turtle System

Faith presents the complete Turtle trading system as a worked example of systematic trading design. The system had two variants: System 1 (shorter-term) and System 2 (longer-term).

The Complete Turtle Trading Rules:

ComponentSystem 1System 2
Entry - LongBuy on 20-day highest high breakoutBuy on 55-day highest high breakout
Entry - ShortSell on 20-day lowest low breakoutSell on 55-day lowest low breakout
Entry Filter (S1 only)Skip if prior breakout was a winnerNo filter
Stop Loss2N from entry2N from entry
Exit - Long10-day lowest low20-day lowest low
Exit - Short10-day highest high20-day highest high
Position Size1% ATR unit1% ATR unit
PyramidingAdd at each 0.5N in favor, max 4 unitsAdd at each 0.5N in favor, max 4 units

The System 1 Filter Explained:

System 1 had a unique feature: the Turtles were instructed to skip a 20-day breakout signal if the prior 20-day breakout had been profitable. The logic was that after a profitable breakout, the market might be in a range-bound period where breakouts were more likely to fail. However, if a System 1 signal was skipped and the market continued to trend, the 55-day breakout (System 2) would catch the trade. This ensured that no significant trend was missed entirely.

This filter is an example of what Faith later calls a "regime filter" - an attempt to distinguish between trending and mean-reverting market conditions. Modern implementations might use volatility measures, ADX, or market microstructure data from tools like Bookmap to achieve similar regime detection.

Market Selection:

The Turtles traded a diversified portfolio of highly liquid futures markets across multiple sectors:

  • Currencies: British Pound, Deutsche Mark, Swiss Franc, Japanese Yen, Canadian Dollar
  • Metals: Gold, Silver, Copper
  • Energy: Crude Oil, Heating Oil, Unleaded Gas
  • Interest Rates: T-Bonds, T-Notes, Eurodollars
  • Grains: Corn, Soybeans, Wheat, Soybean Oil, Soybean Meal
  • Softs: Coffee, Cocoa, Sugar, Cotton
  • Meats: Live Cattle, Feeder Cattle, Hogs (Lean Hogs)

The diversification across uncorrelated markets was integral to the system. Trend-following systems have low win rates (typically 35-45%) and depend on catching occasional large trends. By trading many markets simultaneously, the Turtles increased the probability that at least one market would be trending strongly at any given time.


Part IV: System Design, Robustness, and the Perils of Curve Fitting

Chapter 11: Lies, Damn Lies, and Backtests

This chapter contains Faith's most important contribution to trading education outside the Turtle rules themselves. He delivers a devastating critique of how most traders design and test trading systems, arguing that the vast majority of backtested systems are worthless because they are curve-fitted to historical data.

The Curve-Fitting Problem:

Curve fitting occurs when a trading system is optimized to perform well on historical data but has no genuine predictive power for future markets. Faith explains that this happens because traders confuse pattern recognition with genuine edge. Given enough parameters and enough historical data, a computer can always find a combination that would have been profitable in the past. But this "profitability" is an artifact of optimization, not evidence of a real market inefficiency.

Faith provides a memorable analogy: curve fitting is like trying to find your car keys under a streetlight because that is where the light is, rather than where you dropped them. The light represents historical data; the keys represent genuine market edge.

Degrees of Freedom and the Optimization Trap:

Faith introduces the concept of degrees of freedom in system design. Each parameter in a trading system represents a degree of freedom - a dimension along which the system can be adjusted to fit historical data. The more parameters a system has, the more likely it is to be curve-fitted.

Number of ParametersCurve-Fitting RiskExample
1-2LowSimple moving average crossover
3-5ModerateMoving average crossover with volatility filter and ATR stop
6-10HighMulti-indicator system with optimized lookback periods, thresholds, and filters
10+Very HighNeural network or machine learning models with many features

The Turtle system itself had remarkably few parameters: the breakout lookback period (20 or 55 days), the exit lookback period (10 or 20 days), the stop distance (2N), and the pyramid interval (0.5N). This simplicity was deliberate. Dennis understood that a simple system with few parameters was far more likely to be robust than a complex system optimized across many dimensions.

The Four Horsemen of Backtesting Apocalypse:

  1. Optimization Bias: The tendency to select parameter values that performed best in backtesting, which inevitably overfits to historical noise.

  2. Data Mining Bias: Testing many different system concepts on the same data set. If you test 100 different system ideas, some will appear profitable by chance alone. Faith notes that a 95% confidence level applied across 100 tests would be expected to produce 5 false positives.

  3. Survivorship Bias: Backtesting on markets that still exist today, ignoring markets that went to zero or were delisted. This inflates historical returns because the worst-performing instruments are excluded.

  4. Look-Ahead Bias: Using information in the backtest that would not have been available at the time the trade decision was made. Common examples include using adjusted price data, backtesting with split-adjusted stock prices, or using economic data that was later revised.

Chapter 12: On Solid Ground - Building Robust Systems

Having demolished naive backtesting, Faith presents his framework for designing systems that are genuinely robust - meaning they work across different markets, different time periods, and different market conditions.

The Robust System Design Framework:

PrincipleDescriptionImplementation
SimplicityFewer parameters reduce curve-fitting riskLimit system to 3-5 parameters maximum
UniversalitySystem should work across many markets without re-optimizationTest on diverse portfolio; do not market-specific optimize
Parameter StabilitySmall changes in parameters should not dramatically change resultsTest parameter sensitivity; avoid "peaks" in optimization landscapes
Out-of-Sample ValidationTest on data the system has never seenReserve 30-40% of data for out-of-sample testing
Monte Carlo SimulationTest system robustness under randomized conditionsRandomize trade order, skip random trades, add slippage
Walk-Forward AnalysisRepeatedly optimize on rolling windows and test on subsequent periodsSimulates real-world experience of optimizing and then trading

Faith emphasizes parameter stability as particularly important. If a 20-day breakout system is profitable but a 19-day or 21-day system is not, the 20-day result is almost certainly curve-fitted. Robust systems show smooth performance across a range of parameter values, indicating that the edge derives from a genuine market phenomenon rather than a specific numerical coincidence.

The Robustness Spectrum:

Faith conceptualizes system robustness as a spectrum rather than a binary characteristic:

LevelCharacteristicRiskExample
FragileWorks only with specific parameters on specific markets during specific time periodsExtreme"This RSI+MACD combo worked on ES from 2015-2018"
BrittleWorks across some conditions but fails under stressHighTrend system that works in normal volatility but fails in crises
ResilientWorks across most conditions with moderate degradation under stressModerateDiversified trend system with adaptive volatility filters
AntifragileActually improves under stress conditionsLowSystems that profit from volatility spikes and trend acceleration

The Turtle system was designed to be antifragile. Its best performance came during periods of extreme market movement - exactly the conditions that destroyed less robust systems. This is because breakout systems by nature require large moves to generate profits, and the largest moves occur during periods of market stress.


Part V: Psychology, Discipline, and Mastering the Demons

Chapter 13: Bulletproof Systems

Faith addresses the question of how to build systems that survive not just normal market conditions but regime changes - fundamental shifts in market structure that can render previously profitable systems worthless.

He identifies several types of regime changes:

  • Volatility regime shifts: Markets transitioning from low-volatility, mean-reverting environments to high-volatility, trending environments (or vice versa)
  • Correlation regime shifts: Markets that were previously uncorrelated becoming correlated during crises (the "all correlations go to one" phenomenon)
  • Structural regime shifts: Changes in market microstructure due to regulation, technology, or participant composition (e.g., the rise of algorithmic trading)

Faith's solution to regime change is not to try to predict it but to design systems that degrade gracefully when it occurs. This means:

  1. Trading multiple systems simultaneously (system diversification)
  2. Trading across multiple timeframes
  3. Trading across multiple markets
  4. Using conservative position sizing that can survive worst-case scenarios

For AMT/Bookmap traders, this principle translates directly to maintaining multiple playbooks for different market conditions (trending, bracketing, transitioning) and having explicit rules for when to deploy each playbook.

Chapter 14: Mastering Your Demons

The final major chapter returns to psychology, which Faith considers the ultimate determinant of trading success. He draws on his own experience and the divergent outcomes among the Turtles to illustrate how psychological factors overwhelm technical knowledge.

The Turtle Performance Divergence:

All Turtles traded the same system with the same rules. Yet their performance varied dramatically. Some Turtles made substantial profits; others barely broke even; a few lost money. The difference was entirely attributable to execution quality - specifically, the degree to which each Turtle followed the rules during psychologically difficult periods.

Faith identifies four "demons" that sabotage traders:

Demon 1: Fear Fear manifests as the reluctance to enter trades after a losing streak (missing the recovery), the urge to tighten stops beyond what the system prescribes (getting stopped out of good trades), and the tendency to reduce position sizes during drawdowns (reducing exposure precisely when the system is most likely to recover).

Demon 2: Greed Greed manifests as the urge to increase position sizes after a winning streak (increasing risk at the worst time), the desire to override exit signals on profitable trades (hoping for more profit), and the temptation to add parameters or filters to the system to capture more of the historical equity curve (curve fitting).

Demon 3: Hope Hope is the most insidious demon because it feels virtuous. It manifests as holding losing positions beyond the stop level ("it will come back"), adding to losing positions ("averaging down"), and continuing to trade a system that has clearly lost its edge ("it worked before, it will work again").

Demon 4: Regret Regret drives revenge trading, where a trader takes impulsive trades to "make back" losses. It also drives the abandonment of good systems after a drawdown, because the trader regrets not having switched to a different approach earlier.

Quote: "The Turtles were not taught complex systems. They were taught simple systems and the discipline to follow them."

Faith proposes a structured approach to psychological management:

The Demon Management Protocol:

StepActionPurpose
1IdentifyRecognize which demon is currently active
2AcknowledgeAccept the emotional experience without judgment
3SeparateDistinguish the emotional impulse from the correct systematic action
4ExecuteTake the action prescribed by the system, not the action suggested by the emotion
5RecordDocument the incident for later review
6ReviewPeriodically analyze demon incidents to identify patterns and develop countermeasures

Framework 1: The Edge-Execution Matrix

Faith's framework can be synthesized into a two-dimensional model that maps trading success along two axes: edge quality and execution quality.

Poor ExecutionGood Execution
Strong EdgeMediocre results; the edge exists but is destroyed by poor execution. Many Turtles fell here.Excellent results; the Turtle ideal. Faith and the top-performing Turtles operated here.
Weak EdgeCatastrophic results; no edge combined with poor execution guarantees failure. Most retail traders.Consistent small losses; disciplined execution of a system with no edge produces slow death by a thousand cuts.
No EdgeRapid account destructionSlow account destruction; at least capital is preserved longer

The key insight is that strong edge with poor execution often produces worse results than moderate edge with excellent execution. This is counterintuitive but is supported by the Turtle data. A Turtle who followed 80% of the rules with a strong edge underperformed a Turtle who followed 100% of the rules with the same edge, even though both had access to the identical system.


Framework 2: The System Robustness Evaluation Framework

Faith provides a multi-factor framework for evaluating whether a trading system is genuinely robust or merely curve-fitted:

Evaluation CriterionRobust SystemCurve-Fitted System
Parameter Count3-5 parameters10+ parameters
Parameter SensitivityStable performance across a range of valuesPerformance collapses with small parameter changes
Market DiversityWorks across multiple markets and asset classesOnly works on the specific markets tested
Time Period StabilityWorks across different historical periodsWorks only on the optimization period
Out-of-Sample PerformancePerformance degrades modestly out-of-samplePerformance collapses out-of-sample
Monte Carlo ResultsConsistent across randomized scenariosHighly variable under randomization
Theoretical FoundationGrounded in a plausible market inefficiencyNo clear explanation for why it should work
Walk-Forward ConsistencyConsistent results across walk-forward windowsInconsistent results across windows

Faith argues that the theoretical foundation criterion is often overlooked but is among the most important. A system must have a plausible story for why the edge exists. The Turtle system exploited trend persistence caused by human cognitive biases and institutional constraints. This is a well-documented market phenomenon with a clear theoretical basis. A system that finds an arbitrary pattern (e.g., "buy on the third Tuesday after a full moon") may backtest well but has no theoretical foundation and is almost certainly curve-fitted.


Framework 3: The Position Sizing Decision Tree

Faith's ATR-based position sizing system can be generalized into a decision framework applicable to any trading approach:

Step 1: Define Maximum Portfolio Risk

  • Determine the maximum percentage of account equity you are willing to lose in a single day or on a single trade
  • Turtle default: 2% per unit (stop at 2N)

Step 2: Normalize Risk Across Markets

  • Calculate each market's volatility (ATR or equivalent)
  • Size positions so that each position contributes the same dollar risk to the portfolio
  • This prevents high-volatility markets from dominating portfolio risk

Step 3: Apply Correlation Constraints

  • Group markets by correlation
  • Limit total exposure within each correlation group
  • Prevent the portfolio from becoming a single concentrated bet disguised as diversification

Step 4: Apply Directional Constraints

  • Limit total long and total short exposure
  • Prevent excessive directional bias

Step 5: Scale for Drawdown Management

  • During drawdown periods, the ATR-based system automatically reduces position sizes because account equity has declined
  • This creates a natural anti-fragile response: losing money reduces future risk exposure
Decision PointQuestionAction if YesAction if No
New signal receivedDoes the trade fit within single-market limits?Proceed to next checkSkip trade
Correlation checkDoes the trade fit within correlated-market limits?Proceed to next checkSkip trade
Direction checkDoes the trade fit within directional limits?Proceed to next checkSkip trade
Volatility checkHas volatility changed significantly since last calculation?Recalculate N and position sizeUse existing position size
Equity checkHas account equity changed significantly?Recalculate unit size based on new equityUse existing unit size

Comparison: Turtle Trading vs. Modern Systematic Approaches

The following table compares the original Turtle approach to modern systematic trading methodologies, highlighting what has changed and what remains timeless:

DimensionTurtle System (1983-1988)Modern Systematic Trading (2020s)What ChangedWhat Stayed the Same
Entry SignalsDonchian channel breakouts (20/55-day)Machine learning models, order flow analysis, microstructure signalsEntry methodology has become far more sophisticatedThe principle that entries should be based on a quantifiable edge
Exit StrategyTrailing channel exits (10/20-day)Dynamic exits based on volatility, regime detection, order flowExit sophistication has increasedThe principle that exits should be rule-based, not discretionary
Position SizingATR-based (N system), 1% risk per unitVolatility-targeting, Kelly criterion, risk parityMathematical frameworks have become more sophisticatedThe principle that risk must be normalized across positions
Risk ManagementLayered unit limits (4/6/10/12)Portfolio-level VaR, stress testing, factor-based risk modelsRisk models are more granular and dynamicThe principle that risk must be constrained at multiple levels
Market Selection24 liquid futures marketsGlobal multi-asset class, including crypto, options, FXUniverse has expanded dramaticallyThe principle of diversification across uncorrelated markets
ExecutionOpen outcry pit trading, manualAlgorithmic execution, smart order routing, DMAExecution has been almost entirely automatedThe principle that execution quality matters for realized edge
BacktestingLimited computing power, simple testsCloud computing, tick-level data, walk-forward optimizationComputing power has increased exponentiallyThe principle that backtests must guard against curve fitting
PsychologyEntirely manual, willpower-basedAutomated systems reduce psychological burdenAutomation removes much of the psychological challengeThe principle that human psychology is the primary obstacle
DataDaily OHLC price dataTick-level, order book, footprint, Bookmap heatmapData granularity has increased by orders of magnitudeThe principle that trading decisions should be based on market-generated information
TimeframeDays to weeks (position trading)Microseconds to months (full spectrum)Timeframes have expanded in both directionsThe principle that the system must match the trader's timeframe capability

Critical Analysis

Strengths of "Way of the Turtle"

1. Unmatched Authenticity Faith's insider perspective on the Turtle experiment cannot be replicated by any other author. He was not an observer or a journalist reconstructing events from interviews. He was the most successful participant, and his account carries the authority of direct experience. This makes the book a primary source document for one of the most important events in trading history.

2. Integration of Theory and Practice Many trading books are either theoretical (academic treatments of market efficiency, behavioral finance, etc.) or practical (here are the rules, follow them). Faith integrates both. He explains the theoretical foundations of trend following (cognitive biases, market microstructure) while simultaneously providing the exact mechanical rules that implemented those theoretical insights in practice.

3. Exceptional Treatment of Backtesting and System Design Chapters 11 and 12 on backtesting pitfalls and robust system design are among the best available in any trading book. Faith's explanation of curve fitting, degrees of freedom, and parameter sensitivity analysis is rigorous, accessible, and directly actionable. Many traders who read these chapters report that they fundamentally changed how they evaluate trading systems.

4. Honest Treatment of Psychology Faith does not present himself as a psychological paragon. He admits to struggles with the demons he describes and acknowledges that even he did not always follow the rules perfectly. This honesty makes his psychological advice more credible and more useful than the typical trading book's advice to "just be disciplined."

Weaknesses and Limitations

1. Historical Specificity The Turtle experiment occurred in a specific market environment (1983-1988) characterized by strong commodity trends, low algorithmic participation, and wide bid-ask spreads. The specific rules may not work as effectively in modern markets, though Faith acknowledges this. Some readers may struggle to separate the timeless principles from the time-bound specifics.

2. Post-Turtle Career Faith's career after the Turtle experiment was less distinguished than his Turtle performance. He founded a software company that eventually failed, and his subsequent trading results were not publicly documented. While this does not invalidate the lessons in the book, it raises questions about the durability of the Turtle approach and Faith's ability to adapt to changing markets.

3. Survivorship Bias in the Narrative The Turtle experiment itself may suffer from survivorship bias. Dennis selected participants who fit his criteria, trained them in his methodology, and gave them capital during a period favorable for trend following. The experiment does not prove that trend following always works or that anyone can be a successful trader - it proves that selected individuals taught a specific system during a favorable period could be profitable.

4. Limited Treatment of Alternative Approaches Faith presents trend following as the optimal approach to systematic trading and devotes relatively little attention to mean-reversion strategies, market-making approaches, or hybrid methodologies. This creates a somewhat one-dimensional view of systematic trading that may not serve readers exploring other paradigms.

5. Organizational Inconsistency The book alternates between memoir, philosophy, technical instruction, and behavioral finance without always signaling these transitions clearly. Some readers find this engaging; others find it disorienting. A more structured organization might have made the technical content more accessible.


Trading Takeaways for AMT/Bookmap Daytraders

While the Turtle system was designed for multi-day to multi-week position trading, its principles translate directly to intraday trading with AMT and Bookmap tools:

1. Edge Identification Through Order Flow

The Turtle edge derived from trend persistence caused by cognitive biases. AMT/Bookmap traders can identify analogous edges in the order book: absorption at key levels, iceberg order detection, aggressive buying/selling imbalances, and delta divergences. The principle is identical - find a quantifiable, repeatable pattern that produces a positive expected value over many trades.

2. ATR-Based Position Sizing for Intraday Trading

The N system translates directly to intraday trading by using a shorter ATR period (e.g., 14-period ATR on a 5-minute chart). Normalizing position sizes by intraday volatility ensures that each trade carries approximately the same dollar risk regardless of the instrument's current volatility regime.

3. Correlation Awareness

AMT/Bookmap traders who trade multiple instruments (e.g., ES, NQ, YM simultaneously) should apply the Turtle correlation constraint principles. Holding maximum long positions in all three equity index futures is not diversification - it is a single concentrated bet on the equity market.

4. Process Journaling

The Turtle emphasis on process over outcome applies directly to discretionary AMT/Bookmap trading. Traders should journal not just their P&L but the quality of their decision-making process: Was the trade based on a valid AMT setup? Was position sizing appropriate? Was the exit executed according to plan?

5. Regime Detection

The Turtle System 1 filter (skip breakout if prior breakout was profitable) was a primitive regime detection mechanism. Modern AMT/Bookmap traders have far superior tools for regime detection: Bookmap's heatmap reveals whether the market is in a trending (stacked orders on one side) or bracketing (heavy orders on both sides) regime. Trading the right playbook for the current regime is the AMT equivalent of the Turtle filter.

6. Drawdown Management

The automatic position reduction during drawdowns (because unit size is based on current equity, and equity has declined) is a principle that every intraday trader should adopt. Many daytraders increase size after losses to "make it back" - the exact opposite of what the Turtle system prescribes.

7. Multi-Timeframe Context

Just as the Turtles used both 20-day and 55-day breakout systems to capture trends at different timescales, AMT/Bookmap traders should maintain context across multiple timeframes. A Bookmap scalp should be taken in the direction indicated by the higher-timeframe AMT auction structure, not against it.


The Turtle Trader's Self-Assessment Checklist

Use this checklist to evaluate whether you are applying the Turtle principles to your own trading:

Edge and System Design:

  • I can articulate my trading edge in one sentence with a clear theoretical basis for why it should persist
  • My trading system has fewer than 5 optimizable parameters
  • I have tested my system on out-of-sample data (data it has never seen during development)
  • Small changes in my parameters do not dramatically change my system's performance
  • My system works across multiple markets or instruments without re-optimization
  • I have conducted Monte Carlo simulation to understand my system's worst-case drawdown
  • I can explain the source of my edge (cognitive biases, structural features, liquidity dynamics) without referencing historical backtest results

Risk Management:

  • I size positions based on current volatility, not a fixed dollar amount or fixed number of shares/contracts
  • I have predefined maximum risk limits at the single-trade, daily, and portfolio levels
  • I track correlation between my positions and limit exposure to correlated bets
  • I reduce position sizes automatically when my account is in drawdown
  • I never risk more than a predefined percentage of my account on any single trade
  • I have a maximum daily loss limit after which I stop trading for the day

Psychology and Execution:

  • I follow my trading rules on at least 90% of signals, including during drawdown periods
  • I evaluate my trading decisions based on process quality, not outcome
  • I maintain a trading journal that records both the trade details and my psychological state
  • I can identify which "demon" (fear, greed, hope, regret) is most likely to affect me in specific situations
  • I do not increase position sizes after a losing streak to "make back" losses
  • I do not abandon my system after a drawdown that falls within historically normal parameters
  • I take breaks from trading when I recognize that emotional decision-making has taken over

Performance Measurement:

  • I calculate risk-adjusted returns (Sharpe, MAR ratio, or equivalent), not just raw P&L
  • I compare my actual performance to the theoretical performance of my system to identify execution gaps
  • I review my performance monthly and quarterly with statistical rigor, not subjective impressions
  • I track my win rate, average win, average loss, and expectancy per trade
  • I monitor whether my system's edge is decaying over time by tracking key performance metrics on a rolling basis

Key Quotes with Commentary

"Good trading is not about being right, it's about trading right."

This is perhaps the most important sentence in the book. It encapsulates the shift from prediction-based trading (which is a form of fortune-telling) to process-based trading (which is a form of probability management). A trader who is "right" 80% of the time but lets losers run and cuts winners short will underperform a trader who is "right" 35% of the time but cuts losers quickly and rides winners far.

"The secret of trading is that there is no secret."

Dennis told this to the Turtles during training, and Faith uses it to dismantle the mystique around trading success. The Turtle rules were simple - embarrassingly simple by the standards of modern quantitative finance. The power was not in the complexity of the rules but in the discipline of execution and the soundness of the risk management framework.

"The Turtles were not taught complex systems. They were taught simple systems and the discipline to follow them."

This quote directly challenges the prevalent assumption that more sophisticated systems produce better results. In Faith's experience, the opposite was often true. Complex systems were more likely to be curve-fitted, harder to execute consistently, and more prone to breaking down in unfamiliar market conditions.

"Don't worry about where the prices are going. Worry about what you are going to do when they get there."

This reframes the trader's task from prediction to preparation. Instead of asking "Will the market go up or down?" the Turtle asks "What is my plan if the market goes up? What is my plan if it goes down? What is my plan if it goes sideways?" This preparation-based approach is directly applicable to AMT/Bookmap trading, where the trader identifies key levels and predefined responses before the market reaches those levels.

"Losers average losers."

Attributed to Paul Tudor Jones but echoed by Faith, this maxim directly contradicts the instinct to add to losing positions. The Turtle system explicitly prohibited adding to losers and required adding to winners (pyramiding). This is psychologically difficult because adding to a winning position feels risky (what if it reverses?), while adding to a loser feels rational (the price is lower, so it is a better deal). The Turtle framework recognizes this as a cognitive trap.


Connection to Broader Trading Literature

"Way of the Turtle" sits at the intersection of several important threads in trading literature:

Trend Following Canon: The book is part of the trend-following tradition alongside Michael Covel's "Trend Following," Ed Seykota's writings, and the works of Jerry Parker and other CTAs. Faith's contribution is unique because he provides the specific rules and the insider perspective that other trend-following books often lack.

Behavioral Finance Application: Faith's treatment of cognitive biases draws on the academic work of Kahneman, Tversky, and Thaler but applies it specifically to trading contexts. This makes the book a practical bridge between behavioral finance theory and trading practice.

System Design Methodology: The chapters on backtesting, curve fitting, and robustness place the book in conversation with Robert Pardo's "The Evaluation and Optimization of Trading Strategies," Perry Kaufman's "Trading Systems and Methods," and David Aronson's "Evidence-Based Technical Analysis."

Trading Psychology: The psychological framework connects to Mark Douglas's "Trading in the Zone," Brett Steenbarger's "The Psychology of Trading," and Denise Shull's "Market Mind Games." Faith's approach is less therapeutic than these authors but more integrated with systematic trading methodology.


Further Reading

BookAuthorWhy Read ItRelationship to "Way of the Turtle"
"Trend Following"Michael CovelComprehensive overview of the trend-following industry and its practitionersBroader context for the trend-following philosophy Faith practiced
"The Complete TurtleTrader"Michael CovelAlternative account of the Turtle experiment from an outsider's perspectiveDifferent perspective on the same events; useful for triangulation
"Trading in the Zone"Mark DouglasDeep exploration of the probabilistic mindset Faith describesExpands on the psychological framework; more depth on mental models
"Market Wizards"Jack SchwagerInterviews with top traders including Richard DennisDennis's own words about the Turtle experiment and his trading philosophy
"The Evaluation and Optimization of Trading Strategies"Robert PardoRigorous treatment of walk-forward analysis and system validationExtends Faith's chapters on backtesting with more mathematical depth
"Evidence-Based Technical Analysis"David AronsonStatistical framework for evaluating trading signalsFormalizes the anti-curve-fitting arguments Faith makes
"Thinking, Fast and Slow"Daniel KahnemanThe foundational work on cognitive biases that Faith referencesProvides the academic foundation for Faith's psychological framework
"Markets in Profile"James DaltonAuction Market Theory framework for understanding market structureComplementary framework; AMT provides the market reading that the Turtle system replaced with mechanical rules
"Mind Over Markets"James DaltonMarket Profile fundamentals and day type classificationFoundation for understanding how markets organize themselves through auctions
"Antifragile"Nassim Nicholas TalebFramework for building systems that benefit from disorderTheoretical basis for Faith's concept of bulletproof (antifragile) trading systems

Conclusion

"Way of the Turtle" is one of the essential texts in the canon of systematic trading literature. Its value lies not in the specific Turtle rules - which are historical artifacts of a particular era in commodity trading - but in the meta-principles that those rules embodied. Edge identification, risk normalization through volatility-based position sizing, parameter-sparse system design, out-of-sample validation, and the psychological discipline to execute through drawdowns are principles that apply to every systematic approach, from multi-week trend following to intraday AMT/Bookmap scalping.

Faith's unique contribution is the integration of these principles into a coherent framework, validated by one of the most famous experiments in trading history. The book demonstrates that trading success is not about finding the perfect entry signal or the magic indicator. It is about building a complete system - encompassing entries, exits, position sizing, risk limits, and psychological management - that has a genuine edge and can be executed consistently over thousands of trades.

For AMT/Bookmap daytraders, the most actionable takeaway is this: your edge may come from order flow analysis, delta divergences, or absorption patterns rather than from Donchian channel breakouts. But the framework around that edge - how you size positions relative to volatility, how you constrain risk at multiple levels, how you validate your approach against curve fitting, and how you maintain discipline through inevitable losing periods - should follow the principles Faith articulates. The technology has changed. The markets have changed. Human psychology has not. And it is human psychology - both the cognitive biases that create edges and the cognitive biases that prevent traders from exploiting them - that remains the fundamental force driving all market outcomes.

The Turtle experiment proved that trading can be taught. "Way of the Turtle" teaches it. Whether you internalize and execute the lessons is, as Dennis would say, entirely up to you.

Log in to mark this book as read, save it to favorites, and track your progress.

GreenyCreated by Greeny