Key Takeaways
Build from an observed market premise, not a clever indicator.
Start with a hypothesis. Kaufman treats technical and fundamental analysis as complementary. Fundamentals explain forces such as supply, demand, and policy; technical rules help time decisions and define manageable risk. Prices may anticipate news before it becomes public, so a trading idea should begin with an observable market behavior, not a convenient formula. The book's goal is to help readers choose and test a program, not crown one method universally best.
Make the system inspectable. A complete trading program must specify its data, time frame, signals, trade selection, execution costs, risk controls, and monitoring. A simple premise is easier to test than a complicated one, and each component should be evaluated before components are combined.
What's valuable here is the separation of explanation from prediction. A story about why prices should move is not itself evidence that a rule can profit from the move. The discipline resembles scientific model building: define a claim, specify what would disconfirm it, and test it against observations not used to invent it. Markets also adapt. Once a pattern becomes widely traded, its profitability can shrink, so a sound rationale is a starting point, not permanent proof.
Treat every signal as probability, and every estimate as uncertain.
Measure the spread of outcomes. The standard deviation describes how widely observations cluster around their average. In a normal distribution, about 68% of values lie within one standard deviation, 95.5% within two, and 99.7% within three. These are reference points, not promises about market prices, which can be skewed and unusually extreme.
Respect the sample. A few trades cannot establish a system's likely performance. The book warns that small samples carry substantial error, and illustrates why frequency distributions or median values can be more useful than standard deviation when prices have long tails. A price series with a brief, dramatic spike can distort its average and make ordinary conditions look unlike themselves.
This is a practical lesson in statistical humility. A probability estimate is only as useful as its assumptions and sample. Financial returns often have fatter tails than a normal curve predicts, so the apparent rarity of a large move can be overstated. A median or empirical frequency chart helps expose skew, but neither removes the problem of limited history. The strongest interpretation is comparative: quantify uncertainty, then avoid treating any calculated probability as a guarantee.
Choose the tool only after naming the price behavior to capture.
Match method to movement. A moving average smooths prices to reveal direction, while a breakout method waits for price to exceed a prior high or low. Countertrend methods instead seek reversals, but still need a view of the prevailing trend. These approaches answer different questions and carry different risks. A tool's complexity does not make it more effective.
Choose the time frame deliberately. Five-minute data can contain substantial noise; weekly data emphasizes longer movements and may already dictate a slower style. A short-term oscillator can help time an entry into a longer-term trend, but it should not be mistaken for the trend itself. The underlying observation should determine the tool and the interval.
This is a model-selection problem, not an indicator beauty contest. In statistical learning, a model that is too simple misses structure, while one that is too intricate fits noise. The practical task is to identify the behavior that matters, then use the least complicated method that captures it. Multiple time frames help because they separate direction from timing, but their signals should be treated as related evidence, not independent confirmation.
Trade crop seasonality as a changing supply calendar, not a fixed recipe.
Start with the crop cycle. Grain prices often face harvest pressure when newly produced supply reaches the market, then become sensitive to planting and growing-season weather. Soybeans and corn can therefore show recurring low-price periods near harvest and greater uncertainty during crop development. Storage costs and export demand also shape prices, so the pattern is not simply a calendar effect.
Check whether the pattern persists. Seasonality can be estimated from monthly averages, percentage changes, or prices adjusted for trend. Medians can reduce distortion from a few extreme years. The book also separates bullish, bearish, and nonseasonal years, showing why a long-run average may conceal sharply different annual paths. Changing production, including South American competition, can alter familiar patterns.
Seasonality is strongest when tied to a physical mechanism, such as planting, storage, or harvest, rather than an unexplained month-of-year effect. That makes it a useful prior expectation, not a trading command. Climate, technology, trade, and policy can shift the pattern. This resembles causal forecasting: the calendar suggests where to look, while current supply evidence determines whether the historical tendency is relevant this year.
Use price structure for levels, and volatility for room to trade.
Mark the market's boundaries. Chartists use support and resistance to identify prices where earlier advances or declines stalled. A channel places parallel boundaries around a move; a breakout beyond them may signal a change in direction. Point-and-Figure charting records meaningful price changes without assigning equal importance to elapsed time. Its box size and reversal requirement determine both signal sensitivity and the risk of a trade.
Allow for ordinary fluctuation. A band around a trend can filter minor penetrations, but wider bands produce fewer signals and increase the distance to the entry. Narrow bands react sooner yet invite false breakouts. Price levels are guides, not walls: a market can pass through support or resistance when new information overwhelms prior trading behavior.
These tools connect chart reading with market microstructure. Orders often cluster near visible highs, lows, and round levels, which can make those areas matter in practice. Yet clustered orders can also create abrupt breakouts once liquidity is consumed. A boundary is therefore both a potential barrier and a place where movement may accelerate. The useful question is not whether a level is certain, but how much risk is justified if it fails.
Read volume as participation evidence, never as a verdict.
Separate activity from direction. Rising volume during a price move is traditionally treated as confirmation, while open interest indicates outstanding futures positions. Neither measure proves who is right or which way prices must go. Individual contract volume can rise or fall as traders roll positions into a nearer delivery, so total market volume is often more informative for studying overall activity.
Account for the clock. Volume follows intraday patterns, with activity often concentrated near the opening and close. Tick volume, the number of price changes, can serve as a timely but imperfect proxy when actual volume is unavailable. Volume indicators such as On-Balance Volume assign activity according to price direction, but their usefulness still depends on testing and interpretation.
Volume is evidence about participation, not a transparent record of intention. Every futures trade has both a buyer and a seller, so an increase cannot identify which side is better informed. This is a market-microstructure distinction: transaction activity measures interaction, while price change measures the result. Volume can strengthen a price-based argument, but using it as an independent vote risks counting the same market event twice.
Adapt to volatility and time frame, but distrust a perfect fit.
Change speed when conditions change. Kaufman's Adaptive Moving Average uses an Efficiency Ratio, comparing net price movement with the sum of absolute price changes over the same interval. A directional move produces a higher ratio and a faster trend; erratic movement produces a lower ratio and a slower one. Other adaptive methods vary their speed using relative volatility or momentum.
Keep direction and timing distinct. A longer time frame can identify the broad trend while a shorter one supplies an entry. The book's multiple-screen examples combine these roles rather than asking one indicator to do everything. Adaptive calculations can respond to changing conditions, but their periods and thresholds still need testing across different markets and regimes.
Adaptive methods resemble feedback control: a system changes its response speed as the environment changes. That can reduce lag in directional markets and avoid overreacting to noise. The risk is added flexibility. Each new parameter gives an optimization more ways to fit historical accidents. Adaptation is most credible when its rule has a clear rationale and remains stable across distinct periods, rather than merely delivering a better backtest.
Test execution and price shocks, not just a beautiful backtest.
Test for robustness. Kaufman recommends a clear hypothesis, verified calculations, out-of-sample data, and broad testing rather than searching only for the single best result. In one comparison, six trend systems were tested on Eurodollars over ten years; performance generally declined as trading became more frequent. That finding highlights how costs and market noise can erase the apparent edge of fast strategies.
Include the real trade. Commissions, slippage, order timing, liquidity, and monitoring all affect results. A price shock can dominate a historical record, making a system look more profitable or less risky than it would be in live trading. After trading begins, actual fills must be compared with expected performance, and the rules or position size adjusted when the difference is persistent.
Backtesting is a form of measurement, not a certificate of future success. Trying many parameter combinations creates a multiple-comparisons problem: some will look exceptional by chance. Out-of-sample checks help, but repeated feedback from the test period gradually turns it into part of the fitting process. A credible strategy needs a plausible mechanism, realistic execution assumptions, and performance that survives more than one market regime.
Protect capital before chasing returns; diversify genuinely different risks.
Size for survival. Futures leverage can make a small price change produce a large gain or loss relative to deposited margin. Risk control therefore begins with the amount committed to each trade, the liquidity of the market, and the capital available to withstand losses. The book discusses individual trade risk, leverage, probability of ruin, and the compounding of positions as connected decisions.
Diversify by behavior. Holding several markets or trading methods can reduce risk when their returns are genuinely different. Simply adding more contracts from closely related markets may provide little protection. The book's central warning is practical: a strategy can become ruinous through excessive leverage, while even careful risk management cannot turn a poor trading premise into a good one.
This is a question of survival, not bravado. Expected return matters only if the investor can endure the path required to earn it. Portfolio theory formalizes the value of combining assets whose returns do not move together, but correlations can rise during market stress. Diversification should therefore be tested under adverse conditions, and position size should reflect uncertainty, not just the margin demanded by an exchange.
Analysis
Trading Systems and Methods is best understood as a technical reference organized around a practical question: how can a trader turn observations about prices into a tested program without confusing a historical pattern for a dependable edge? Its sweep is unusually broad, moving from averages, regression, and probability to chart formations, cycles, seasonality, spreads, volume, adaptive methods, artificial intelligence, testing, and capital control. That breadth is the source of its value and its central summarizing difficulty. It is not a single recipe but a map of competing ways to interpret market behavior.
The book's strongest contribution is methodological. It repeatedly asks the reader to state a premise, isolate components, test them independently, account for costs, and monitor live performance. The advice that complexity does not imply superiority is especially important in an era when computing power can test thousands of combinations. A trading rule can look persuasive because the past has been searched until it supplies a flattering example. The corrective is not to reject computation, but to treat it as a disciplined experiment with explicit assumptions and a skeptical reading of results.
The treatment also makes clear that markets are not stationary laboratory systems. Participants change, policies change, liquidity changes, and a method's success can depend on the regime in which it is applied. Trend following and countertrend trading are not contradictory doctrines so much as different responses to different price structures. Multiple time frames, volatility adjustment, and diversification are attempts to manage that instability, though each introduces assumptions that should themselves be tested.
A useful modern reading is to distinguish three questions that trading discussions often blur: Does a pattern exist? Can it be detected before it ends? Can it be traded after costs and risk? Kaufman's wide catalogue helps generate hypotheses, but only the latter two establish practical value. The enduring lesson is epistemic as much as financial: measurement can improve judgment, yet no measurement eliminates uncertainty or substitutes for controlled exposure.
Review Summary
Readers find Trading Systems and Methods comprehensive but dense. Many appreciate its breadth and depth, particularly for experienced traders seeking advanced strategies. Some note it's better as a reference than a cover-to-cover read. The book's technical focus appeals to some but may be overwhelming for others. Reviewers praise its thoroughness in covering technical analysis tools, though some wish for more guidance on application. Overall, it's seen as a valuable resource for technically-driven traders, albeit requiring significant time investment to fully utilize.
People Also Read
Glossary
Efficiency Ratio
Measures directional efficiencyA ratio comparing the net price change over a chosen interval with the sum of the absolute price changes within it. Values near one indicate movement mostly in one direction; values near zero indicate substantial back-and-forth movement. Kaufman uses it to adjust a moving average's speed, making the average faster in directional markets and slower in noisy ones.
Directional Movement
Separates upward and downward movementA Wilder method that compares changes in successive daily highs and lows to estimate upward and downward directional pressure. The values are smoothed and scaled relative to a true-range measure. They can be used to assess trend direction and strength, or combined with the Parabolic Time/Price System in a trading strategy.
Parabolic Time/Price System
Accelerating stop-and-reverse trend systemA trend-following method in which a stop-and-reverse level advances toward price as a trade becomes profitable. Its acceleration factor increases when prices make new extremes, reducing the lag of the stop. The system is designed to stay in the market and reverse when the stop is penetrated.
Point-and-Figure
Charts price changes without timeA charting method that records price movement in columns of Xs and Os, using a chosen box size and reversal requirement. Time is not plotted at equal intervals. The box size controls sensitivity, while the number of boxes required for reversal helps define a trend change and the associated trade risk.
Market Profile
Maps time spent at pricesA way of organizing intraday market activity by marking the prices traded during successive time periods. The resulting distribution highlights areas where trading persisted, called value areas, and distinguishes them from prices that were visited briefly. The method uses these patterns to interpret balance, initiative movement, and possible trend development.
Commodity Selection Index
Ranks markets for trend tradingWilder's measure for comparing commodity markets by combining directional movement, average true range, contract value, margin, and commission considerations. Its purpose is to identify markets that may offer a favorable opportunity for trend-following trades relative to the capital and risk involved.
Step-Forward Testing
Tests parameters on future segmentsA procedure that selects system parameters using one block of historical data, then applies those parameters to the next, unseen block. The process moves forward through time and combines the results from these successive out-of-sample periods. It aims to approximate how a strategy would have performed as new data became available.
Optimal f
Best fixed fraction to riskA position-sizing concept that seeks the fraction of capital to commit on each trade to maximize the growth of the account, based on the historical distribution of wins and losses. Because it depends heavily on observed outcomes and can imply aggressive exposure, its estimate is sensitive to data quality and the possibility of future losses larger than those recorded.
Elliott's Wave Principle
Interprets recurring price-wave patternsA market analysis framework that interprets price movement as sequences of waves in a primary direction, followed by corrective waves. It subdivides larger patterns into smaller ones and uses Fibonacci relationships to estimate potential proportions and timing. Identifying the current wave can be subjective, especially before a pattern is complete.
FAQ
What is "Trading Systems and Methods" by Perry J. Kaufman about?
- Comprehensive trading systems guide: The book offers a thorough exploration of trading systems, methods, and the tools needed to develop or select a robust trading program, with a focus on futures but applicable to other markets.
- Emphasis on systematic, quantitative trading: Kaufman prioritizes quantitative and systematic approaches, including trend-following, countertrend, breakout, and spread systems, rather than discretionary trading or market psychology.
- Bridging theory and practice: It covers both technical and fundamental analysis, research skills, risk management, and practical system development, aiming to close the gap between trading expectations and real-world results.
- Testing and validation focus: The book stresses rigorous system testing, validation, and adaptation to ensure trading strategies are robust and reliable.
Why should I read "Trading Systems and Methods" by Perry J. Kaufman?
- Authoritative and detailed resource: Kaufman draws on decades of experience, providing clear explanations, formulas, and programming examples suitable for both beginners and professionals.
- In-depth technical coverage: The book covers a wide range of technical methods, from basic moving averages to advanced topics like neural networks, genetic algorithms, and adaptive techniques.
- Practical system development: Readers are guided through system design, testing, risk management, and performance monitoring, all essential for real-world trading success.
- Bridges theory and application: Kaufman emphasizes the importance of adapting systems to real market conditions and using computer tools for optimization and robustness.
What are the key takeaways from "Trading Systems and Methods" by Perry J. Kaufman?
- Systematic approach is essential: Successful trading relies on well-tested, systematic methods rather than intuition or market lore.
- Testing and validation are critical: Robust backtesting, out-of-sample validation, and step-forward testing are necessary to avoid overfitting and ensure real-world performance.
- Risk management is non-negotiable: Effective leverage, diversification, stop-losses, and position sizing are vital to protect capital and maximize returns.
- Adaptation to market conditions: Adaptive techniques and regular system review help maintain effectiveness as market dynamics change.
What are the most notable quotes from "Trading Systems and Methods" by Perry J. Kaufman and what do they mean?
- "The market reflects all the jobber knows about the condition of the textile trade... it is better informed on crops than the farmer or even the Department of Agriculture." — Highlights that market prices incorporate all available information, often more than any individual participant.
- "Your answer is only as good as its weakest point." — Stresses the importance of rigor and thoroughness in system development and testing.
- "Speculation is a difficult business, not one to be taken casually." — Warns that trading requires discipline, skill, and realistic expectations, not casual or reckless behavior.
- "Conserve Your capital and wait for the right time." — Emphasizes patience and capital preservation as keys to long-term trading success.
How does Perry J. Kaufman define and compare technical and fundamental analysis in "Trading Systems and Methods"?
- Fundamental analysis: Focuses on supply-demand, economic reports, political events, and long-term price forecasts to understand market direction.
- Technical analysis: Studies price, volume, and open interest patterns, often using computerized methods, to forecast short- and long-term price movements.
- Integration benefits: Kaufman notes that many professionals combine both approaches, using technical analysis for timely signals and fundamental analysis for broader context.
- Practical application: The book provides tools and methods for both types of analysis, allowing traders to choose or blend approaches as needed.
What are the main types of moving averages and their uses according to "Trading Systems and Methods" by Perry J. Kaufman?
- Simple Moving Average (SMA): Averages prices over a fixed period, smoothing data but introducing lag; useful for identifying general trends.
- Exponential Moving Average (EMA): Weights recent prices more heavily, reducing lag and adapting faster to price changes; better for volatile markets.
- Weighted and double smoothing: Variations that adjust weighting schemes or smooth the moving average itself to improve trend detection and reduce noise.
- Adaptive moving averages: Methods like KAMA adjust their sensitivity based on market volatility, providing more responsive trend signals.
How does Perry J. Kaufman explain momentum and oscillator indicators in "Trading Systems and Methods"?
- Momentum as speed: Measures the rate of price change over a fixed interval, helping identify the strength and direction of trends.
- Oscillators normalized: Indicators like RSI and stochastic oscillate between fixed bounds (e.g., 0-100), signaling overbought or oversold conditions.
- Divergence and smoothing: Momentum divergence can signal trend reversals, while smoothing techniques reduce noise and improve signal quality.
- Adaptive oscillators: Adjust calculation periods or centering values to avoid prolonged false signals in changing market conditions.
What is cycle analysis and how is it applied in "Trading Systems and Methods" by Perry J. Kaufman?
- Trigonometric and spectral analysis: Uses sine and cosine functions, Fourier analysis, and Maximum Entropy Spectral Analysis (MESA) to identify and model market cycles.
- Seasonality and cycles: Recognizes recurring patterns linked to natural or economic cycles, such as planting/harvesting or planetary cycles.
- Practical application: Kaufman advises removing trend and seasonality to isolate cycles, which can then be used for timing entries and exits.
- Cycle channel index and phase angle: Tools like the cycle channel index and phase angle help track cycle progression and identify turning points.
How does Perry J. Kaufman describe and use charting techniques like point-and-figure and swing charts in "Trading Systems and Methods"?
- Point-and-figure charting: Plots price movements without regard to time, using Xs and Os to filter noise and highlight trends and reversals.
- Swing charts: Define trends by price reversals of a minimum size, filtering out minor fluctuations and clarifying trend changes.
- Trading signals and formations: The book details classic buy/sell signals, trendlines, and price objectives derived from chart patterns.
- Integration and risk management: Kaufman explains how to use these charts for stop-loss placement, position scaling, and combining with other methods for robust trading.
What are adaptive techniques in trading systems, as explained by Perry J. Kaufman in "Trading Systems and Methods"?
- Dynamic parameter adjustment: Adaptive methods modify parameters like moving average periods or stop-loss levels based on current volatility or market noise.
- Examples of adaptive systems: Includes Kaufman's Adaptive Moving Average (KAMA), Chande's VIDYA, and the Parabolic Time/Price System.
- Improved robustness: Adaptive techniques help reduce whipsaws, improve timing, and maintain effectiveness across different market regimes.
- Careful design required: While powerful, adaptive methods require thorough testing to avoid overfitting and ensure reliability.
How does Perry J. Kaufman recommend developing, testing, and validating trading systems in "Trading Systems and Methods"?
- Stepwise development: Start with a clear hypothesis, test simple components independently, then combine and refine for a complete system.
- In-sample and out-of-sample testing: Optimize parameters on historical data, then validate on unseen data to check for consistency and robustness.
- Step-forward (walk-forward) testing: Sequentially optimize and test on rolling periods to simulate real-time trading and avoid overfitting.
- Avoid feedback loops: Never reuse out-of-sample data for optimization, as this undermines the validity of future performance estimates.
What risk management and money management strategies does Perry J. Kaufman recommend in "Trading Systems and Methods"?
- Leverage management: Use margin and leverage carefully, adjusting position sizes according to risk tolerance and market volatility.
- Diversification: Spread risk across different markets and systems with low correlation to reduce overall portfolio volatility.
- Stop-loss and equity monitoring: Implement individual trade stops and monitor overall equity to prevent catastrophic losses and survive losing streaks.
- Conservation of capital: Risk only a small portion of total capital on any single trade to ensure long-term survival.
How does Perry J. Kaufman address advanced topics like neural networks, genetic algorithms, and pattern recognition in "Trading Systems and Methods"?
- Neural networks: Mimic brain-like structures to learn patterns from historical data, requiring careful input selection, training, and validation to avoid overfitting.
- Genetic algorithms: Use evolutionary principles to optimize trading strategies by mutating and combining parameter sets for improved performance.
- Pattern recognition: Forms the basis of many trading systems, from traditional charting to computerized analysis of price sequences and intraday patterns.
- Cautions and best practices: Kaufman warns that these advanced methods can find spurious relationships if not properly controlled, emphasizing the need for multiple runs and out-of-sample testing.
Download PDF
Download EPUB
.epub digital book format is ideal for reading ebooks on phones, tablets, and e-readers.