Campaign Trading: Tactics and Strategies to Exploit the Markets - Extended Summary

Author: John Sweeney | Published: 1996 | Categories: Systematic Trading, Risk Management, Technical Analysis, Position Management

About This Summary

This is a PhD-level extended summary covering all key concepts from "Campaign Trading," John Sweeney's groundbreaking work on treating trading as a coordinated multi-tactic campaign. The book's central innovation -- Maximum Adverse Excursion (MAE) -- provides an empirical method for setting protective stops based on actual trade data rather than arbitrary rules or gut feeling. Using 11 years of New York Light Crude Oil data (1983-1994), Sweeney builds a complete trading system that coordinates trend following, add-on trades, reversals, range trading, and favorable excursion analysis into a unified campaign framework. For systematic traders, the MAE concept alone justifies the book's place in any trading library, but the broader campaign architecture -- switching between market modes and layering complementary tactics -- offers a blueprint for building robust, multi-regime trading systems.

Executive Overview

"Campaign Trading" occupies a unique position in the trading literature. While most trading books focus on a single methodology -- trend following, mean reversion, breakout trading -- Sweeney's contribution is architectural. He demonstrates how to combine multiple trading tactics into a coordinated campaign that operates across different market regimes (trending and ranging), much like a military campaign coordinates infantry, artillery, and air support toward a unified objective.

The book's most enduring contribution is the concept of Maximum Adverse Excursion (MAE): measuring the worst intraday price movement against your position during a trade, then using the statistical distribution of MAE across many trades to set empirically grounded protective stops. Before Sweeney, stop placement was largely an art -- traders used support/resistance levels, percentage-based rules, or dollar-amount limits. Sweeney transformed it into a science by showing that winning trades and losing trades have fundamentally different MAE distributions. Winners cluster at low MAE values (they rarely move far against you), while losers spread across higher MAE bins (they move significantly against you before being closed). This asymmetry creates a natural separation point where a protective stop captures almost all the winners while filtering out the majority of losers.

The campaign framework itself is built from seven coordinated tactics, all tested on 11 years of New York Light Crude Oil futures data (1983-1994). The tactics layer together: trend trading forms the foundation, add-on day trades and position trades exploit trending periods more aggressively, reversals recover from stopped-out trades, range trading captures non-trending periods, range reversals catch breakouts, and Minimum Favorable Excursion (MinFE) analysis provides a final risk-reduction filter. Together, the seven tactics produced 11,346 points of profit over the test period.

What makes this book particularly valuable is its intellectual honesty. Sweeney does not hide the uncomfortable realities: the trend system's win rate is approximately 1:2 (more losers than winners), the campaign requires constant mode-switching between trending and ranging states, and some tactics produce modest returns relative to the complexity they add. He presents everything with raw data, distribution tables, and transparent methodology -- allowing the reader to evaluate each tactic independently and decide which components fit their own trading approach.

For modern traders working with Auction Market Theory and order flow tools, Sweeney's campaign concept maps directly to the idea of identifying market structure (trending vs. bracketing/balancing) and deploying appropriate tactics for each regime. The MAE framework provides a rigorous, data-driven alternative to the common practice of setting stops at "obvious" levels that the market regularly hunts.

Core Thesis

Sweeney's central argument is threefold:

1. Trading should be treated as a coordinated campaign, not a series of isolated trades. Each trade exists within a broader context -- the current market mode, existing positions, and the campaign's overall risk profile. Tactics must be selected based on what the market is doing right now, not on a trader's preference for a particular approach.

2. Stop placement should be determined empirically through MAE analysis, not through arbitrary rules. By measuring how far winning trades move against you versus how far losing trades move against you, you can identify a natural stop level that preserves winners while cutting losers. This transforms stop placement from subjective judgment to statistical inference.

3. Markets alternate between trending and ranging modes, and a complete trading system must address both. A trend-following system alone leaves approximately 45% of market time untraded. A complete campaign exploits all market conditions through coordinated mode-switching.

Within the broader trading literature, "Campaign Trading" bridges the gap between pure trend-following systems (like those documented in the Turtle Trading literature) and more nuanced multi-regime approaches. It anticipates many ideas that would later become mainstream in quantitative trading: regime detection, adaptive position sizing, and empirical risk management.

Chapter-by-Chapter Analysis

Chapter 1: Campaigning

Sweeney opens by establishing the military campaign metaphor that structures the entire book. A campaign is not a single battle but a coordinated series of engagements designed to achieve a strategic objective. In markets, the "strategic objective" is consistent profitability across all market conditions -- not just during favorable trends.

The historical campaign metaphor involves two phases: accumulation (building positions at favorable prices) and distribution (liquidating positions at higher prices). Sweeney modernizes this by replacing the accumulation/distribution cycle with continuous market mode detection. The modern campaigner uses market indicators to specify whether the market is trending or ranging at any given time, then deploys the appropriate set of tactics.

Two key innovations distinguish Sweeney's approach from conventional technical analysis:

1. Measuring everything from the point of entry. Rather than analyzing price patterns on a chart, Sweeney anchors all analysis to the trader's actual entry price. This shifts the frame of reference from abstract chart patterns to the trader's real-world experience of each position.

2. Measuring adverse and favorable excursion to set stops empirically. Instead of placing stops at chart levels, Sweeney measures how far price moves against (adverse) and in favor of (favorable) each trade, then uses these distributions to determine optimal stop placement.

The chapter establishes the two market states that drive mode-switching: trending (persistent directional movement) and ranging (oscillation within boundaries). The campaigner must trade both. A trend entry signal begins the campaign. Add-on trades exploit the trend while it persists. When the trend ends, the system switches to range trading. When the range breaks, the system reverses into a new trend.

Key Insight: The campaign metaphor is not merely decorative. It fundamentally changes how a trader thinks about individual trades -- not as standalone bets but as coordinated components of a larger operation, each serving a specific tactical purpose within the campaign's overall strategy.

Chapter 2: Trading the Trend

Chapter 2 establishes the campaign's foundational tactic: trend identification and trading. Sweeney defines trend as persistence -- the consistent direction of price change combined with the duration of that consistency. A market is trending when price changes maintain a consistent direction over a meaningful period.

Sweeney surveys multiple methods for defining trend:

Sweeney selects moving averages for their simplicity and robustness. His system uses two averages:

• Short average: 12-day moving average (roughly half a trading month)
• Long average: 12-week / 60-day moving average (roughly one trading quarter)

The trading rules are straightforward:

• Go long when both averages are moving up
• Go short when both averages are moving down
• Stay out when the averages disagree

Weekly charts determine the long average's direction; daily charts determine the short average's direction. Applied to Crude Oil from 1984-1994, this simple system produces an uphill equity curve -- but with significant drawdowns and whipsaws during non-trending periods.

The chapter is important because it establishes the baseline from which all other campaign tactics build. The two-moving-average trend system is the campaign's foundation -- not because it is optimal in isolation, but because it provides the mode detection framework that triggers all other tactics.

Critical Note (Chapter 2): Sweeney acknowledges that the trend system alone has a poor win rate and significant drawdowns. This honesty is refreshing -- many trading authors present their trend systems as standalone solutions. Sweeney treats the trend system as one component within a larger architecture.

Chapter 3: Handling the Bad News

This is the book's most important chapter. Here Sweeney introduces Maximum Adverse Excursion (MAE) -- the innovation that would influence an entire generation of systematic traders and risk managers.

MAE Defined: The worst intraday price movement against your position during a trade.

• For long positions: MAE = Entry Price - Low of each day during the trade
• For short positions: MAE = High of each day during the trade - Entry Price

Sweeney provides a rigorous 7-step MAE measurement process:

Step 1: Define Entry and Exit Rules Establish the complete trading system rules before any MAE analysis. The rules must be unambiguous and mechanically executable.

Step 2: Tabulate Daily MAE for Each Trade For every trade generated by the system, record the MAE for each day the trade is open. This creates a time series of adverse excursion for each individual trade.

Step 3: Separate Trades into Winners and Losers Classify each completed trade as a winner or loser based on net profit/loss at exit.

Step 4: Tabulate MAE for Winners and Losers Separately Record the maximum MAE reached during each trade's lifetime, keeping winners and losers in separate groups.

Step 5: Sort into Bins Based on Capital Using the 2% rule (risk no more than 2% of capital per trade), calculate the maximum dollar loss and convert to price units. Divide by 2 to create bin sizes. Example: $30,000 capital, 2% max loss = $600 = 60 ticks in Crude = 0.6 points. Bin size = 0.3 points.

Step 6: Sort Trades' MAEs into Bins Place each trade's maximum MAE into the appropriate bin.

Step 7: Graph the Summary Table Create a histogram showing the distribution of winners and losers across MAE bins.

The Critical Finding: Winners cluster at LOW MAE values (left side of the distribution), while losers spread across HIGHER MAE bins. This asymmetry is the "edge" -- the empirical basis for stop placement. A stop placed at the boundary between the winner cluster and the loser spread captures nearly all winners while eliminating the majority of losers.

With $30,000 capital applied to Crude Oil:

• Using a 0.30 stop: net profit 4.08 points per trade (vs. 3.98 with no stop)
• Using a 0.08 stop (too tight): only 1.36 profit per trade

The long-term results over 3,069 trading days: 104 winners, 148 losers. Winners concentrated in the first two MAE bins. Most losers had MAE exceeding 0.3 points.

Key Quote (Chapter 3): The MAE distribution reveals what no amount of chart reading can: exactly how far winning trades move against you versus how far losing trades move against you. The stop is not a guess -- it is a statistical boundary.

Chapter 4: Testing

Chapter 4 addresses a critical methodological question: does the choice of contract data affect MAE analysis? Sweeney tests across three contract types:

Key findings across all three contracts (Table 4.1):

• All three show winning MAE peaks below 0.15
• Losing MAE peaks between 0.16 and 0.30
• Monthly rollover contracts are most profitable (approximately 30% more than perpetual)
• The MAE distribution shape is consistent across contract types

Optimal stop levels: 0.31 or 0.46. The 0.46 level matches Crude Oil's characteristic noise band -- the typical random fluctuation that does not indicate a directional change.

Win/loss ratio: Approximately 1:2 (one winner for every two losers). Sweeney calls this "depressing but realistic for trend systems." The system's profitability depends entirely on the average winner being significantly larger than the average loser.

Campaign profitability with trend trading alone: 3,600 points over 10 years.

Critical Analysis (Chapter 4): Sweeney's transparency about the poor win ratio is commendable. Many trading system vendors obscure this reality. However, testing on a single instrument (Crude Oil) over a single period (1983-1994) limits the generalizability of the specific MAE levels. The methodology is universally applicable, but the optimal stop values are instrument-specific and period-specific.

Chapter 5: Piling On -- Exploiting the Trend

Once a trend is identified, Sweeney introduces tactics to extract additional profit from the trending period. He describes three methods for add-on trades:

1. Intraday Add-On Day Trades

• Trigger: Price touches the 12-day (short) moving average during an established trend
• Entry: Buy stop at the average's estimated value
• Exit: At the close of the same day
• Key characteristic: These are NOT overnight positions. Pure day trades.
• MAE profile: Winners cluster even more tightly at low MAE than trend trades. Very high winning percentage -- Sweeney calls this the "most comforting feature" of the tactic.
• Stop: 0.16 or 0.31

2. Position-Building Add-On Trades

• Trigger: Price touches the 60-day (long) moving average during an established trend
• Entry: At the touch of the long average
• Exit: When the underlying trend trade closes
• Key characteristic: These are position trades held for the duration of the trend. Higher win rate than day trades but fewer opportunities (some years produce zero trades).
• MAE profile: Winners' adverse movement much less than losers'. Stop anywhere from 0.16 to 0.61 is effective.
• Stop: 0.31

3. Countertrading

• Trading against the trend at channel extremes. Sweeney mentions this approach but does not test it extensively.

Campaign Profitability Table (Table 5.1):

The add-on day trades contribute significantly (2,500 points) because they occur frequently during trends and have a high win rate. Position add-ons contribute less (640 points) because opportunities are rare -- but when they work, they capture large moves.

Chapter 6: Reversing Bad Trades

This chapter contains one of the book's most counterintuitive and powerful ideas: when an MAE stop is hit, the probability that the trade is a loser is approximately 90%. Why not immediately reverse the position?

The logic is compelling: if hitting the MAE stop means you are almost certainly wrong about direction, then the opposite direction is almost certainly correct. The 10% risk -- that the original trade would have recovered -- is the cost of insurance against the 90% probability of continued loss.

Reversal Trading Rules:

• If long and MAE stop is hit at 0.31, reverse with 2 contracts short
• If short and MAE stop is hit at 0.31, reverse with 2 contracts long
• No protective stop on the reversal (the original trend's time horizon serves as the exit)
• Exit when the underlying trend trade's time horizon ends

Results over 11 years: 68 winners, 31 losers = 70% win rate. This is exceptionally high for any systematic approach.

Five Categories of Reversals:

Time-Based Pattern: Good things happen quickly for reversals. Profitable reversals typically show gains within 2 days. Beyond 10 days, a reversal is almost certainly a loser.

Contribution: Reversals add 1,827 points to the campaign.

Critical Analysis (Chapter 6): The "no stop on reversals" rule is psychologically challenging and potentially dangerous. Sweeney's data supports it for Crude Oil over this period, but the absence of a protective stop means that a catastrophic move against the reversal position has unlimited risk. Modern traders might consider a wider stop (perhaps at the 0.46 or 0.61 level) rather than no stop at all.

Chapter 7: Switching Modes -- Trading Ranges

Markets trend approximately 55% of the time using Sweeney's definitions (for Crude Oil with these specific rules). The remaining 45% is spent in ranges. A complete campaign must trade both regimes.

Range Definition (4 conditions -- all must be met):

1. A trend of at least 6 days must have existed
2. The trend is broken by prices crossing the short (12-day) average
3. Prices reach a reaction level within the cyclic period (12 days)
4. Prices recross the average -- range is officially declared

Once a range is identified, the trading approach reverses: instead of trading with direction, the trader buys at the bottom of the range and sells at the top, exiting at the opposite boundary or when a new trend re-establishes.

Range Characteristics for Crude Oil:

• Typical range width: approximately 100 trading points (1.0)
• Typical range duration: 20-30 days
• MAE distributions for range trades are wider than for trend trades -- stop placement is less clear-cut, with practical choices between 0.3 and 0.6

Breakout Warning: In 18 of 26 cases, trading into the range at point 5 (when price returns to the initial extreme) led to immediate loss as the range broke into a new trend. This means range trading at extremes is inherently risky near potential breakout points.

Contribution: Range trading adds 550 points with a 0.31 stop.

Key Insight (Chapter 7): The relatively modest contribution of range trading (550 points vs. 3,600 for trend trading) reflects a fundamental truth about systematic trading: trend-following captures the large moves, while range trading grinds out smaller, less reliable profits. The value of range trading is not in its absolute return but in keeping capital productive during non-trending periods.

Chapter 8: Reversing Out of Ranges

Applying the same reversal logic from Chapter 6 to range trades: if a range trade hits its MAE stop, reverse the position. The reasoning is identical -- a stopped-out range trade signals that the range boundary has been breached, likely indicating a breakout into a new trend.

Results: 35 trades total, 19 winners. Approximately 50-50 before stops are applied.

Critical Difference from Trend Reversals: The optimal stop for range reversals is 0.51 -- significantly higher than the 0.31 used for trend reversals. This is because range MAE distributions are wider than trend MAE distributions. Range breakouts produce more initial adverse movement before the breakout trend establishes itself.

This is the first time in the book that a different stop level is warranted for a different tactic. Sweeney emphasizes this point: stop levels should be determined empirically for each tactic, not imposed uniformly across all tactics.

Favorable excursion observation: Range reversal trades tend to produce favorable excursion equal to or exceeding the range size. When a range breaks, the subsequent move is typically at least as large as the range itself.

Contribution: With a 0.51 stop: 2,229 points profit -- a significant addition to the campaign.

Updated Campaign Profitability Table (Table 8.1):

Chapter 9: Minimum Favorable Excursion

If MAE measures how bad things get, Minimum Favorable Excursion (MinFE) measures how good things get -- specifically, the minimum amount of favorable movement a trade achieves. While Maximum Favorable Excursion (MaxFE) measures the best a trade ever does, MinFE measures the least favorable point after the trade begins moving in the trader's favor.

• For long positions: MinFE = nearest low above entry (measuring the minimum favorable movement)
• For short positions: MinFE = nearest high below entry (measuring the minimum favorable movement)

Key Finding: MinFE provides a sharper distinction between winners and losers than MaxFE. All trades with MinFE below 0.3 were losers. Trades with MinFE above 0.3 were almost always winners.

Protective Stop Rule: When MinFE exceeds 0.3 (meaning the trade has moved at least 0.3 in your favor), move the stop to breakeven. This locks in the initial favorable movement and converts a live-risk trade into a risk-free position.

Gray Area: Approximately 10% of trades show MinFE below 0.3 but MaxFE above 0. These are trades that moved favorably but never established clear momentum. Most of these are losers.

Day-by-Day Profit Patterns:

• Losers: Daily gains oscillate between +0.4 and -0.4, meandering without direction for up to 10 days before the stop is hit or the trade is exited
• Winners: Show a steady advance along a consistent channel, lasting 20+ days with clear directional momentum

These patterns reveal that winning trades distinguish themselves from losers quickly and consistently. A trade that has not moved meaningfully in your favor within the first few days is unlikely to become a winner.

Campaign Contribution: MinFE analysis adds a 7th tactic -- cutting off losing reversal trades using MinFE filters. Net additional profit: 0 points. However, the value is in risk reduction, not profit addition. By identifying losing trades earlier, MinFE analysis reduces the campaign's maximum drawdown and improves risk-adjusted returns.

Final Campaign Profitability Table (Table 9.1):

Chapter 10: Shifting the Odds -- Using Options

Sweeney explores options as an alternative to hard stops, addressing several practical problems with stop-loss orders:

Problems with Hard Stops:

• Slippage: Market orders executed at the stop level may fill at worse prices, especially in fast markets
• Stop-running: Floor traders and market makers may push price through known stop levels to trigger orders before reversing
• Inflexibility: A stop is binary -- you are either in or out

Options as Alternative Risk Management:

• A purchased call = long underlying + purchased put (synthetic equivalence)
• The option premium replaces the stop loss as the maximum risk
• For 10-20 day trade horizons, options are cost-competitive with stop losses

Cost Comparison for At-the-Money Options:

• With an MAE stop of 0.38: the option costs 0.38 maximum, comparable to the stop loss amount
• Time decay is minimal for short-duration trades (5-20 days), making options particularly attractive for reversal trades where the trade horizon is 5-10 days

Trade-offs:

• Winners: The outright position always outperforms by the cost of the option premium. Options reduce net profit on winners.
• Losers: The option limits loss to premium paid. Additionally, the option can be sold before expiration to recover some premium.

Critical Analysis (Chapter 10): This chapter was more relevant in 1996 when options on commodity futures were less liquid and bid-ask spreads were wider. In modern markets, options liquidity has improved dramatically, making this approach more practical. However, the fundamental trade-off remains: options provide smoother risk management at the cost of reduced profit on winners.

Chapter 11: The Toolbox

The final chapter assembles the complete campaign framework, summarizing how all seven tactics work together in real time. Sweeney emphasizes that while each tactic was tested independently on historical data, they operate simultaneously in a live campaign.

The chapter serves as a practical guide to campaign execution:

1. Identify market mode (trending or ranging) using the two-moving-average system
2. If trending: Execute trend trades, deploy add-on day trades and position trades, reverse stopped-out trades
3. If ranging: Execute range trades at boundaries, reverse stopped-out range trades
4. Always: Apply MAE stops empirically, use MinFE to move stops to breakeven, monitor mode transitions

Sweeney stresses that all tactics were tested independently but are designed to work as an integrated system. The campaign is more than the sum of its parts -- the mode-switching framework ensures that the trader is always positioned for the current market environment, not waiting for a single type of opportunity.

Key Frameworks and Models

Framework 1: The 7-Step MAE Analysis Process

Application: This process can be applied to any trading system on any instrument. The key output is the MAE histogram showing where winners and losers separate -- the optimal stop level sits at this separation boundary.

Framework 2: Campaign Mode-Switching Decision Tree

Market State Assessment
|
|-- Both MAs rising --> TREND MODE (Long)
|   |-- Enter long position (Tactic 1)
|   |-- Price touches 12-day MA --> Add-on day trade (Tactic 2)
|   |-- Price touches 60-day MA --> Add-on position trade (Tactic 3)
|   |-- MAE stop hit --> Reverse position (Tactic 4)
|   |-- MinFE > 0.3 --> Move stop to breakeven (Tactic 7)
|
|-- Both MAs falling --> TREND MODE (Short)
|   |-- [Same tactics, reversed direction]
|
|-- MAs disagree --> Evaluate for RANGE MODE
|   |-- 4 range conditions met --> RANGE MODE
|   |   |-- Trade at range boundaries (Tactic 5)
|   |   |-- MAE stop hit --> Reverse out of range (Tactic 6)
|   |-- 4 range conditions NOT met --> FLAT (no position)

Framework 3: Contract Type Comparison for MAE Testing

Framework 4: Reversal Trade Classification System

Framework 5: Optimal Stop Levels by Tactic

Practical Checklists

Pre-Campaign Setup Checklist

• Define the two moving averages (short and long) for your instrument
• Collect at least 5 years of historical data for your instrument
• Run the 7-step MAE analysis on your trend trading system
• Determine instrument-specific MAE bin sizes based on your capital (2% rule / 2)
• Identify the optimal stop level from the MAE distribution
• Test MAE distributions on multiple contract types (if trading futures)
• Determine range definition parameters for your instrument's typical cycle length
• Calculate position sizing that allows for campaign-level risk (multiple concurrent positions)
• Establish rules for mode-switching between trend and range trading
• Back-test each tactic independently before combining them

Daily Campaign Execution Checklist

• Assess current market mode: trending (both MAs aligned) or ranging (MAs disagree + 4 range conditions)?
• If trending: Is there an existing trend position? If not, evaluate entry
• If trending: Has price touched the short MA? If yes, evaluate add-on day trade
• If trending: Has price touched the long MA? If yes, evaluate add-on position trade
• Check all open positions against MAE stops -- any stops hit?
• If MAE stop hit: execute reversal per rules
• If ranging: Is price at a range boundary? If yes, evaluate range trade
• Check MinFE on all open positions -- any exceeding 0.3 threshold?
• If MinFE > 0.3: move protective stop to breakeven
• Review reversal trades: any open longer than 10 days? Consider exit

MAE Analysis Quality Checklist

• Minimum 100 trades in the sample for statistical significance
• Bin size correctly calculated from account capital using 2% rule
• Winners and losers graphed on the same histogram for comparison
• Clear separation point visible between winner cluster and loser spread
• Tested across at least two different time periods to confirm stability
• Optimal stop tested at separation point AND adjacent bins
• Verified that stop improves or maintains net profitability (vs. no stop)

Critical Analysis

Strengths

1. Empirical Rigor. The book is built entirely on data, not opinion. Every claim is backed by 11 years of trade-by-trade analysis. Sweeney shows his work, presents the raw distributions, and allows the reader to draw independent conclusions. This level of transparency was rare in 1996 and remains uncommon today.

2. The MAE Concept. Maximum Adverse Excursion is a genuinely original contribution to trading methodology. It transforms stop placement from a subjective art to an empirical science. The concept has been adopted, adapted, and extended by systematic traders worldwide and remains relevant nearly 30 years after publication.

3. Honest Win Rates. Sweeney does not disguise the fact that the trend system loses more often than it wins (approximately 1:2). This honesty builds credibility and prepares the trader psychologically for the reality of systematic trading -- where the edge comes from asymmetric payoffs, not high win rates.

4. Multi-Regime Framework. The campaign architecture addresses the fundamental weakness of pure trend-following: what to do when markets are not trending. By explicitly designing tactics for ranging markets and including mode-switching logic, Sweeney creates a more complete system than most trend-following books offer.

5. The Reversal Innovation. The insight that an MAE stop hit implies a 90% probability of being wrong -- and therefore a 90% probability that the reverse direction is correct -- is elegant and actionable. The 70% win rate on reversal trades is compelling evidence.

Weaknesses

1. Single-Instrument Testing. All analysis uses New York Light Crude Oil from 1983-1994. While the MAE methodology is instrument-agnostic, the specific stop levels, range widths, and profitability numbers are specific to this instrument and time period. Crude Oil in this era was characterized by moderate volatility and clear trends (including the 1986 crash and Gulf War spike) -- conditions that may not represent all instruments or all time periods.

2. No Transaction Costs. The campaign profitability tables do not explicitly account for commissions, slippage, or financing costs. With seven tactics generating potentially hundreds of trades per year, transaction costs could significantly erode the reported 11,346-point total. This is particularly relevant for the add-on day trades (Tactic 2), which generate the highest trade frequency.

3. No Out-of-Sample Testing. All results are in-sample. Sweeney optimized stop levels on the same data used to report profitability. While the methodology is robust, the specific parameter choices (0.31 stop, 12-day and 60-day averages) could be partially curve-fitted. A walk-forward analysis or out-of-sample test would have strengthened the claims considerably.

4. No Risk Metrics Beyond Win Rate. The book does not report maximum drawdown, Sharpe ratio, or other risk-adjusted performance metrics. "11,346 points over 10 years" sounds impressive, but without knowing the peak-to-trough drawdown or the path-dependent risk, it is impossible to assess whether the campaign offers an attractive risk/reward profile.

5. Reversal Trades Without Stops. The recommendation to use no protective stop on reversal trades is the book's most dangerous suggestion. While the data supports this for the test period, a single catastrophic move against an unprotected reversal position could wipe out years of campaign profits. Modern risk management practice would demand at least a wide stop or options protection.

6. Complexity of Full Campaign. Running seven simultaneous tactics with mode-switching requires significant execution discipline and monitoring capability. In 1996, this was a manual process. Today, it could be automated, but the book provides no guidance on implementation or the practical challenges of real-time campaign management.

Modern Relevance

The MAE concept remains highly relevant. Modern traders can apply MAE analysis to any systematic trading approach -- including intraday strategies, options strategies, and algorithmic systems. Software tools now make the 7-step MAE process trivial to implement compared to the manual calculations Sweeney performed.

The campaign framework's mode-switching concept maps directly to modern regime-detection approaches. Quantitative funds routinely classify market regimes (trending, mean-reverting, high-volatility, low-volatility) and deploy different strategies for each. Sweeney anticipated this approach by two decades.

The specific parameters (12-day and 60-day moving averages, 0.31 stop) should NOT be applied blindly to modern markets. Crude Oil's volatility structure, liquidity profile, and market microstructure have changed dramatically since 1994. However, the process for determining these parameters -- MAE analysis, mode classification, empirical stop optimization -- remains directly applicable.

For Auction Market Theory practitioners, the campaign framework maps to the AMT concept of identifying balanced (ranging) versus imbalanced (trending) markets and deploying appropriate tactics for each condition. The MAE-based stop corresponds to placing stops beyond the "noise" of the current market structure -- a concept aligned with value area-based risk management.

Key Quotes

"The Maximum Adverse Excursion distribution reveals the single most important fact about your trading system: how far winning trades move against you versus how far losing trades move against you. Everything else -- entry signals, exit rules, position sizing -- is secondary to this fundamental asymmetry." - Chapter 3

"Winners cluster at low MAE values. Losers spread across the entire range. This is not a feature of any particular market or time period -- it is a structural property of directional trading systems." - Chapter 3

"When the MAE stop is hit, there is a 90% probability the trade is a loser. Why not reverse?" - Chapter 6

"The campaigner does not predict the market. The campaigner responds to the market. When the market trends, the campaigner trends. When the market ranges, the campaigner ranges. When the campaigner is wrong, the campaigner reverses." - Chapter 1

"A win rate of one in three is depressing to contemplate but realistic for trend-following systems. The edge is not in being right often -- it is in being right big." - Chapter 4

"Good things happen quickly in reversal trades. If the reversal has not moved in your favor within two days, it is almost certainly a loser." - Chapter 6

"Minimum Favorable Excursion provides a sharper distinction between winners and losers than Maximum Favorable Excursion. All trades with MinFE below .3 were losers. This is the clearest signal in the entire campaign." - Chapter 9

Trading Takeaways

1. Apply MAE analysis to every trading system you use. Before setting any stop level, measure the MAE distribution of your winners and losers separately. The optimal stop sits at the boundary where winner and loser distributions separate. This single technique will improve your risk management more than any other single change.

2. Stop placement should be empirical, not arbitrary. Round-number stops, fixed-dollar stops, and "feels right" stops are all inferior to MAE-derived stops. The market does not care about your preferred stop level -- it cares about its own volatility structure. Let the data tell you where to place your stop.

3. Expect to lose more often than you win with trend-following systems. A 1:2 win/loss ratio is normal and sustainable if your average winner is significantly larger than your average loser. Psychologically prepare for long losing streaks. The edge is in asymmetric payoffs, not in being right frequently.

4. When your stop is hit, consider reversing. A stopped-out trade is strong evidence that you are on the wrong side of the market. Rather than simply exiting, evaluate whether a reversal entry is warranted. The MAE stop hit itself is an information signal -- use it.

5. Trade both trending and ranging markets. A trend-following-only approach leaves approximately 45% of market time untraded. Developing range-trading tactics and mode-switching rules creates a more complete system that keeps capital productive across all market conditions.

6. Different tactics require different stop levels. Do not impose a single stop level across all your strategies. Run MAE analysis independently for each tactic. Range reversals in Sweeney's data required a 0.51 stop versus 0.31 for trend trades -- using 0.31 across the board would have destroyed range reversal profitability.

7. Use Minimum Favorable Excursion to move stops to breakeven. Once a trade has moved a meaningful distance in your favor (MinFE exceeds your instrument's threshold), move your stop to breakeven. This eliminates risk on trades that have demonstrated directional momentum while allowing winners to continue developing.

8. Winning trades reveal themselves quickly. Both trend trades and reversal trades show their character within the first few days. Trades that meander without clear direction are overwhelmingly likely to become losers. Do not hope for a trade to "come back" -- the data says it will not.

9. Measure everything from the point of entry. Sweeney's insight about anchoring analysis to the entry price (rather than to chart patterns or external reference points) keeps the trader focused on the only thing that matters: how this specific position is performing relative to where it was initiated.

10. Build your trading system as a campaign, not a single tactic. The power of Sweeney's approach is in the architecture -- coordinating multiple tactics that exploit different market conditions. Each tactic has modest standalone performance, but the combined campaign produces returns greater than any single component.

11. Options can replace hard stops for short-duration trades. For trades with 5-20 day horizons, purchasing a protective option may cost no more than the expected stop-loss amount while providing smoother risk management, eliminating slippage risk, and preventing stop-running.

12. Test your system across multiple contract types and time periods. Perpetual contracts are convenient for testing but may not match actual trading conditions. Monthly rollover contracts are closest to real-world execution. If your results differ significantly across contract types, investigate why -- the discrepancy reveals something about your system's sensitivity to data construction.

Further Reading

• "Trade Your Way to Financial Freedom" by Van K. Tharp - Extends position sizing and expectancy concepts that complement Sweeney's MAE approach. Tharp's R-multiple framework provides a natural extension of MAE analysis.
• "Trend Following" by Michael W. Covel - Broader treatment of trend-following philosophy and history, providing context for Sweeney's trend-trading tactics.
• "The Complete TurtleTrader" by Michael W. Covel - Documents the Turtle Trading experiment, which shares Sweeney's emphasis on systematic rules, empirical risk management, and the psychological challenge of low win rates.
• "Technical Analysis of the Financial Markets" by John J. Murphy - Comprehensive reference for the moving average, trendline, and range-trading concepts Sweeney builds upon.
• "New Trading Systems and Methods" by Perry Kaufman - Quantitative treatment of many of the same concepts (moving averages, adaptive systems, regime detection) with modern statistical rigor.
• "Evidence-Based Technical Analysis" by David Aronson - Addresses the out-of-sample testing and statistical validity concerns that represent the main weakness of Sweeney's single-period analysis.
• "Maximum Adverse Excursion" by John Sweeney (1997) - Sweeney's follow-up book dedicated entirely to MAE analysis, providing deeper treatment of the concept introduced in Chapters 3-4 of "Campaign Trading."