Technical Analysis of the Financial Markets — John J. Murphy
Chapter 2 - Dow Theory: pages 48–59.
1. Macro Overview & Strategic Value
Chapter 2 traces the origin of technical analysis back to Charles Dow’s turn-of-the-century Wall Street Journal editorials and distills them into six tenets that Murphy treats as the direct ancestor of modern trend-following systems. The strategic payoff for a practitioner is that nearly every core building block used later in the book — trend definition, confirmation/divergence logic, volume-based confirmation, and the “ride the trend until reversal” default — traces back to a framework built over a century ago, meaning a trader who internalizes Dow Theory is effectively pre-loading the assumptions behind every later chapter’s tools.
The chapter’s structural function is to formalize what counts as a trend and what counts as a valid signal. Dow’s three-trend hierarchy (primary/secondary/minor) and his insistence on cross-confirmation between the Industrial and Rail (now Transportation) Averages give a practitioner an explicit, falsifiable signal-generation process rather than a vague “the market looks bullish” impression — this is the first appearance in the book of a genuinely testable, rules-based trading framework.
Equally important strategically is Dow’s tolerance for lag: the theory intentionally sacrifices the first and last 20-25% of a move in exchange for higher-confidence participation in the middle of established trends. This directly foreshadows the core expectancy trade-off in every trend-following system built later — accept late entries and late exits in exchange for capturing large, high-probability trend segments rather than trying to pick tops and bottoms.
2. Core Concepts & Mechanics
- Averages discount everything — Dow’s original formulation of the discounting premise from Chapter 1, applied specifically to index-level price action rather than individual securities; implies index-level signals already reflect aggregate economic conditions.
- Three-tiered trend structure — Primary (year+), secondary/intermediate (3 weeks–3 months, retracing ~1/3 to 2/3 of the prior primary leg), and minor (<3 weeks) trends form a nested hierarchy; a trader must identify which tier they’re trading to avoid mistaking a correction for a reversal.
- Three-phase major trend model — Accumulation (informed early buying), public participation (trend-followers pile in as news improves), and distribution (informed money exits as sentiment peaks); operationally this tells a trader that by the time news confirms a trend, the best entry window is likely already behind them.
- Cross-index confirmation requirement — No valid bull/bear signal exists unless both the Industrial and Transportation Averages break to new highs/lows; divergence between the two is read as “prior trend still intact,” giving traders an explicit veto rule against premature reversal calls.
- Volume-trend alignment — Volume should expand in the direction of the prevailing trend (rising on up-moves in an uptrend, rising on down-moves in a downtrend); used only as secondary confirmation, since Dow’s actual signals were generated from closing prices alone.
- Trend continuation default — A trend is assumed intact until a definitive reversal signal appears, operationalizing the “trend in motion stays in motion” principle from Chapter 1 into an explicit decision rule: default to holding, not to fading.
- Failure vs. nonfailure swings — A failure swing (lower high fails to exceed the prior peak before breaking the prior low) is a weaker reversal signal than a nonfailure swing (new high made, then a lower high forms before the prior low breaks); this distinction is a direct precursor to modern swing-structure and market-structure-break analysis.
- Closing-price-only validation — Dow required a close beyond a prior peak/trough for a signal to count; intraday penetrations were disregarded, meaning a strict Dow-Theory trader ignores intraday noise entirely.
- Documented lag/capture trade-off — Dow Theory historically missed 20-25% of a move before signaling, but captured roughly 67-68% of major moves in the Industrials, Transports, and S&P 500 from 1920-1975 — a quantified expectancy statement about the cost of confirmation-based entries.
3. Technical Terminology & Reference Table
| Term | Operational Definition |
|---|---|
| Primary trend | Dominant multi-year market direction; the “tide” in Dow’s tide/wave/ripple analogy |
| Secondary (intermediate) trend | Corrective move lasting 3 weeks–3 months, typically retracing 33-66% (commonly ~50%) of the prior primary leg |
| Minor trend | Short-term fluctuation lasting under 3 weeks; noise relative to the intermediate trend |
| Accumulation phase | Early-stage informed buying before sentiment or news turns bullish |
| Distribution phase | Late-stage informed selling while public sentiment and news are still bullish |
| Confirmation | Requirement that two related averages (or indicators) both signal the same trend direction before it’s considered valid |
| Divergence | Situation where related averages disagree, read as evidence the prior trend is still in force |
| Failure swing | Reversal pattern where price fails to exceed the prior peak before breaking the prior trough — weaker signal |
| Nonfailure swing | Reversal pattern where price exceeds the prior peak, then forms a lower high before breaking the prior trough — stronger signal |
| Lines (in the averages) | Horizontal, sideways consolidation patterns; modern equivalent of “rectangles” |
4. The Author’s Market Philosophy
Murphy presents Dow’s model as fundamentally trend-persistent and capture-oriented rather than prediction-oriented — the goal was never to call exact tops and bottoms, but to reliably identify and ride the large middle portion of major moves, accepting confirmation lag as the cost of higher signal reliability. Participant behavior is modeled as a phase-based information cascade: informed capital acts first (accumulation/distribution) while the broader public reacts only once trends are already visible in price and news, meaning edge in this model comes from earlier trend recognition rather than earlier information. Edge generation, in Murphy’s reading of Dow, is structural rather than predictive — it comes from disciplined confirmation rules (cross-index agreement, volume alignment, closing-price-only signals) that filter out false signals at the cost of some lag, not from forecasting turning points ahead of the crowd.
5. Systemic & Portfolio Integration
Dow Theory is presented as the direct structural ancestor of every trend-following and momentum system discussed later in the book, since the primary/secondary/minor trend hierarchy and the confirmation/divergence framework reappear explicitly in later chapters on trend concepts and continuation patterns. The documented ~67-68% historical capture rate also functions as an early expectancy benchmark, illustrating that a system doesn’t need to catch entire moves to be profitable over time — a foundational concept for systematic risk management and position-sizing frameworks built later in the book.
6. Active Recall Evaluation
- Why does Dow require confirmation from two separate averages rather than relying on a single index’s signal alone — what specific failure mode does this rule guard against?
- Explain why a failure swing is considered a structurally weaker reversal signal than a nonfailure swing, in terms of what information each pattern does or doesn’t confirm.
- Dow Theory sacrifices the first and last 20-25% of a move in exchange for what benefit — and how does this trade-off generalize to modern systematic trend-following?
- Why does Murphy argue that the minor (near-term) trend, which Dow considered largely unimportant, becomes critical for futures traders specifically?
- How does the three-phase model (accumulation/participation/distribution) imply an information asymmetry between different classes of market participants, and what does that suggest about when the “public” typically enters and exits a trend?
Answer Key (spoiler)
- Requiring two averages to confirm each other guards against a false signal generated by movement isolated to one sector or index — if only one average breaks out while the other lags or diverges, Dow treats the prior trend as still intact rather than risking a premature reversal call based on a single data series.
- A failure swing shows only a lower low without a higher high preceding it, meaning momentum weakened before even retesting the prior peak — a nonfailure swing shows a genuine higher high first, so the subsequent lower high and broken low represent a clearer, better-confirmed shift in structure rather than a move that never regained strength in the first place.
- The trade-off buys higher-confidence entries and exits by waiting for confirmed trend establishment and confirmed reversal, at the cost of not catching the extreme early and late portions of a move; modern systematic trend-followers accept the same lag deliberately, since the large “fat middle” of a trend is argued to be more reliably profitable than attempting to time exact turning points.
- Because futures traders typically trade the intermediate trend rather than holding through full primary trends (unlike Dow’s original equity-index framework), minor swings become the practical tool for timing entries and exits within that intermediate move rather than being dismissed as noise.
- The model implies informed/early capital enters during accumulation while sentiment is still negative and exits during distribution while sentiment is still positive, meaning the broad public typically buys only after the trend and supporting news are already well established — placing most public participation squarely in the middle, most exposed phase of the cycle.