Knowledge base · Concept

Dow theory

Educational reference from the platform knowledge base — written agent-readable first, rendered here for humans. Mechanics, not advice: nothing here is a recommendation to buy or sell any security.

Dow theory

Definition

Dow theory is technical analysis’s founding framework — codified from Charles Dow’s and William Hamilton’s 1900-1929 Wall Street Journal editorials (Rhea’s 1932 text is the canon). Its core claims: markets move in three simultaneous trends (primary/tide, secondary/wave, minor/ripple); primary trends have accumulation, public- participation, and distribution phases; INDICES MUST CONFIRM each other (industrials and transports making new highs together — the original cross-confirmation rule); volume confirms trend; and a trend persists until definitive reversal evidence. Uniquely for a century-old charting framework, it has a peer-reviewed positive audit: Brown-Goetzmann-Kumar reconstructed Hamilton’s actual calls and found risk-adjusted value.

How it works / structure

  • The trend taxonomy (the durable contribution): the three-timescale decomposition — a primary trend carrying secondary corrections carrying daily noise — is the ancestor of every multi-timeframe method in this KB (lens-technical horizon discipline; indicator-sma ribbons operationalize it).
  • The confirmation principle (the testable core): divergence between related indices (one making highs the other refuses) as trend-health evidence — the logic generalized into modern breadth analysis (indicator-breadth-advance-decline); the transports-confirm-industrials specific has weakened as the economy de-industrialized (labeled — the PRINCIPLE outlived its original pair).
  • The audit (Brown-Goetzmann-Kumar): Hamilton’s 255 editorials (1902-1929), scored blind and replayed — positive risk-adjusted performance, primarily via being defensively out during declines; the finding is careful and bounded: ONE practitioner’s application, in one era, with value concentrated in risk timing rather than return chasing.
  • Engine-relevant residue: phase taxonomy (accumulation/distribution as regime states), cross- asset confirmation gates, and trend-until-proven-otherwise as a null (the trend-following school’s ancestor — style-trend-following-school).

When it applies

Multi-timeframe analysis design (the decomposition is the template); breadth/confirmation gate construction (the principle in modern instrumentation); trend-regime classification (phase language maps to the platform’s regime states); historical literacy (most chart vocabulary descends from here — reading the field requires its grammar).

Risk profile & failure modes

  • Subjectivity at the edges: “definitive reversal evidence” invites hindsight reading — the documented general critique of narrative charting; modern use requires parameterized reversal definitions.
  • The original pair’s decay: transports no longer proxy the economy’s shipment layer as they did in 1910 — mechanical industrials/transports confirmation is an antique (labeled); the principle needs modern pairs.
  • Era-boundedness of the audit: one practitioner, pre-1930 — the positive finding does not license the framework generically; it licenses testing it.
  • Lore accretion: a century of secondary literature attached claims Rhea never made — the KB cites the codified text and the audit, not the folklore.

Evidence & limits

Rhea (1932) is the primary text; Brown-Goetzmann-Kumar (1998) is the peer-reviewed audit — a rare positive, bounded result for a charting framework. Modern mechanical implementations of “Dow theory signals” vary so widely that cohort claims about “the” theory are unmeasurable (labeled); the platform tests specified versions only.

Falsifiable-thesis examples

Illustrations only, not signals:

  • “A parameterized confirmation gate (index high requires equal-weight counterpart high within 20 sessions) improves trend-entry quality vs ungated entries in replay (confirmation-principle test)” — falsified by the paired replay.
  • “Divergence states (cap-weighted highs unconfirmed by breadth) precede below-median 3-month forward returns (modern-pair restatement)” — falsified by the conditional distribution.

Cross-references

  • The modern instrumentation: indicator-breadth-advance-decline, indicator-sma
  • The descendant school: style-trend-following-school, strategy-momentum
  • The framework family: ta-elliott-wave, ta-wyckoff-method
  • The lens: lens-technical

Sources

  • Rhea, R. (1932), The Dow Theory — Barron's — the codified statement of Hamilton's editorials
  • Brown, S., Goetzmann, W. and Kumar, A. (1998), The Dow Theory: William Peter Hamilton's Track Record Reconsidered — Journal of Finance 53(4), 1311-1333

The agent cites this page.

Inside the platform, this entry is live context: the AI reasons from it, quotes it, and grades against it. Make your case.

Inquire about founding membership