Dow Theory

The man who never wrote a book

Almost everything in the last five lessons — trends, peaks and troughs, support/resistance, the idea that markets have personalities at different timeframes — traces to one man writing newspaper editorials around the year 1900.

Charles Dow founded Dow Jones & Company with Edward Jones in 1882. On July 3, 1884, he published the first stock market average of its kind: eleven stocks — nine railroads and two manufacturers — that he thought summarized the country's economic health. In 1897 he split the index into two — 12 Industrials and 20 Rails — which is the ancestor of the 30-stock Dow Jones Industrial Average that you still see quoted on the evening news.

Dow never wrote a book on his own theory. He died in 1902. In 1903, S.A. Nelson compiled Dow's editorials into The ABC of Stock Speculation and coined the term "Dow's Theory." William Peter Hamilton — Dow's associate and successor at the Wall Street Journal — formalized the ideas in 1922's The Stock Market Barometer. Robert Rhea extended them in 1932's The Dow Theory.

Richard Russell compared Dow's contribution to stock-market theory with Freud's contribution to psychiatry. That's not entirely ironic: Dow was working out what the market was actually doing by reading its own tape, and writing down the patterns he saw. Those patterns have held up for 120 years.

The six tenets

Murphy reproduces them faithfully. Here they are, each with the verbatim Dow/Hamilton wording and the mechanical consequence you can still use today.

1. The averages discount everything

The sum and tendency of the transactions of the Stock Exchange represent the sum of all Wall Street's knowledge of the past, immediate and remote, applied to the discounting of the future. — Hamilton, paraphrasing Dow

Everything knowable is already in the price. Interest rates, earnings, sentiment, fear, news that hasn't broken yet but insiders know about — all of it is being processed by people making trades, and the trades are what produce the tape. You don't need to compile macro indicators, bank clearings, or commodity price indexes separately; the market already did that.

This is the foundational premise of all modern technical analysis. It's also the assumption that lets you do your job without a Bloomberg terminal.

2. The market has three trends

Dow defined an uptrend as a situation in which each successive rally closes higher than the previous rally high, and each successive rally low also closes higher than the previous rally low. In other words, an uptrend has a pattern of rising peaks and troughs.

The same definition you learned in the Trends lesson — it came from here. But Dow's fuller framing is that trends coexist at three scales, which he compared to the sea:

Murphy, channeling Dow's own metaphor:

An observer can determine the direction of the tide by noting the highest point on the beach reached by successive waves. If each successive wave reaches further inland than the preceding one, the tide is flowing in.

Practical consequence: a 3-week pullback inside a multi-year uptrend is a wave, not a turning tide. Pros don't confuse those. Amateurs do.

3. Major trends have three phases

This is the tenet that makes Dow a pattern-reader and not just a trend-definer. Every major bull market, he claimed, unfolds in three distinct phases:

Click through the phases to see them on the chart. Each is characterized not just by price structure but by who is buying and what the news looks like:

1. Accumulation — after a bad decline, informed investors — "who recognize that the market has assimilated all the so-called bad news" — start buying from discouraged holders. Volume is quiet. The chart is boring. News is terrible and everybody's convinced the bottom hasn't come yet.

2. Public participation — price breaks out on expanding momentum. News improves. Trend-followers and technicians enter. This is the longest phase and the only one that classical trend systems are designed to capture.

3. Distribution — newspapers print "increasingly bullish stories." Economic news is "better than ever." Speculative volume climbs. Meanwhile, the same informed investors who accumulated at the bottom quietly sell into the euphoria. One final higher high, then price rolls over.

Students of Elliott Wave theory will recognize the structure: R.N. Elliott elaborated on Rhea's version of Dow Theory and got to the five-wave sequence from here. Wyckoff took the same framework in a different direction. Both trace back to Dow.

4. The averages must confirm each other

Dow's specific application, since he built the Industrials and the Rails (now the Transports):

Dow, in referring to the Industrial and Rail Averages, meant that no important bull or bear market signal could take place unless both averages gave the same signal, thus confirming each other.

The logic: the economy produces goods (Industrials) and moves those goods (Transports). A bull market without Transports confirming — i.e. factories producing but shipments stagnant — is probably not a real bull market. Either the goods aren't moving, or nobody's buying them.

Murphy, on the timing:

He did not believe that the signals had to occur simultaneously, but recognized that a shorter length of time between the two signals provided stronger confirmation.

Specific historical example Murphy cites: "At the start of 1997, the Dow Transports confirmed the earlier breakout in the Industrials. The following May, the Dow Industrials confirmed the earlier new high in the Transports." Six months of inter-average confirmation. That 1997–2000 bull market was the real deal.

Modern consequence: whenever you take a broad-market signal, check whether a correlated proxy agrees. SPY making new highs while QQQ and IWM don't? Suspicious. Tech highs while semiconductors lag? Question the breakout.

5. Volume must confirm the trend

Volume should expand or increase in the direction of the major trend. In a major uptrend, volume would then increase as prices move higher, and diminish as prices fall.

This is Dow's secondary rule — he based his actual buy/sell signals on closing prices alone, not volume. Volume was a confirmation filter, not a trigger. A breakout on thin volume is suspect; a breakout on expanding volume has participation behind it.

Bulkowski would nuance this a century later (his 150,000-sample test found volume is only a weak predictor of breakout success), but the directional framing — volume should move with the trend, not against it — still holds.

6. A trend is assumed to be in effect until it gives definite signals that it has reversed

An object in motion tends to continue in motion until some external force causes it to change direction. The odds usually favor that the existing trend will continue.

Newton, applied to the tape. This is the tenet that underpins every trend-following system ever written: don't exit a trend without a structural reason. A single down day, a single indicator crossover, a single scary news headline — none of those qualify.

What qualifies? Murphy names them: broken trendlines, broken support/resistance levels, broken swing structure, moving-average break of a long-tested MA. Each has its own false-signal rate. None is perfect. All are structural signals — price violating a level that has held for a long time.

Closes only

A detail that matters more than it sounds:

Dow relied exclusively on closing prices. He believed that averages had to close higher than a previous peak or lower than a previous trough to have significance. Dow did not consider intraday penetrations valid.

Any intraday spike that pokes through a level and then retreats is noise, per Dow. Only closes count. Modern traders often forget this and get head-faked by intraday violations. When you mark your own levels, use close-based confirmations unless you have a specific reason not to.

Failure vs. nonfailure swings

Dow wrote specifically about two reversal patterns, which Murphy labels failure and nonfailure swings:

• Failure swing: a new rally fails to exceed the previous peak, then price drops through the previous trough. Two lower peaks, two lower troughs. Clear sell signal. The weaker setup.
• Nonfailure swing: the new rally exceeds the previous peak, then price drops through the previous trough. Only lower lows exist so far, not lower highs. Some Dow theorists wait for a lower high before confirming the reversal. The stronger setup, but requires more patience.

The general message: structural reversals look different from single-bar reversals. If you can't see at least two peaks and two troughs making a lower or higher pattern, you don't have a Dow-grade signal.

The criticism and the answer

Dow Theory has been criticized forever for being slow. Murphy's response:

On average, Dow Theory misses 20 to 25% of a move before generating a signal… Those who criticize Dow Theory for failing to catch actual market tops and bottoms lack a basic understanding of the trend-following philosophy.

The stat we covered in the Trends lesson, here in its original context. From 1920–1975, Dow Theory signals captured 68% of the moves in the Industrial and Transportation Averages and 67% of those in the S&P 500 Composite (source: Barron's). Missing the first and last quarters of a move is not a bug — it is the definition of this system.

If you want to catch tops and bottoms, you want something other than Dow Theory. (And you'll probably regret not using Dow Theory.)

What Dow was actually trying to do

The surprising punchline, from Murphy:

Dow apparently never intended to use his theory to forecast the direction of the stock market. He felt its real value was to use stock market direction as a barometric reading of general business conditions.

Dow built a macroeconomic thermometer, not a trading system. The trading applications came later. He was reading the market to understand the economy. The fact that people like you and me now use his framework to plot entry and exit points is downstream of something he never intended — which is itself a pretty Dow-like observation about how markets actually use information.

Quick check

What you now know

• Dow Theory is the origin. Trends, peaks and troughs, S/R, multi-timeframe logic — they all trace to Dow's 1900-era Wall Street Journal editorials, codified after his death.
• Six tenets, each still in working order:
  1. The averages discount everything.
  2. Three trends (primary, secondary, minor) coexist.
  3. Major trends move in three phases — accumulation, public participation, distribution.
  4. The averages must confirm each other.
  5. Volume should move in the direction of the trend.
  6. A trend persists until it gives a definite reversal signal.
• Dow used closes only. Intraday pokes don't count. Something to remember when you're tempted to react to a wick.
• Failure vs nonfailure swings — two structural reversal patterns, one weaker, one stronger. Both require two peaks and two troughs before you call the trend over.
• ~68% capture, 20–25% missed. By design, not by defect. If you want tops and bottoms, you don't want Dow Theory.
• Dow wasn't trying to build a trading system. He was building a barometer for the economy. Everything we've taught is downstream of his framework applied to a purpose he didn't originally intend.

Next: Wyckoff Phases. Richard Wyckoff elaborated Dow's accumulation/distribution into a much more specific five-phase schema, with named events inside each phase. Where Dow is the skeleton, Wyckoff is the anatomy.