Pring builds every chart pattern out of three plain building blocks: trend classification, support/resistance, and trendlines, and explains the crowd psychology behind each one clearly. His one real backtest, in the final chapter, covers only two of the fifteen patterns he teaches.
The one idea
The book's real argument is that a chart pattern is not a shape you memorize, it's a picture of a specific psychological standoff between buyers and sellers. Support and resistance form where enough people remember getting burned at a price to act differently the next time they see it. A trendline is just a ruler laid across a series of those decisions. Once you see patterns this way, Pring argues, you stop hunting for textbook-perfect shapes and start reading the underlying trend, which is what he says the professionals actually do.
The book organizes technical analysis into layers. Trend classification (primary, intermediate, short-term) tells you what magnitude of reversal you're even looking for. Support and resistance zones tell you where the crowd is likely to fight. Trendlines connect those fights into a line you can watch break. Volume tells you whether the fight was real or hollow. Only after all four are in place does Pring walk through individual patterns (rectangles, head-and-shoulders, triangles, broadening formations, and a long section on one- and two-bar reversals) as applications of the same four ideas, not as a new rulebook each time.
The last two chapters are different in kind from the rest of the book. Chapter 17 is a practical checklist for telling a real breakout from a whipsaw. Chapter 18 reports an actual quantitative study Pring commissioned from Recognia: 5,235 computer-identified patterns across four U.S. industry sectors, 1982 to 2003. It is the only place in the book where a claim is tested against data rather than illustrated with a hand-picked chart.
What it actually teaches
Pring's approach is not one linear checklist, it is four building blocks applied in this order every time you evaluate a chart:
CLASSIFY THE TREND YOU'RE READING
Primary trends run the length of the business cycle, roughly 3.6 years trough to trough, with bull and bear legs of 1 to 2 years each. Intermediate trends last six weeks to nine months. Short-term trends run three to four weeks. The size of trend a pattern is reversing sets how long the pattern should take to form: three-plus months for a primary reversal, three to six weeks for an intermediate one, one to two weeks for a short-term one.
MAP SUPPORT AND RESISTANCE FROM PRIOR ACTION
Look for previous highs and lows, round numbers, and points where a trendline or moving average intersects price. A zone's strength depends on three things: how much volume traded there, how fast and far the prior move into it ran, and how much time has passed since it last mattered. Support that breaks becomes resistance on the way back up, and resistance that breaks becomes support on the way down.
DRAW THE TRENDLINE ON AT LEAST TWO TOUCHES
A line through a single point is not a trendline, Pring calls it 'worse than drawing nothing at all' because it looks meaningful and isn't. Significance comes from three factors: the length of the line, the number of times it has been touched or approached, and how shallow its angle is. Use closing prices, not intraday highs and lows, since closes better reflect who was willing to hold overnight. Note that a rising trendline breaks sooner on a logarithmic scale and a falling trendline breaks sooner on an arithmetic scale, so the choice of scale changes your signal.
DON'T ACT UNTIL THE BREAK IS DECISIVE
Pring cites the old 3 percent penetration rule from the early 1900s but says it's too blunt for intraday charts and too arbitrary across different volatility levels. His practical filter is to require the price to hold beyond the boundary for at least two bars, one bar closing past the line can just be exhaustion. Watch the trend above the one you're breaking out on: contra-trend breakouts (a bullish pattern completing inside a primary bear market) fail far more often than pro-trend ones.
MEASURE THE OBJECTIVE FROM THE PATTERN'S HEIGHT
Project the vertical distance between the pattern's two boundaries from the breakout point, in the direction of the break. Treat this as a minimum, not a ceiling, price often runs to multiples of the objective. Always do this on a logarithmic scale: on an arithmetic chart the same projection can show an objective of zero or less, which Pring calls 'clearly a very unlikely possibility.' Check volume on the way out: a genuine breakout should show rising activity, and a break on shrinking volume is a warning that the move is running on a lack of sellers rather than real buying.
What it looks like Monday morning
A reader who takes this book seriously stops drawing trendlines through single price touches and starts asking, for every level on the chart, how much volume traded there and how many times price has already bounced off it. They switch their charting software to logarithmic scale before measuring any pattern objective, since Pring shows the arithmetic scale can produce a nonsense target. And before acting on any breakout, they check the trend one timeframe up: a bullish pattern completing inside a larger downtrend gets treated as lower-probability by default, not equal odds.
What they don't do, because the book itself doesn't support it, is treat any of this as a tested trading system. Chapter 18's numbers apply only to head-and-shoulders and double top/bottom patterns in four sectors between 1982 and 2003, with no stop-losses, commissions, or slippage built in. Everything else, rectangles, triangles, broadening formations, one- and two-bar reversals, is explained through hand-picked chart examples, not backtested.
| Trade | Result | Capital after |
|---|---|---|
| 1 | -50% | $50.00 |
| 2 | +50% | $75.00 |
| 3 | +10% | $82.50 |
| 4 | +10% | $90.75 |
| 5 | +10% | $99.83 |
Four winning trades in a row after one 50 percent loss (an 80 percent win rate) still ends below the starting balance, which is why Pring treats the false-breakout stop as more important than the pattern's win rate.
Trading losses usually happen when the trader is positioned in a countercyclical position against the main trend.Martin J. Pring, Martin Pring on Price Patterns
Where it fails
- The one real test covers two patterns out of fifteen. Chapter 18's 5,235-pattern study, run by Recognia across four sectors (financial, energy, transportation, retail) from 1982 to 2003, tested only head-and-shoulders and double top/bottom patterns. Rectangles, triangles, broadening formations, and the entire second half of the book on one- and two-bar reversals were never quantitatively tested; every claim about them rests on selected chart examples.
- The backtest explicitly excludes stop-losses and trading costs. Pring states directly that the research 'permitted a theoretically unlimited loss from short sales or a 100 percent loss from long positions' and did not model prudent money management, commissions, or slippage. The headline result, that 70 percent of bullish patterns in bull markets reach their price objective within five pattern-lengths, measures whether the target was ever touched, not whether a disciplined trader using stops would have captured the move.
- The sample knows the primary trend in advance; you won't. The study classified each period as a primary bull or bear market 'with the benefit of hindsight' before scoring a pattern as pro-trend or contra-trend. In real time nobody has that information. The contra-trend numbers are the honest finding here: bullish breakouts in bear markets reached objective only 55 percent of the time versus 70 percent for pro-trend signals, a fact that arrives on page 344 of a 360-page book, not in the marketing copy.
- Every numeric rule is immediately walked back to 'use judgment.' The 3 percent breakout-confirmation rule, the two-bar hold rule, and the test for a 'decisive' breakout are each introduced and then qualified as depending on volatility, time frame, and experience within the same page. A reader looking for a mechanical, backtestable ruleset, which the book's subtitle promises as 'the definitive guide,' gets a framework for judgment calls instead.
- The window ends in 2003 and was never revisited. This is the only edition of the book; the copyright page and ISBN confirm no later printing. The 21-year test window (1982-2003) predates decimalization's effect on tick size, algorithmic execution, and the correlation-driving rise of sector ETFs, all of which changed how patterns in individual stocks behave, and none of it is addressed because there was no revision to address it in.
- The chart examples are chosen to prove points already made. Outside Chapter 18, every worked example (the NCR head-and-shoulders, the Nabors outside bar, the Newmont Mining triangle failure) is selected because it illustrates a rule Pring has just stated, not sampled at random. That's normal for a teaching book, but it means the vivid, memorable examples and the one systematic test in the book are pulling in different directions: the examples make the patterns look reliable, and the test shows contra-trend versions failing close to half the time.
Who it is for
Buy it if
Buy this if you want the clearest available plain-English explanation of why support, resistance, and trendlines work the way they do, tied to real crowd psychology rather than arbitrary shapes. It rewards a reader who will treat Chapter 18's numbers as directional evidence about pro-trend versus contra-trend odds, not as a tradeable system.
Skip it if
Skip it if you already understand pattern psychology and want a mechanical, backtested rulebook with stop-loss logic and win rates you can plug into a system. The book tests only two of its fifteen patterns, excludes stops and costs from that test, and hedges nearly every numeric threshold as a judgment call.
As an Amazon Associate I earn from qualifying purchases. It costs you nothing extra, and it does not change what gets recommended or what the verdict says.