Teaches four Fibonacci-ratio chart patterns (AB=CD, Gartley 222, Butterfly, Three Drives) with exact retracement levels and worked trade examples. The patterns are precisely defined; the claim that they make money is not backed by any data the reader can check.
The one idea
The book's whole system rests on one geometric shape: a price swing that moves, retraces, then moves again (AB=CD), measured against a handful of Fibonacci ratios (.382, .50, .618, .786 for retracements; 1.272, 1.618, 2.0, 2.618 for extensions). Every other pattern in the book, the Gartley '222', the Butterfly, the Three Drives, is that same AB=CD leg bolted onto a fourth point (X) that anchors it. Learn to spot the three-leg shape and the four-leg shape, and you can supposedly spot all of them.
Each pattern comes with hard invalidation rules, not vague guidelines. A Gartley pattern is dead if point C exceeds point A, or if B or D exceed X. A Butterfly pattern requires D to extend past X, capped at 2.618 of the XA leg (beyond that, the authors say the pattern is 'negated'). This is the book's real strength: it turns a subjective-sounding idea ('this chart looks like it might turn') into a checklist a trader can apply the same way every time.
The other half of the book is trade mechanics: where to place the entry limit order, where the stop-loss goes, when to scale out, and how to size the position as a percentage of account capital. The authors walk through more than a dozen real trades (S&P E-mini, Google, Pfizer, soybeans, crude oil, Wal-Mart, Intel, the euro) showing entry, stop, and exit prices to the tick or the penny.
What it actually teaches
The book's approach, stripped to its working parts, runs in this order every time a setup forms:
SPOT THE LEG STRUCTURE
Find the swing points that form the pattern's skeleton: three points (A, B, C, with completion at D) for a plain AB=CD, or four (X, A, B, C, D) for a Gartley or Butterfly. The AB leg is the impulse move; BC is the correction; CD is the leg you're trying to project.
CHECK THE FIBONACCI RETRACEMENT
Measure where BC (or AB, in a Gartley/Butterfly) retraces to. The book's four working levels are .382, .50, .618, and .786. A shallow .382 retracement in a fast market signals a strong trend; a deeper .618 or .786 retracement signals a normal AB=CD completion. On the Gartley, C must not exceed A; on the Butterfly, the AB leg typically sits at .618 or .786 of XA.
PROJECT THE COMPLETION POINT
For a plain AB=CD, take the AB leg's point spread, multiply by 1.272 or 1.618, and add it to (or subtract it from) point C to get the D target. For a Gartley, D must land inside X (never past it) and typically completes near the .786 of XA. For a Butterfly, D must extend beyond X, projected to 1.272 or 1.618 of the XA leg, with 2.618 as the hard ceiling before the pattern is voided.
READ THE WARNING AND CONFIRMATION SIGNS
Before trusting the completion point, watch for warning signs: gaps in the CD leg, wide range bars (roughly twice normal size) near D, or tail closes at the pattern's end, all of which suggest the move will overshoot. Confirmation signs, tweezer tops/bottoms, a gap in the trade's direction, or simply waiting one full time bar before entering, are the authors' answer to that risk.
SIZE THE POSITION AND PLACE THE STOP
Risk 1 to 3 percent of total trading capital per trade (the book's own bracket, aimed at novices on the low end). Divide the dollar risk by the per-share or per-point stop distance to get position size. Example given: a $100,000 account risking 3 percent ($3,000) with a $2.00 stop yields 1,500 shares. Never risk more than roughly 10 percent of a stock's price on the stop; 5 percent is called 'reasonable.'
SCALE OUT AT THE FIBONACCI TARGETS
Exit the first portion of the position at a profit equal to the amount risked, then move the stop to breakeven, converting the trade to risk-free. Take the second exit at the .618 retracement of the full AD swing, with an optional third exit at .786 or .382 if trading a three-part scale-out.
What it looks like Monday morning
A reader who takes this seriously starts by drilling one pattern, the AB=CD, on a single market's daily chart until the shape is recognizable without hesitation, then adds the Gartley '222' once that's automatic. The book's own advice is to concentrate on one or two patterns in one or two markets before adding more, which is a reasonable brake on a system this pattern-rich.
On Monday morning that means pulling up a chart, marking the last clean swing high and low, and checking whether the retracement sits at .382, .50, .618, or .786 before doing anything else. It also means writing down the invalidation rule for whatever pattern is forming (C cannot exceed A on a Gartley, D must exceed B on an AB=CD) before placing any order, and calculating position size from a fixed percentage of account risk rather than a round number of shares or contracts.
| Pattern | Key retracement leg | C-point rule | D completion vs. X |
|---|---|---|---|
| AB=CD | BC retraces .382/.50/.618/.786 of AB | No X point; CD extends 1.27 to 2.0+ of AB | No X; D must exceed B |
| Gartley "222" | AB retraces to .618 or .786 of XA | C cannot exceed A | D must stay inside X (cannot exceed it) |
| Butterfly | AB retraces .618-.786 of XA (or .382/.50) | C cannot exceed A | D must extend past X, to 1.272-1.618 (max 2.618) |
All three patterns share the same AB=CD skeleton; what changes is whether there's an X anchor point and whether D lands inside or beyond it.
Trade what you see, not what you believe.Larry Pesavento and Leslie Jouflas, Trade What You See
Where it fails
- The core statistical claim is never shown, only asserted. The book states Gartley's original 1935 claim of roughly 70 percent accuracy and says 'testing weekly, daily, and intraday patterns over the past 40 years has proven Gartley's original premise was indeed accurate,' but no sample size, date range, win/loss data, or methodology is given anywhere in the book. A reader has to take the authors' word for the single most important number in the whole system.
- Every worked example is picked, not sampled. Each chapter shows two or three winning trades and, to its credit, usually one failed one, but the failures shown are the lucky, small-loss kind (stopped out for $240 to $400) rather than the disastrous, gap-through-the-stop kind the text itself warns about repeatedly, especially for the Butterfly pattern. There is no way to know from the book how these dozen or so hand-picked charts relate to the pattern's actual hit rate across hundreds of setups.
- The method depends on subjective judgment the book cannot fully specify. Warning signs like a 'wide range bar' are defined loosely as 'twice normal size,' and confirmation relies on reading tweezer tops, tail closes, and thrust, all visual judgment calls that two traders looking at the same chart could reasonably disagree on. That subjectivity makes the system hard to backtest or automate, which is exactly why the win-rate claim above can't be independently checked.
- The trade examples are frozen in a market structure that no longer exists. Every price example dates to 2005 through 2007: crude oil at $74 to $78, Google at $384, T-bonds capped at 101, S&P E-mini in the 1200 to 1400 range. Harmonic pattern scanners (MT4/MT5 indicators, TradingView scripts) that auto-detect these exact same AB=CD, Gartley, and Butterfly shapes have proliferated since publication, which plausibly erodes whatever edge came from being one of the few traders who could spot them by eye.
- The book assumes a discretionary day trader with screen time, not a systems builder. Every setup requires watching intraday charts in real time to catch gaps, tail closes, and confirmation bars as they form. There is no code, no spreadsheet, and no mechanical rule set precise enough to hand to a computer; a reader hoping to build an automated strategy from this book will have to make dozens of arbitrary judgment calls the authors leave to 'feel.'
Who it is for
Buy it if
You already trade discretionarily off price charts and want a structured, rule-based vocabulary (specific Fibonacci ratios and invalidation rules) for patterns you may already trade on gut feel. It also suits someone building a harmonic-pattern scanner who needs the exact ratio definitions Gartley, Butterfly, and AB=CD are built from.
Skip it if
You want statistical proof an edge exists before risking money, or you're looking for a systematic, backtestable strategy you can code without filling in dozens of subjective judgment calls yourself. Skip it too if you have no chart-reading experience; the book assumes you can already read price bars and gaps fluently.
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.