Trade What You See: How to Profit from Pattern Recognition
Larry Pesavento and Leslie Jouflas · 2007 · 225 pages
Pesavento and Jouflas teach five geometric price patterns built from Fibonacci ratios, each with defined entry, stop, and target rules. The patterns are genuinely learnable in a weekend; the 70 percent win claim behind them is not.
The one idea
The book's whole system rests on one idea: price swings are not random, they repeat in measurable proportions tied to the Fibonacci ratios .618, .786, 1.272, and 1.618. Every chart pattern in the book, the AB=CD, the Gartley 222, the Butterfly, and the Three Drives, is the same basic X-A-B-C-D swing structure with a different completion ratio at the far end. Learn to label the swing and measure the retracement, and you can apply the same method to a stock, a futures contract, or a currency pair on any time frame.
Each pattern comes with a rule for what invalidates it (a leg exceeding a prior point, for example) and a rule for where the trade completes, which the authors then pair with a fixed dollar or percentage stop-loss and a two- or three-part scale-out at the .618 and .786 retracement levels. The appeal is that a trader can, in theory, know the exact risk and the exact reward zone before the pattern even finishes forming.
The other half of the book, roughly a third of it, is not chart method at all. It covers thinking in probabilities, tracking win/loss statistics, sizing positions to 1 to 3 percent of capital, and building a written trading plan. That material is standard risk-management advice, competently explained, but it is not what a reader buys this specific book for.
What it actually teaches
The core method is a repeatable sequence for spotting, measuring, and trading each harmonic pattern.
LABEL THE SWING X-A-B-C-D
Mark the starting high or low as X, then label each subsequent swing point A, B, C, and D as price moves. The AB=CD pattern uses only A, B, C, D (three legs); the Gartley and Butterfly patterns add the X anchor point (four legs) so the AB=CD sits inside a larger XA swing.
CHECK THE BC RETRACEMENT
The BC leg should correct to one of the book's primary Fibonacci levels: .382 (strong trend), .50, .618, or .786. A retracement that stops short at .382 signals a fast, extended CD leg to come; a deep .618 or .786 retracement signals a CD leg roughly equal to AB.
PROJECT THE COMPLETION RATIO FOR THE SPECIFIC PATTERN
Each pattern has its own valid completion zone at D: an AB=CD completes when CD equals AB or extends it 1.272 to 2.00 times; a Gartley 222 completes at the .786 retracement of the XA leg and D can never exceed X; a Butterfly is by definition a failed Gartley where D pushes past X to the 1.272, 1.618, or 2.00 extension of XA, with 2.618 as the point beyond which the pattern is negated and the trend probably just continues.
READ THE WARNING AND CONFIRMATION SIGNS
Before treating a pattern as tradable, look for gaps in the CD leg, wide range bars (roughly twice normal size), and tail closes near the completion point, all of which suggest the pattern will overshoot. Confirmation instead comes from a tweezer top or bottom, a gap in the direction of the intended trade, or simply waiting one full time bar past the completion point before entering.
SIZE THE TRADE TO 1 TO 3 PERCENT OF CAPITAL
Pick a stop-loss price first, then divide your fixed dollar risk (1 percent of a $100,000 account is $1,000) by the per-share or per-contract stop distance to get position size. If the resulting risk on a valid pattern doesn't fit inside that percentage, the book's instruction is to pass on the trade rather than shrink the stop.
SCALE OUT AT THE RETRACEMENT LEVELS AND MOVE THE STOP TO BREAKEVEN
The book's default trade management takes a first partial exit at the .618 retracement of the completed swing (often close to the amount originally risked), moves the stop to breakeven at that point, and takes a second exit at the .786 retracement or at a 1.272/1.618 extension for Butterfly trades.
What it looks like Monday morning
A reader who takes this seriously starts by printing charts and hand-drawing the X-A-B-C-D swings on whatever they already trade, because the authors are explicit that this pattern recognition is a trained-eye skill, not software output. The realistic Monday-morning change is adding a Fibonacci retracement/extension tool to the charting platform and running the invalidation checklist (does C exceed A, does D exceed X) on the next few setups that show up, rather than trading blind pullbacks.
The other concrete change is adopting the 1 to 3 percent position-sizing formula and the two-rule discipline at the end of the money management chapter: always use a stop-loss order, never move it further away. That part of the book is worth keeping regardless of what a reader concludes about the patterns themselves.
| Pattern | Legs | D completion (from Chapter text) | Invalidated if |
|---|---|---|---|
| AB=CD | 3 (A-B-C-D) | CD equals AB, or extends it 1.272 to 2.00+ times AB | D fails to exceed B, or BC retraces more than 1.00 of AB |
| Gartley 222 | 4 (X-A-B-C-D) | .786 retracement of the XA leg | D exceeds X, or C exceeds A, or B exceeds X |
| Butterfly | 4 (X-A-B-C-D) | 1.272 to 2.618 extension beyond X of the XA leg | D fails to exceed X, or the extension passes 2.618 |
All three patterns must contain a valid AB=CD leg inside them; the book treats the Gartley and Butterfly as that same three-leg pattern anchored to an added X point.
Rule #1: Always use stop-loss orders. Rule #2: Never violate Rule #1.Pesavento and Jouflas, Trade What You See
Where it fails
- The 70 percent win rate has no visible data behind it. The book repeats Gartley's original claim from his 1935 book, Profits in the Stock Market, that the pattern worked 7 out of 10 times over a 30-year period, and says testing over the past 40 years has proven it accurate, but no sample size, date range, market list, or methodology is given anywhere in this edition. A reader is asked to take the win rate on faith.
- Pattern validity is a judgment call, not a hard rule. The book itself admits about 40 percent of AB=CD patterns are perfectly symmetrical and 60 percent are not, that harmonic numbers are zones to watch rather than exact prices, and that a wide range bar is roughly twice normal size. Two traders following the same chapter can label the same chart differently, which undercuts the claim of a quantified, repeatable edge.
- Every full worked trade example in the book is a winner. Chapters 4 through 6 walk through named trade setups (S&P E-mini, soybeans, Google, Pfizer, crude oil, Wal-Mart), and outside a couple of intentionally labeled failed-pattern examples, the featured trades all end in a profit or a small, controlled stop-out. There is no multi-month track record, drawdown curve, or losing streak shown, so the reader never sees a genuinely bad stretch with this method.
- A third of the book is generic trading-plan material, not the method. Chapters 11 through 14, roughly 55 of 225 pages, cover thinking in probabilities, keeping win/loss statistics, building an annual business plan, and daily mental preparation. It is competent advice but is not specific to harmonic patterns and could sit in almost any trading book from this era.
- Chapters 2 and 3 spend real page count on astrology and Gann lore before the method starts. Before the AB=CD pattern is introduced on page 39, the book covers W.D. Gann's disputed $55 million legend, George Bayer's Egg of Columbus, moon-cycle stock research, and financial astrologer Evangeline Adams. None of it is load-bearing for placing a trade, and it delays the actionable content.
- The book predates the market structure it's teaching people to read. Published in 2007, the S&P E-mini harmonic numbers it cites (5.4, 6.85, 8.7, 10.8 points) and the assumption of clean, hand-drawable swing repetition were formed before high-frequency and algorithmic order flow reshaped intraday price action on the same instruments. The book gives no acknowledgment that this could change the numbers.
Who it is for
Buy it if
A discretionary technical trader who already reads price charts and wants a structured vocabulary for geometric patterns, with concrete Fibonacci levels and defined stop-loss placement instead of vague trend-line calls. Also useful for a swing trader who wants a repeatable checklist for entries and scaled exits.
Skip it if
Anyone who wants a backtested, statistically verified system before risking money should skip this; the win-rate claims are asserted, not shown. Also skip it if you don't have real-time charting software with a Fibonacci tool and the screen time to hand-practice pattern recognition, since the book assumes both.
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.