Fibonacci, Elliott & Harmonics
Fibonacci Trading: How to Master the Time and Price Advantage
Boroden's rule is simple: never trade a single Fibonacci level, only a zone where at least three separate Fibonacci measurements land within a few ticks of each other. She then does the same thing to the calendar, projecting the same ratios forward in trading days to find when a turn is due.
The one idea
Almost everyone who draws a Fibonacci retracement draws one line off one swing and then waits to see if price respects it. Boroden's argument is that a single line is close to worthless, and that the only thing worth trading is a coincidence. Take every meaningful swing on the chart, run three different kinds of Fibonacci measurement off all of them, and look for the places where the resulting levels pile up on top of each other. She calls that pile a price cluster and it needs a minimum of three separate relationships landing in a tight range. Anything less is noise.
The three measurements are different animals and that matters, because it is what makes the overlap non-trivial. A retracement measures how far back into a prior swing price has come (0.382, 0.50, 0.618, 0.786). An extension measures how far past a prior swing price has gone (1.272, 1.618). A projection compares two swings moving in the same direction and asks whether the new one will equal the old one (1.00 and 1.618), which is just a measured move with a Fibonacci label on the 1.618 version. Three unrelated ways of measuring, all pointing at the same price, is her definition of a decision point.
The second half of the book is the part almost nobody else was teaching in 2008. She takes the identical ratios and applies them to the horizontal axis. Count the trading days between two swing highs, multiply by 0.618, and mark that date forward. Do it for every pair of swings on the chart. Where three or more of those dates land within a day or two of each other, you have a time cluster, a window when a reversal is more likely. Her whole thesis is that a price cluster and a time cluster arriving together is the highest-odds moment the chart will ever hand you. She calls it synchronicity.
What it actually teaches
The book gives an explicit five-step sequence in Chapter 16, and every earlier chapter is one step of it expanded.
MARK THE SWINGS
Pick the obvious swing highs and lows. Her rule for relevance is strict and it is where she says students go wrong most often. When you are looking for support after a pullback, run retracements from every prior low up to the single highest high on the chart. When you are looking for resistance after a bounce, run them from every prior high down to the single lowest low. The moment a new extreme is made, everything you drew is dead and has to be redrawn. Old support does not become new resistance: she rejects that idea outright and says to just re-measure instead.
RUN ALL THREE PRICE TOOLS
Retracements at 0.382, 0.50, 0.618 and 0.786, adding 0.236 only when the swing is long. Extensions at 1.272 and 1.618, sometimes 2.618, and 4.236 only on a wildly extended move. Projections at 1.00 and 1.618 using a three-point tool that compares swings in the same direction. Her working set of ratios is 0.382, 0.50, 0.618, 0.786, 1.00, 1.272 and 1.618, with 0.236, 2.618 and 4.236 as occasional extras. The derivations are ordinary algebra: 0.786 is the square root of 0.618, 1.272 is the square root of 1.618, 0.382 is 0.618 squared.
FIND THE SETUP
Three or more relationships overlapping in a tight range is Setup 1, the price cluster. A single 1.00 projection of a prior corrective swing, taken in the direction of the trend, is Setup 2, symmetry. Setup 3 is the two-step, a zigzag where the 0.618 or 0.786 retracement of a to b overlaps the 1.272 or 1.618 extension of c to d and the 1.00 projection of b to c from d, meaning leg bc equals leg de. She only takes setups pointing with the trend, defined loosely as higher highs and higher lows or the reverse. Countertrend clusters are for managing exits, not for entries.
OVERLAY TIME (OPTIONAL)
Two-point time cycles use 0.382, 0.50, 0.618, 0.786, 1.0, 1.272, 1.618 and 2.618 measured high to high, low to low, high to low and low to high. Three-point time projections use 1.0, 1.272 and 1.618. She counts trading days, not calendar days. A time cluster is three or more cycles landing within one to three trading days of each other on a daily chart, or one to three bars intraday, and the tradable window is that span plus one bar on each side. She will not run time work below a 15-minute chart, and never on tick or volume charts because the bars are not equal in time.
TRIGGER, STOP, THREE TARGETS
The zone alone is not a trade. Wait for a trigger, usually a break of a prior swing high or low on a faster chart. Maximum stop is a few ticks beyond the far edge of the cluster. First target is always the 1.272 extension of the swing that ran into the zone, second is 1.618, third is 2.618. Her tolerance for a level counting as hit is 3 to 4 ticks either side, wider in forex or off a very large swing. Her top-shelf version, the ideal setup, is a symmetry projection on a 3 or 5 minute chart where price is on the correct side of the 34 EMA and both the 14-bar and 50-bar CCI are on the correct side of zero.
What it looks like Monday morning
Stop trading single retracement lines. The concrete change is that you draw retracements from every qualifying swing rather than one, and you only mark the chart where three of them overlap. On a stock where the levels are wide, that means a zone of roughly half a percent; on index futures it means five or six ticks. Everything else you drew gets ignored. That one habit is the book's actual deliverable and you can adopt it in an afternoon.
The second change is where your stop goes and where your first target goes. Both become mechanical: stop a few ticks past the far edge of the zone, first target the 1.272 extension of the leg that just ran into it. That gives you a risk-to-reward number before you enter instead of after. If you want the time half, run it on daily charts only at first, counting trading days between swings and multiplying by 0.618, 1.0 and 1.618. Do not attempt her 3-minute intraday version unless you are at the screen all day, because it has to be redrawn from scratch every time a new swing high or low prints.
| Relationship | Swing measured | Arithmetic | Level |
|---|---|---|---|
| 1.0 projection | 12653 high to 12587 low, projected from the 13173 high | 66 points down from 13173 | 13107 |
| 1.0 projection | 12847 high to 12782 low, projected from the 13173 high | 65 points down from 13173 | 13108 |
| 1.272 extension | 13124 low to 13173 high | 49 x 1.272 = 62.3 down from 13173 | 13111 |
| 1.0 projection | 12843 high to 12782 low, projected from the 13173 high | 61 points down from 13173 | 13112 |
The low printed at 13113, one tick above the zone, and a 75-point rally followed, worth $375 per contract. Notice that three of the four legs are ordinary measured moves at a 1.0 ratio: they clustered because the three prior declines were 66, 65 and 61 points, not because of any Fibonacci ratio.
My definition of a price cluster is the coincidence of at least three Fibonacci price relationships that come together within a relatively tight range.Carolyn Boroden, Fibonacci Trading
Where it fails
- She never defines the one number the whole method depends on. Everything hangs on "a relatively tight range," and she quantifies it for time (one to three trading days) but never once for price. Work it out from her own examples and the definition swings wildly. The E-mini Dow cluster runs 13107 to 13112, five points, 0.04 percent of price. The Google cluster runs 483.74 to 486.53, which is 0.58 percent. That is a fifteen-fold difference in what counts as tight, decided by eye. Her stated tolerance for a level being hit is 3 to 4 ticks, but she also writes that the real test is whether you see "a glaring violation or shortfall" when you look at the chart. A rule that resolves by eyeballing is not a rule, it is a habit, and yours will not match hers.
- The obvious objection is that enough lines guarantee a cluster, and she dismisses it rather than answering it. Count what she puts on a chart. Ten ratios, three tools, times every swing. Her Home Depot walkthrough names five swings for retracements alone, which is twenty lines before a single extension or projection is drawn. She warns readers in Chapter 2 that the book's chart images look blurry, then explains why: the levels are "clustering and essentially overlapping one another, making the chart levels difficult to read," and adds "this is something that we actually want to see happen." That is the criticism stated in the author's own words as a feature. When the randomness argument comes up directly, her whole response is a sidebar: she chuckles silently and says it is not worth arguing with people who refuse to do their homework. To her credit she does insist that a cluster is never a trade without a separate trigger, which is a genuine defense. She just never makes it as one.
- Not one number in 304 pages is tested. There is no backtest, no hit rate, no sample size, no expectancy math, no losing-trade tally. The first target "is met a high percentage of the time" and she never says what percentage. She admits it once explicitly about the CCI zero line reject: "I don't have backtested data on this." The same is true of everything else in the book. Worse, look closely at her flagship cluster and three of its four legs are plain measured moves at a 1.0 ratio. They landed within five points of each other because the three prior declines happened to be 66, 65 and 61 points, which has nothing to do with Fibonacci. She even concedes symmetry is "most commonly called a measured move."
- The timing half has no failures in it at all. On the price side she is reasonably honest. She shows a Google projection that did nothing (Figure 5-7) and a Russell chart where two clusters broke in a row (Figure 6-30), and she states plainly that clusters get violated every day. On the time side, every single example in Chapters 11 through 13 works. Fourteen or so time clusters, fourteen or so reversals. She asserts in a sidebar that time clusters fail like price clusters do, then never shows one. Since the time half is the book's selling point and the part you cannot verify from memory, that is where an honest failure example was most needed and where none appears.
- The data is one calm market, and the book shipped straight into the opposite one. Nearly every chart is drawn from 2004 to 2007, with a heavy concentration in January to April 2007: orderly swings, clean higher highs, low volatility. The method needs exactly that. It requires identifiable swings, a directional pattern, and price that walks into a zone rather than jumping over it. The words gap, slippage, commission, overnight, news, earnings and volatility do not appear anywhere in the book. Not once. A max-risk stop placed a few ticks beyond a cluster edge is a stop that gets jumped in a gapping market, and she published in 2008 with nothing to say about it. She does tell you to stand aside when the market goes sideways, which is honest, but chop and crashes are most of what came next.
- The book routes you to paid products for the two things that make it usable. The automated time histogram, which is what makes intraday timing practical rather than impossible, is "only available within the Dynamic Trader program." Dynamic Trader belongs to Robert Miner, her mentor of twenty years, who wrote the book's foreword. The other missing piece is the trigger. She says repeatedly that whether a setup pays comes down to the trigger, then spends Chapter 14 pointing at other people's tools: Woodie's CCI, the TTM squeeze from Carter and Senters, whose chat room she merged with hers. Chapter 17 is several pages of NinjaTrader screenshots. Her own Author Tip explains that redrawing the analysis after every new swing "becomes tedious," which is "one of the reasons traders choose to let me do the work in the chat room." That is the book telling you its method is too laborious to run yourself, and naming the subscription that solves it.
- It quietly does not work for anyone with a job. The ideal setup lives on 3 and 5 minute charts. She updates about sixteen charts down to the 3-minute frame, and every new swing high or low invalidates the projections and forces a redraw. That is a full-time occupation requiring real-time futures data, futures margin, and an uninterrupted trading session. The book never says any of this out loud, and never suggests a version scaled down for someone checking charts twice a day. If you cannot sit at the screen from open to close, the only piece of the method you can actually use is the daily-chart price cluster, which is maybe a third of the book.
Who it is for
Buy it if
You already trade discretionary swings or intraday and you already have an entry trigger you trust, and what you lack is a disciplined way to pick where the trade goes on. This book is that. It is also the right buy if you currently draw one 0.618 line off one swing and treat it as gospel, because the confluence rule is a straight upgrade over that.
Skip it if
Skip it if you want evidence a method works before you fund it, because there is none here. Skip it if you trade around a job, since the parts she is proudest of need all-day screen time. And skip it if you already own Pesavento or Carney on harmonic patterns, because the two-step chapter is a thinner version of the Gartley material and she says so herself.
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.