Crab Pattern — The Deepest Extension, the Best R:R

The Crab is a pattern for those who think the Butterfly reverses too early. Scott Carney described it in 2000 and for years called it his favorite setup — and it's defined by one thing: the DEEPEST extension in the entire harmonic family. Point D doesn't land just past the X extreme, as it does in the Butterfly; it lands at 161.8% of wave XA — far beyond the structure, at a spot where the last wave no longer looks like a breakout, but like panic or euphoria.

And that's exactly where the popular reputation for "the best R:R among harmonics" comes from. Reversals from extremely overshot levels can be violent — a market that has overshot tends to come back fast and deep — and the room for the trip back through the whole structure (B, C, A) is enormous relative to the risk. Sounds like a free lunch? It isn't. You pay for the depth of the extension with the rarity of the setup and the brutal character of wave CD, which you have to sit through without adding to your position "in installments." You'll find the shared primer for the whole family in our harmonic patterns guide; here we take apart the most stretched-out member of the family.

Pattern Structure

The Crab is an extension-type XABCD: the bullish version draws a letter "M" with the last leg dropping deep below point X, the bearish an inverted "W" with D high above X.

The Crab's signature is a contrast: shallow B, extreme D. The market makes a listless correction, seemingly returns to the trend — and then wave CD launches with a force that breaks X and keeps going, all the way to 161.8% of the whole impulse. That disparity sets the Crab apart from the Butterfly (there, B = 78.6% XA, and D only 127.2%) and from the Bat (similar B, but D doesn't break X at all). There's also a "Deep Crab" variant with B at 88.6% XA — the rest of the rules stay the same; we mention it so you don't automatically reject a setup with a deep B, though it's formally a separate variant.

The ±3-pp tolerance applies, but in the Crab the critical level is D: it's the 161.8% XA level that sets it apart from an "overstretched Butterfly." If you have to pick one ratio to enforce strictly, pick this one.

How to Measure Point D

As with every extension-type pattern, D is measured as an extension of wave XA: a grid from X to A, read at 161.8%. The PRZ zone comes from three measurements taken once XA, AB and BC are complete:

1. 161.8% extension of XA — the core and the constitutive condition. Without this level, there's no Crab.
2. 261.8–361.8% extension of AB — a measurement of wave CD relative to AB (alternatively 2.24–3.618 from BC, depending on where C landed). The range is wide, so use it to confirm, not to define the zone on its own.
3. Projection from the structure — check whether 161.8% XA overlaps with an outer level other traders are watching: old higher-timeframe support/resistance, a supply zone, a liquidity cluster. A Crab without external confluence is just a lonely ruler in a vacuum.

Two measurement warnings. First, the further out you project, the more sensitive the result is to your choice of points X and A — an error of a few candles when anchoring the grid shifts D by a distance larger than your entire stop. Draw off clear, unambiguous swings, or not at all. Second, wave CD in the Crab often accelerates toward the end (capitulation/euphoria): price can blow through 161.8% with momentum and only reverse a moment later. So treat D as a zone with a buffer, and enter only after a reaction.

How to Trade It

Entry. No blind limit orders. In the 161.8% XA zone you're catching the market at its most emotional phase of the move — entering without confirmation means standing in front of a runaway capitulation. Wait for a rejection candle with a clear wick, an engulfing candle, a structure shift on a lower timeframe. A hallmark of good Crabs: the reversal comes FAST and aggressively; if price in the D zone just sits and stews, there's no edge here.

Stop loss. X was broken a long time ago, so the boundary is set by the outer ratio: classically around 2.0 XA (some practitioners use 2.24 XA) with a buffer for noise. The stop looks distant in nominal terms, but measure it against the structure: from 161.8% to 2.0 XA is roughly 38% of the length of wave XA — while the room for the trip back on the other side is many times larger. A break of 2.0 XA says plainly: this wasn't an overshoot, it's a new trend.

Targets. This is where the Crab shows why it's loved. First target: 38.2–61.8% retracement of wave CD — and since CD is the longest wave in the family, even this "modest" level sits far from entry. Second: the area around point B, then C. Third, on a full reversal: the area around A. The setup's geometry usually gives an R:R of roughly 1:3–1:5 reaching the second target — provided you calculated it BEFORE entering on the specific setup, rather than trusting an average from the internet. Standard execution template: part of the position at the first target, breakeven, the rest at B/C.

A numerical example, for the memory. A bullish Crab: an XA impulse from 100 to 110. B at 50% (roughly 105), a bounce BC, then wave CD breaks X and slides down to D = 161.8% extension of XA, roughly 93.8. Stop below 2.0 XA (roughly 90) with a buffer: risk of about 4 units. First target at 38.2% retracement of CD (roughly 98.5 — a gain of 4.7), second at B ≈ 105 (a gain of over 11, R:R about 1:2.8), third in the area of A = 110 (R:R about 1:4). The numbers show where the "best R:R" legend comes from — and why a single stop-out on this setup hurts less than it seems, as long as you sized your position off the stop.

Context filter. The best Crabs complete at levels where the overshoot makes systemic sense: a sweep of an important higher-timeframe low/high, an old accumulation zone, or the area around major round price numbers. A Crab in the middle of nowhere is just geometry.

Common Mistakes

Catching the knife before the zone. The costliest mistake: price breaks X, "it's already cheap," you enter at 127.2% — but this is a Crab, not a Butterfly, and half of the wave is still left before D. Tell them apart at point B, and by discipline: until 161.8% XA is reached, there is no pattern.

Averaging into wave CD. CD in the Crab is often capitulation — adding to a position "because it must turn around by now" turns a small stop into a catastrophe. One entry, in the zone, after a reaction.

A nominal stop instead of a structural one. Since D sits far from the structure, it's tempting to use a tight stop "just under the wick." Price in an overshoot zone is jittery; the boundary is 2.0 XA, and you size the position to the stop, not the other way around.

Trading the Crab on M1–M5. Extensions on low timeframes are mostly noise and slippage; on top of that, the setup is rare, so scanners "find" it where it doesn't exist. H1 and up, ideally H4/D1.

Believing "the best R:R" is a statistic. Time for honesty, as throughout this series: R:R is the geometry of the setup, not its win rate. Large-scale research at the Bulkowski standard for harmonics DOES NOT EXIST — the "Encyclopedia of Chart Patterns" doesn't catalog them, and private backtests have small samples and unverifiable recognition rules. Add the subjectivity of drawing (a different X swing = a different Crab, or none at all) and the rarity of the setup, which means building your own statistics takes months. A high R:R with an unknown hit rate can be an edge — or a stop-loss machine. Only your journal decides.

FAQ

What is the Crab pattern? An extension-type XABCD structure described by Scott Carney in 2000. Signature: a shallow B (38.2–61.8% retracement of XA) and the deepest D in the family — 161.8% extension of XA, far beyond point X. The bullish version buys a deep overshoot of a low; the bearish sells an overshot high.

Where does the "best R:R" reputation come from? From geometry: a stop behind 2.0 XA sits relatively close to entry, while the targets (CD retracements, then B, C, A) are far away, because you're traveling back through the longest CD wave in the family. Reaching the area of C usually gives an R:R of 1:3–1:5. That said, it's geometry and a practitioner's opinion, not a measured win rate.

How does the Crab differ from the Butterfly? Butterfly: B = 78.6% XA, D = 127.2% XA — a reversal just past the extreme. Crab: B = 38.2–61.8% XA, D = 161.8% XA — a reversal after a full overshoot. Confusing them is costly: entering "Butterfly-style" on a Crab still building all but guarantees you get stopped out.

FAQ

Rafał — Krypto Bez Ściemy

Trader and founder of Strefa Tradingu. He’s been breaking crypto down on YouTube for years — no hype, no signals, with a focus on market structure and risk management.

kryptobezsciemy.plInstagramYouTubeKlub Zero