Corrective Waves: the A-B-C Structure
After five waves forward the market answers with three corrective waves back, labeled A, B and C, and most of the damage traders suffer inside a trend happens because the B wave convinces them the trend never ended. The corrective structure is the impulse's mirror: smaller, stuttering, emotionally inverted, and far more dangerous per point traveled, because nothing about it feels like a trend while you are inside it. This lesson completes the wave block's foundation: impulses push, corrections answer, and the answer has a shape you can learn to read.
Why Corrections Come in Threes
Impulses push with the larger trend and can travel cleanly, because every dip finds buyers whose bias already points up. Corrections fight the larger trend, and fighting the larger trend is expensive, so they cannot trend cleanly: progress downward has to be rationed into bursts separated by relief rallies. The result is the three-wave shape, a drive down, a rally back, and a final drive, labeled A, B and C. The same shape repeats at every scale, a correction at one degree being a complete A-B-C inside a single wave at the degree above.
| Wave | Its job | Volume character |
|---|---|---|
| Wave A | First real break of the impulse's rhythm | Expanding on the initial push down |
| Wave B | Relief rally, retrace part of A | Fading, thin, unconvincing |
| Wave C | Final drive to the correction's real low | Renewed, sometimes heavy into the low |
Why three and not two, or five? Two waves cannot complete a retrace that must also trap optimism, and five belongs to the trending direction by the framework's own rule of substitution: what moves with the larger trend subdivides as five, what moves against it manages three. The number three is not decoration. It is the minimum structure that can shake out the trend's late arrivals, fake out the early bottom callers, and still finish lower, all while traveling against a current that pushes the other way.
Wave A: The First Real Break
Wave A does what every prior dip in the impulse declined to do: it breaks the rhythm. Inside waves three and five, pullbacks held above prior swing lows; wave A violates that habit, often on expanding volume, and the expansion matters because it shows real two-sided trade for the first time in the leg. Sellers are no longer merely absent, they are present in size.
The crowd's read of wave A is the source of everything that follows: a pullback within the uptrend, a chance to buy the dip at better prices, one more higher low in a sequence that has delivered nothing but higher lows. The read is not stupid, it is the read that worked four times in a row. That is what makes the correction effective. It arrives looking exactly like every ordinary dip before it, and the crowd buys it on the strength of a pattern that is, from this bar forward, no longer in force.
Wave B: The Trap
Wave B is a relief rally that retraces part of wave A, commonly half to six tenths, sometimes deeper in slower corrections. Its signature is divergence between price and participation: price recovers, sometimes impressively, but volume fades, breadth narrows, and the leaders that carried the impulse decline to lead. The market goes up on less and less, which is the definition of a move nobody is paying to join.
The crowd calls the all clear here. Analysts describe a healthy reset, dip buyers congratulate themselves, and the narrative machinery of the old trend, which never really stopped, resumes at full volume. The trap is the point of the whole structure: wave B exists to gather the maximum number of hopeful longs at the worst possible location before wave C removes their support. In the worked example ahead, the rally to 144 on fading volume is this exact moment, the entire correction's purpose compressed into one deceptive upward leg.
One blunt sentence: a rally nobody trades hard is not a recovery, it is an inventory clearing.
Wave C: The Resolution
Wave C is the final drive to the correction's real low, typically comparable to wave A in length, often equal and sometimes extending past it. The hope gathered in wave B becomes C's fuel: trapped longs from the top, new dip buyers from the B rally, and momentum sellers who finally have confirmation all sell into a market that is running out of Greater Fools in both directions at once.
Wave C ends where selling exhausts, and that ending is precisely where the next impulse will eventually be born, which is why patient traders watch C with more interest than they traded A with. What confirms the correction has ended is not a hunch or an indicator pin: it is a new impulse structure breaking above the B wave extreme with participation. In plain terms, price must climb back through the trap's high on volume that behaves like wave 3 volume, not like wave B volume. Until that happens, every rally inside the correction is a candidate for another C-degree decline.
A Worked Example: Counting the Correction
Stay with the hypothetical instrument from the previous two lessons, invented numbers throughout. The impulse topped at 154. Wave A declines to 134, twenty points, with volume expanding on the break and the first violation of the advance's pullback rhythm. Wave B rallies to 144, recovering exactly half of wave A, on volume that fades as price rises and breadth that narrows to the usual laggards. Wave C declines to 124, twenty points, nearly equal to wave A, taking out the A low and finishing the structure.
Audit the correction the way the impulse was audited. The A-B-C traveled thirty points against an advance of fifty four, a bit more than half of the impulse, and the low at 124 held far above the 100 origin, so nothing at this degree cancels the larger bullish structure. No impulse rule was available to check, because a correction is not an impulse, and its validation is proportional and structural: B's half retrace, C's equality with A, the shape completing where exhaustion, not a price rule, says so. The count implies what comes next: when a new five-wave structure arrives, its wave 1 will be starting a larger degree advance, with the whole 100-to-154 rally as the wave beneath it.
The measured-move arithmetic deserves the write-down it gets in practice. A at twenty points projected from B's high at 144 gave a C target of 124, and the market arrived there almost exactly. Targets of this kind are not promises, they are expectations, and expectations are what let a trader size a C-degree long sensibly instead of catching a falling instrument with full weight.
Wave B's thin rally is the false breakout mechanic in relief form: strength that exists because nobody is testing it.
Practical Rules for Corrections
Do not trade wave B with trend expectations. It is a counter-trend rally inside a counter-trend move, the lowest-quality long in the entire framework, and the volume character tells you so in advance: B rises on fading participation by definition. Traders who insist on buying B are the liquidity wave C consumes, and the framework offers them no invalidation that is both close and meaningful.
Size wave C as a measured move from wave A, projected from B's extreme, and treat deep below that projection as information rather than tragedy: a correction that runs past the origin of the entire prior impulse cancels the larger count, and in the worked example that line sits at 100, far below the actual low at 124. Between those boundaries, the corrective count does its quiet work, and the discipline it enforces is the one this whole level has repeated: write the levels first, let structure verify, and let the crowd's mood confirm what price already said.
Correction Questions
How deep do corrections usually go?
The classic band for a correction against the prior impulse is roughly a third to a bit more than half of it, with the fifty percent zone as the statistical center of gravity. The percentage is context, not authority: the structure's own levels, the origin of the impulse, and B's measured projection carry more weight than any single retracement number.
Can a correction take longer than the impulse it corrects?
Yes, and it often does, because corrections fight the larger trend and must ration their progress. A sharp five-week impulse can be answered by a three-month A-B-C. Time symmetry is not part of the structure; when duration stretches far beyond the impulse, the framework starts reading the pattern as complex, a topic the next lesson takes up.
What is the difference between a correction and a reversal?
Degree and destination. A correction answers one impulse and holds above its origin; a reversal ends the larger trend and takes out the structure that defined it. In the worked example, the tell was the 100 origin: above it, the A-B-C was an answer, below it, the whole higher-degree count would have been cancelled. Structure decides, sentiment argues.
Do A-B-C shapes appear inside waves too?
Constantly. Each wave at one degree contains a full structure at the degree below, so a single wave 2 often resolves as a complete three-wave correction of its own. This nesting is why traders count one degree cleanly rather than three degrees loosely, and why the same shapes repeat from multi-year charts down to the hourly ones.
The plain A-B-C is the skeleton, but corrections arrive wearing different clothes, and each outfit changes the B wave's face. The next lesson names the standard shapes, zigzags, flats and triangles, and shows how each one hides or displays the trap.