Wyckoff Accumulation and Distribution: Which One Is It?
Wyckoff accumulation and distribution are mirror images. What each phase and event means, and why you cannot tell them apart before Phase C.
Wyckoff accumulation is a sideways trading range after a decline, which Richard Wyckoff read as large operators absorbing supply before a markup. Wyckoff distribution is the same range shape after an advance, with those operators unloading before a markdown. The two are mirror images, and that is the practical problem: while a range is still forming, the structure you can see is close to identical in both, and the first evidence that separates them does not arrive until Phase C or later.
So the useful answer to "which one is this?" is an order in which the evidence arrives, plus a way to grade your own labels afterwards.
Wyckoff accumulation and distribution are market-structure phases, and they have no formula. The Accumulation/Distribution indicator, usually shortened to A/D, is a different object: a volume-based line with a published formula, named after Marc Chaikin.
What is Wyckoff accumulation?
A trading range that forms after a sustained decline, in which Wyckoff held that large operators were buying from a public that had given up. The range is not drawn by eye: it is defined by events.
The canonical modern write-up of the method is the ChartSchool tutorial at StockCharts, written by Roman Bogomazov and Michael Lipsett, with schematics Bogomazov drew and the late Henry "Hank" Pruden edited. Its sentence on the boundaries is the one most explanations skip: "The lows of the SC and the ST and the high of the AR set the boundaries of the TR." TR is the trading range, and SC, ST and AR are events defined further down.
Wyckoff's framing device for the buyer is the "Composite Man", a single imagined operator standing in for all the large interests, used so that a chart can be read as one actor's campaign rather than a crowd.
That tutorial is written about stocks, and it states its own scope: "Although this article focuses exclusively on stocks, Wyckoff's methods can be applied to any freely-traded market in which large institutional traders operate, including commodities, bonds and currencies."
What is Wyckoff distribution?
The same range shape after a sustained advance, in which the large operators are selling into public demand before a markdown. It has the same five phases, the same length and the same amount of structure as accumulation.
Explanations of the method tend to shrink the distribution half or drop it. Distribution is where a reader who learned accumulation first gets caught.
Wyckoff accumulation vs distribution: how do you tell which one you are in?
While the range is still forming, usually you cannot, and the sources that teach the schematic say so themselves. What you get instead is a ladder of evidence that arrives in a fixed order.
Rung one is the trend into the range. It is available immediately and uses price only, so write it down first. Financial Tech Wiz puts it this way: "The only thing that separates them is context: a range that forms after a long advance is far more likely to be distribution or re-accumulation than fresh accumulation, and the direction of the trend into the range is your first piece of evidence about which one you are looking at." CMC Markets names the same factor, calling it "The broader trend context, relative volume patterns and the character of the range boundaries". Rung one does not settle it: a range that forms after an advance still leaves two candidates, distribution or re-accumulation.
Rung two is Phase B, where the two halves are hardest to separate. Capital.com's accumulation Phase B reads "Phase B: price moves sideways. This is the cause phase. The composite man buys secretly keeping volume low." Its distribution Phase B reads "Phase B: price moves sideways. This is the cause phase for the distribution part of the cycle. The composite man sells slowly here." The observable half is word for word identical, and the half that differs names who is doing what, which is the thing the reader is trying to work out. Margex mirrors its two Phase B sentences the same way, "Phase B serves as the "cause" for a new uptrend and is where the bulk of Wyckoff accumulation takes place from large players and institutional investors" against the same sentence with downtrend and distribution swapped in. City Traders Imperium states it outright: "Phase B remains the same as in accumulation, to build a cause, this time, however, in preparation for a new downtrend." So does the canonical tutorial, in its own distribution section: "The points about Phase B in distribution are similar to those made for Phase B in accumulation, except that the large interests are net sellers of shares as the TR evolves". Financial Tech Wiz adds: "Phase B looks like a completed range to an impatient eye."
Rung three is that the Phase B differences which do exist are volume readings, not shapes. The canonical tutorial says that in accumulation "the volume on downswings within the TR tends to diminish", while in distribution "SOWs are usually accompanied by significantly increased spread and volume to the downside". LuxAlgo gives accumulation's Phase B a positive signature in the same terms, "declines that progressively shrink in volume and spread while rallies hold more of their ground". Both are volume comparisons, which matters if your instrument has no reliable volume figure.
Rung four is the direction of the Phase C test, the first difference you can see in price alone. Capital.com states both sides in a line each: "Phase C: the price drops below the range." against "Phase C: the price rises above the range." A spring undercuts support and recovers; an upthrust pushes above resistance and fails.
Rung five is resolution: which boundary finally breaks, and on what spread. FXIFY reduces it to two lines. Accumulation is where "Any drop below the range comes back inside, and price rises to the creek", and redistribution is where "A drop below the range holds, and rallies fail below the creek".
Both of the no-event variants of the schematic have no Phase C event at all, so if you are in one of those, your first structural evidence is the Phase D or Phase E break, which is rung five.
LuxAlgo's own FAQ asks how you tell accumulation from distribution while the range is still forming and answers: "Often you cannot with confidence, which the Wyckoff literature itself concedes." It goes on: "The differential evidence arrives late: phase C and D behavior such as a spring that holds versus an upthrust that fails, rallies gaining volume versus losing it, and which boundary finally breaks on expanding spread. Until then, labels are working hypotheses." Its page on the distribution mirror puts it shorter: "Mid-range the two can look nearly identical; resolution behavior separates them."
What are the Wyckoff phases, A to E?
Five phases, the same five in both halves. Phase A stops the old trend, Phase B builds the cause, Phase C tests, Phase D confirms, Phase E leaves the range.
| Phase | Accumulation | Distribution | What you can see | What is inferred |
|---|---|---|---|---|
| A | The decline stops. PS, SC, AR and ST appear, and three of them set the range boundaries | The advance stops. PSY, BC, AR and ST appear | Wide bars, a climactic low or high, a fast reaction, a retest that holds | That large operators have started a campaign |
| B | Builds the cause for a markup | Builds the cause for a markdown | Sideways swings between the boundaries, usually several retests | Who is absorbing and who is distributing |
| C | The test of remaining supply, often a spring below support | The test of remaining demand, a UT or UTAD above resistance | Which boundary gets broken and whether price recovers back inside | Whether supply or demand is exhausted |
| D | Demand dominant. SOS advances, LPS pullbacks. "During Phase D, the price will move at least to the top of the TR" | Supply dominant. SOW declines, LPSY rallies. "During Phase D, price travels to or through TR support" | A sequence of higher lows, or of lower highs | That the range is ending |
| E | "In Phase E, the stock leaves the TR" and markup runs | Price leaves the range and markdown runs | Price outside the range, pullbacks that hold | Nothing much. By here the label is settled |
The quoted cells are the canonical tutorial's own wording. Phase D and Phase E are distinct in that source, and the distinction is practical: in Phase D price is still at the edge of the range, in Phase E it has left. Capital.com moves the break forward into Phase D, writing "Phase D: the markup begins. Price breaks resistance with high volume." and then leaving Phase E as "Phase E: the price is in a clear uptrend."
What are the Wyckoff events, and what do the letters mean?
Nine abbreviations on the accumulation side and eight on the distribution side. The definitions follow the canonical ChartSchool tutorial.
Accumulation:
- PS, preliminary support. The first substantial buying after a prolonged down-move. Volume increases and the price spread widens. It is early, and on its own it does not stop the decline.
- SC, selling climax. The point where selling pressure climaxes and heavy or panicky public selling is absorbed by larger interests near a bottom. Price often closes well off the low.
- AR, automatic rally. Selling pressure has drained, so a wave of buying pushes price up easily, helped by short covering. The high of this rally defines the top of the range.
- ST, secondary test. Price revisits the SC area to test the balance of supply and demand. On a confirming test, volume and spread are significantly smaller. There are usually several.
- Spring, or shakeout. The canonical definition: "a spring is a price move below the support level of the TR (established in Phases A and B) that quickly reverses and moves back into the TR". The same page explains why it works on people: "It is an example of a bear trap because the drop below support appears to signal resumption of the downtrend."
- Test. A follow-up probe for supply. A successful one typically makes a higher low on lower volume.
- SOS, sign of strength. A price advance on a widening spread and relatively higher volume, often just after a spring.
- LPS, last point of support. The low of the pullback after an SOS, on reduced spread and volume. There can be more than one, despite the name.
- BU, back-up. The pullback to the old top of the range. The canonical source gives it a named author: BU is "short-hand for a colorful metaphor coined by Robert Evans, one of the leading teachers of the Wyckoff method from the 1930s to the 1960s", who "analogized the SOS to a "jump across the creek" of price resistance".
FXIFY builds the creek vocabulary out in full: "The top of the range is called the creek. It is the resistance line set by the AR high", the break above it is where "this is called a jump across the creek (JAC)", and BU is "A pullback after the SOS to the old top of the range. It is one kind of LPS."
Distribution:
- PSY, preliminary supply. Large interests begin unloading in quantity after a pronounced up-move. Volume expands and spread widens.
- BC, buying climax. Urgent public buying is filled by professional interests near a top, usually on marked increases in volume and spread.
- AR, automatic reaction. The selloff after the BC. Its low defines the bottom of the range.
- ST, secondary test. Price revisits the BC area to test demand. It may take the form of an upthrust.
- UT, upthrust. The canonical source states it as the exact inverse: "a UT is the opposite of a spring. It is a price move above TR resistance that quickly reverses and closes in the TR."
- UTAD, upthrust after distribution. The counterpart to the spring and terminal shakeout. It comes late in the range and tests new demand after a break above resistance.
- SOW, sign of weakness. A down-move to or slightly past the bottom of the range, usually on increased spread and volume.
- LPSY, last point of supply. After a SOW, a feeble rally on narrow spread that cannot advance. These are the last waves of distribution before markdown.
AR and ST are the same two abbreviations in both halves, with the same letters and a different meaning: automatic rally and automatic reaction.
Wyckoff accumulation schematic 1, 2 and 3: which is which?
There is no schematic 3. The canonical source publishes four diagrams in total: "These concepts are illustrated in the following four schematics; two depicting common variants of accumulation TRs, followed by two examples of distribution TRs."
What the numbers mean, in that source, is whether the Phase C event is present. For accumulation, "springs and terminal shakeouts are not required elements: Accumulation Schematic 1 depicts a spring, while Accumulation Schematic 2 shows a TR without a spring". For distribution, "a UTAD is not a required structural element: the TR in Distribution Schematic #1 contains a UTAD, while the TR in Distribution Schematic #2 does not".
"Schematic 2" names two different objects, because sources count on different bases. Capital.com counts kinds rather than diagrams: "There are 2 main types" of schematic, "1 each for Wyckoff accumulation and distribution", and "Each schematic has 5 phases, A to E." On that basis, type 2 means distribution. On the canonical page and on CMC Markets, schematic 2 means the variant with no Phase C event, and CMC calls it exactly that, "This is sometimes called a Type 2 accumulation schematic." Neither is a typo. If you quote a schematic number to another trader, say which basis you mean.
The numbered diagrams are not Wyckoff's own. LuxAlgo states that "the tidy diagrams numbered #1 and #2 were formalized by later teachers of the course and popularized in modern Wyckoff texts", and the canonical page credits its schematics to Bogomazov and Pruden rather than to Wyckoff. The vocabulary descends from the course, and the pictures are teaching aids built afterwards.
Is the spring required?
No. The schematic 2 variants have no spring and no UTAD, and the ranges in them resolve anyway. Most explanations of the method say so: CMC Markets, "Not all accumulation ranges include a spring"; Margex, "A spring is not required as part of Wyckoff accumulation theory"; Phemex, for the mirror, "UTAD is optional"; TrendSpider, which gives the variant its own heading, "Schematic 2 (no Spring)". FXIFY says what replaces it: "In Schematic 2, there is no spring, and a higher low inside the range acts as the test."
No spring means no Phase C direction to read, so your first structural evidence of which half you are in is pushed out to the Phase D break.
Wyckoff accumulation in an uptrend: re-accumulation, and why it is the hard case
A range that forms inside an existing uptrend is re-accumulation, a pause before the next leg up, and it is the hardest case here because the evidence that normally labels a range is the evidence that goes missing.
The canonical source has the specific answer: "In a re-accumulation TR (which occurs during a longer-term uptrend), the points representing PS, SC and ST are not evident in Phase A. Rather, in such cases, Phase A resembles that more typically seen in distribution (see below)." It adds that "Phases B-E generally have a shorter duration and smaller amplitude than, but are ultimately similar to, those in the primary accumulation base."
Read that against the boundary rule, where the SC and ST lows and the AR high are what fix the range edges. In re-accumulation, two of those three events are not there. So you lose the range boundaries as events, Phase A now looks like the opposite schematic, and rung one of the ladder points the wrong way: the range sits inside an uptrend, which is exactly where distribution forms.
The mirror case is in the same source: "In a redistribution TR within a larger downtrend, Phase A may look more like the start of an accumulation TR (e.g., with climactic price and volume action to the downside)."
City Traders Imperium names the error without resolving it: "Reaccumulation is the continuation of a sustained upward move, allowing the big players to open more positions" and "This is where traders mistake Reaccumulation for Accumulation." CMC Markets reaches it as a question, "The danger is misreading reaccumulation as distribution (or redistribution as accumulation)", and its answer is the three unquantified factors quoted in rung one. Financial Tech Wiz is blunter: these ranges "look identical to accumulation and distribution on a naked chart", and context is all you get.
So in re-accumulation you are missing the Phase A events that would normally draw the range, your trend context points at the wrong answer, and what is left is the Phase B volume comparison and the resolution. That is a thin evidence base, and a good reason to write the call down before the range resolves rather than after.
What does Wyckoff accumulation look like on a chart?
A flat band of sideways price after a drop, with a sharp spike low near its left edge, a fast rally that sets the top, one or more retests of the low that hold, and then higher lows pressing the top of the band.
Resist the urge to treat that as a shape to match. LuxAlgo says: "The schematic is a sequence and a vocabulary, not a template price must obey: real ranges skip, repeat, and blur events, and labels like SC or LPS are provisional until markup confirms them." Velotrade says the same: "Do not force the schematic. Not every range is accumulation."
A spring and a liquidity sweep describe a similar move, and the difference is that Wyckoff's spring sits at a defined place in a labelled range rather than anywhere a level gets taken out. The "smart money" idea in SMC and ICT traces back to Wyckoff's Composite Man.
How would you test your own Wyckoff reading, bar by bar?
Label ranges on past data with the right edge hidden, write the call down before you step forward, then step and score it. A label written after the resolution is not a test of anything.
Financial Tech Wiz gives the instruction in full: "Log the phase you called, the event you saw, and the level that would prove you wrong, then come back after the range resolves and find out whether your labels were actually predictive." It adds: "A Wyckoff read is a hypothesis with an invalidation level, not a prediction. A spring that closes back below the range low is a failed spring, and that is the moment to be out."
What to record, per range, before you step forward:
- The range high and the range low, and which events you used to set each.
- The phase you are calling, and the date of the bar you called it on.
- The event you think you just saw, by abbreviation.
- The level that would prove the call wrong.
- Which half you are calling: accumulation, distribution, re-accumulation or redistribution.
Then step forward and record what happened: which boundary broke, how far price travelled, and whether the event you named held its definition.
ChartLabs replays historical candles one at a time, so the call goes down before you can see how the range resolved. Chart text notes, a Notes column and a note button in the Orders panel, notes in the CSV export, and keyboard shortcuts for the trend line, horizontal line, horizontal ray, vertical line and Fibonacci retracement all help with the logging, and charts reopen on the period you last used per symbol if you are reading one range on two timeframes.
Our markets are forex, indices, commodities including gold and oil, and crypto. Historical data comes from Dukascopy at tick level where available, cleaned to remove anomalies and built into 1-minute candles, reaching back up to around 22 years on the oldest forex pairs, with other instruments starting later.
How many labelled ranges is enough to tell you anything is a separate question. For the general version of the loop, our beginner's guide to backtesting a strategy walks through it, forex backtesting covers the data and session issues that change results, and manual backtesting builds better traders is the argument for doing it by hand rather than reading a report.
What you cannot check this way, and what you would need
The volume half. Every Wyckoff event is defined partly by volume, so a price-only replay can only grade the price-structure half of your reading.
Financial Tech Wiz states the constraint: "Every Wyckoff event is defined by a volume characteristic. A spring without the volume signature is a support break that happened to bounce. If your platform does not show volume for that instrument, use a different method." LuxAlgo adds the instrument-specific problem: "spot forex has no centralized volume, so practitioners substitute tick volume or currency-futures volume". Before you lean on a volume reading, check what your own chart actually shows for the instrument in front of you, and what that figure is counting.
What a replay can check is the price-structure half: where the boundaries were, which phase you called and when, which level would have proved you wrong, which boundary actually broke, and how far price went afterwards. That is a narrower test than the full method.
Start checking your own labels
Pick a market you already watch, find three ranges in its history, and label each one with the right edge hidden before you step forward.
ChartLabs is one plan at $20 a month or $200 a year, with a 7-day free trial and no card needed to start. Start the free trial and step through your first range.
Related Topics
Related
Backtesting a Trading Strategy: A Beginner's Guide
What backtesting is, why it matters, and how to run one step by step: defining your rules, choosing clean historical data, stepping through the chart trade by trade, logging results, and reading the metrics that tell you whether your strategy has a real edge.
Forex Backtesting: Test Your Strategy on Real Data
A practical guide to backtesting forex strategies: why session dynamics, spread variation, and pair-specific behavior change your results, what clean historical data actually means, and a five-step process for testing a strategy before you risk real capital.
Manual Backtesting Builds Better Traders
Stepping through historical charts by hand teaches what an automated backtest report cannot: pattern recognition, what a normal drawdown feels like, and the discipline to follow your own rules when it is uncomfortable.