Accumulation/Distribution Indicator: The Running Total Chaikin Money Flow Was Built to Fix
The Accumulation/Distribution Line scores every bar by where its close landed inside that bar's own high-low range, multiplies the score by volume, and adds the result to a total that never resets. The three-step formula worked through six hand-checkable bars, the trend-confirmation and divergence reads, the gap blind spot that makes the line rise on a day a stock fell 6%, the lineage that produced Chaikin Money Flow, a four-way comparison against OBV, CMF and MFI, and why Williams Accumulation/Distribution is a different indicator entirely.
Most volume indicators ask whether volume showed up. The Accumulation/Distribution Line asks something narrower: inside each bar, did the close finish nearer the high or nearer the low, and how much stock changed hands while that happened. Score the answer between +1 and -1, multiply by the bar's volume, add it to a running total that never resets, and you have the whole indicator. That construction gives it one genuinely useful property and one genuinely strange one, and both fall out of the same three lines of arithmetic. The strange one is why Chaikin Money Flow exists at all.
Quick Answer
The Accumulation/Distribution Line is a running total of volume weighted by where each bar closed inside its own high-low range. Score the close from +1 at the high to -1 at the low, multiply that by the bar's volume, and add the result to the previous total. A rising line reads as accumulation, a falling line as distribution, and only the slope carries meaning, never the number on the axis.
Everything below is the detail that paragraph skips: the exact three-step formula with six bars of arithmetic you can check by hand, the trend-confirmation read, both divergences, the gap blind spot that makes the line rise on a day a stock falls hard, the direct lineage from this indicator to Chaikin Money Flow, and the disambiguation from Williams Accumulation/Distribution, which shares most of a name and almost none of a formula.
What Is the Accumulation/Distribution Indicator?
The Accumulation/Distribution Line, usually shortened to A/D Line or ADL, is a cumulative volume-flow indicator developed by Marc Chaikin and plotted as a single line in a pane beneath price. Its premise is that where a bar closes inside its own range tells you something about who won the session. A close pinned to the high means buyers held the bid into the bell. A close scraping the low means sellers did. Weight that judgment by how much volume traded, and sum the result across every bar on the chart.
Two words in that description do most of the work. Cumulative, because the line is an unbounded running total rather than an oscillator with a ceiling and a floor, so it has no overbought level and no threshold to trade against. And own range, because the comparison happens entirely inside a single bar. The previous bar is not consulted at any point, which sounds like a technicality and turns into the most consequential thing about the indicator further down the page.
Nothing in the formula is specific to equities. It plots on futures, crypto and any instrument with a reported volume figure, on any timeframe, though as with every volume tool the output is only as good as the volume input. Where it fits in a broader chart read is the same place any flow indicator fits, which is a supporting role behind price, and the wider technical analysis overview sets out that ordering better than a single-indicator page can.
How Is the A/D Line Calculated?
Three steps, in this order, every bar. There are no settings.
Step 1, Money Flow Multiplier: MFM = ( ( Close - Low ) - ( High - Close ) ) / ( High - Low )
Step 2, Money Flow Volume:MFV = MFM × volume for that period
Step 3, the line: A/D = previous A/D value + current Money Flow Volume
That three-step construction matches the StockCharts ChartSchool entry on the Accumulation/Distribution Line and Fidelity's technical indicator guide page for Accumulation/Distribution, and unlike most indicators there is nothing to argue about in the setup, because there is no lookback, no smoothing and no multiplier to tune.
The Money Flow Multiplier is the part worth internalising. It runs from +1 to -1 and it is purely a position score. A close exactly at the high returns +1, so the entire bar's volume is added. A close exactly at the low returns -1, so the entire bar's volume is subtracted. A close at the midpoint returns 0, and the bar contributes nothing at all no matter how many shares traded. Anywhere in the upper half of the range is positive, anywhere in the lower half is negative, and the value scales linearly between them.
The Money Flow Multiplier is a position score, nothing else
Here is the arithmetic over six bars. The prices and volumes are invented, chosen so every multiplier lands on a round number you can verify in your head, and the total is anchored at zero before bar 1.
| Bar | High / Low / Close | Multiplier | Volume | Money Flow Volume | Running A/D |
|---|---|---|---|---|---|
| 1 | H 50.00 / L 49.00 / C 49.75 | +0.50 | 1,000,000 | +500,000 | 500,000 |
| 2 | H 51.00 / L 50.00 / C 50.25 | -0.50 | 1,200,000 | -600,000 | -100,000 |
| 3 | H 52.00 / L 50.00 / C 52.00 | +1.00 | 2,000,000 | +2,000,000 | 1,900,000 |
| 4 | H 53.00 / L 51.00 / C 51.00 | -1.00 | 1,500,000 | -1,500,000 | 400,000 |
| 5 | H 48.00 / L 46.00 / C 47.75 | +0.75 | 3,000,000 | +2,250,000 | 2,650,000 |
| 6 | H 48.50 / L 47.50 / C 48.00 | 0.00 | 900,000 | 0 | 2,650,000 |
Work bar 1 through by hand and the pattern is clear. Close 49.75 minus low 49.00 is 0.75. High 50.00 minus close 49.75 is 0.25. Subtract, get 0.50, divide by the 1.00 range, and the multiplier is +0.50, so half the million shares get added. Bar 3 closed exactly at its high, so the multiplier is +1.00 and all two million shares count as accumulation. Bar 4 closed exactly at its low for a multiplier of -1.00 and the mirror result. Bar 6 is the quiet one worth noting: 900,000 shares traded, the close landed dead on the midpoint of the range, and the total did not move by a single unit.
Bar 5 is where this post actually starts. Hold it for a moment, because it needs its own section.
How Do You Read an A/D Chart?
Start with the rule that saves the most confusion: the number on the axis means nothing. The total began wherever your platform started counting, which is a function of loaded history rather than anything about the stock. Two charts of the same symbol with different amounts of history produce different A/D values forever, with identical shapes over the overlapping stretch. Comparing A/D levels between two stocks is not a weak read, it is not a read at all.
What you read is slope and structure. A rising A/D Line says money flow volume has been net positive over that stretch, which is the conventional accumulation read: bars have tended to finish nearer their highs, weighted by how much traded. A falling line says the opposite, and gets called distribution. Both StockCharts and Fidelity frame it that way, and the useful version of the read is the comparison against price rather than the direction on its own.
Trend confirmation is the base case. Price making higher highs and higher lows while the A/D Line makes higher highs and higher lows is agreement, and the conventional interpretation is that the uptrend is likely to continue. Invert everything for a downtrend: lower price lows alongside lower A/D lows say the selling has participation behind it. Because the A/D Line is a line like any other, you can treat its peaks and troughs as structure and compare them to the levels price is actually working against rather than reading the pane in isolation.
Confirmation is a weaker claim than it sounds, and it is worth saying so. In a rising market more bars close in the upper half of their range, so a rising A/D Line in an uptrend is close to mechanical a lot of the time. The information sits in the cases where the mechanical expectation fails, which is the whole reason divergence gets the attention it does, and it is the same reason volume is more useful as a disagreement check than as a confirmation stamp on a momentum chart.
What Do A/D Divergences Signal?
A divergence is price and the A/D Line making opposite statements about the same swing. There are two of them and they are mirror images.
Bearish divergence
Price pushes to a new high. The A/D Line fails to make a new high, or turns down outright. Fidelity's guide puts the read plainly: when price continues to make higher peaks and Accumulation/Distribution fails to make higher peaks, the uptrend is likely to stall or fail. The mechanical translation is that the second push was made on bars closing lower in their ranges, or on less volume behind the strong closes, or both. It signals a weakening advance, not a completed top.
Bullish divergence
Price drops to a new low. The A/D Line does not confirm it, making a higher low or continuing to climb. Same source, mirrored: when price continues to make lower troughs and Accumulation/Distribution fails to make lower troughs, the downtrend is likely to stall or fail. The read is that the second flush happened with closes landing higher in their ranges than the first one did, so the selling pressure driving price down is easing.
Two honest qualifications, because the examples in most write-ups get picked after the fact. Divergence is not a timing tool. Nothing in the construction says how long a disagreement is allowed to run, and plenty of them resolve by the A/D Line catching up to price rather than by price turning. And a divergence on its own is one signal, which is why every serious treatment adds a trigger: a break of structure, a trendline break, a moving average cross, something that fires. Deciding how many of those you require before you act is the subject of how much confluence a setup needs, and a single volume-flow line is nowhere near a system.
The A/D Line diverged and you want to take the other side. That is a reason to look, not a graded trade.
Upload the screenshot and SnapPChart reads that single image against a fixed rubric, then returns a setup grade, an entry, a stop with the reasoning behind the level, targets, and the reward-to-risk those levels imply. Arguing with the grade is cheaper than arguing with the fill.
Grade this chartThe Gap the A/D Line Cannot See
Back to bar 5. The stock closed the previous session at 51.00. The next day it opened at 47.00, traded a range of 46.00 to 48.00, and closed at 47.75 on three million shares. That is a loss of 3.25 points, roughly 6.4%, on heavy volume. Anyone looking at the price chart would call that distribution without hesitating.
The A/D Line added 2,250,000 to the total.
Nothing malfunctioned. The multiplier only ever looks at the close relative to that bar's own high and low, and 47.75 sits three quarters of the way up a 46.00 to 48.00 range, so the multiplier is +0.75 and three quarters of the volume gets counted as accumulation. The previous close of 51.00 appears nowhere in the formula. Neither does the open. Fidelity states the consequence directly: the A/D Line ignores the change from one period to the next, so a security could gap down and close significantly lower while the Accumulation/Distribution Line rises, as long as the close finished above the midpoint of the high-low range. StockCharts makes the same point about the formula completely ignoring the period-to-period change.
Price fell 6.4%. The line went up.
This is not an edge case to route around. It is the definition, and it fires hardest on exactly the days that matter most: earnings gaps, guidance cuts, halt reopens, anything that moves a stock between sessions rather than during one. Those are also the highest-volume bars on the chart, so the largest single contributions to the running total are frequently the ones most likely to score the opposite of what happened to the stock that day.
The practical handling is to know which question you asked. The A/D Line answers "inside each session, did closes finish near highs or near lows, weighted by size." It does not answer "is this stock going up." On a chart with few gaps those two questions produce similar-looking answers and the distinction is academic. On a gap-heavy chart they come apart badly, and the indicator that does account for the overnight move is On-Balance Volume, which tests each close against the previous close instead and therefore treats that same session as a full three million shares of selling. Neither one is broken. They are answering different questions, and on gap days they will tell you opposite things.
Why Chaikin Money Flow Exists
Marc Chaikin built both indicators, and the relationship between them is the cleanest piece of lineage in the volume-flow family. Chaikin Money Flow uses the identical Money Flow Multiplier, multiplied by the identical volume figure, to produce the identical Money Flow Volume value for each bar. Steps 1 and 2 are the same arithmetic on both indicators. Everything that separates them happens in step 3.
The A/D Line adds each bar's money flow volume to an unbounded total that never resets. Chaikin Money Flow sums money flow volume over a rolling window, conventionally 20 or 21 periods because a trading month is roughly 21 sessions, then divides by total volume over that same window. Cumulative forever against rolling window. That is the entire differentiating axis between the two.
What the window buys is forgetting. A cumulative total has no mechanism for deciding that a bar is old, so a gigantic earnings bar from two years ago still sits in the A/D Line's current value at full weight, and the slope you measure across the last three months is being drawn on top of every bar that preceded it. A rolling sum drops that bar off the back of the window and the indicator stops carrying it. That also means Chaikin Money Flow can step on a quiet day purely because an old high-volume bar aged out of the lookback, which is the cost of the same design choice. Dividing by window volume buys the second property: a bounded -1 to +1 output where zero is a real dividing line, so unlike the A/D Line the number itself is readable.
Both inherit the gap blind spot, because both are built on a multiplier that never looks at the previous close. Windowing fixes the memory problem. It does not fix the thing bar 5 demonstrated.
The Four Volume-Flow Indicators, Side by Side
Four indicators get described with roughly the same sentence about money flowing into and out of a stock, and the sentence is useless for telling them apart. The mechanics do it in one table.
| Mechanic | OBV | A/D Line | CMF | MFI |
|---|---|---|---|---|
| What each bar is scored against | The previous bar's close | Its own high and low range | Its own high and low range | Typical price vs the previous bar's typical price |
| How much of the bar's volume is applied | All of it, in one direction or the other | A fraction, set by the multiplier between -1 and +1 | A fraction, set by the same multiplier | All of it, after multiplying by typical price |
| Does it ever forget old bars? | No. Cumulative from the first loaded bar | No. Cumulative from the first loaded bar | Yes. A rolling window, conventionally 20 or 21 | Yes. A rolling window, conventionally 14 |
| Bounded output? | No, unbounded running total | No, unbounded running total | Yes, -1 to +1 by construction | Yes, 0 to 100 by construction |
| Does the absolute number mean anything? | No, the anchor is arbitrary | No, the anchor is arbitrary | Yes, zero is a real dividing line | Yes, 80 and 20 are the conventional bands |
| Is the price level in the formula? | No, only the direction of the close | No, only the close's position in the range | No, only the close's position in the range | Yes, flow is scaled by typical price |
| Can it see a gap? | Yes, the gap is inside the close-to-close test | No, the prior close is not an input | No, the prior close is not an input | Yes, typical price is compared period to period |
| Where one huge old bar sits | Permanently in the total | Permanently in the total | Drops out when the window rolls past it | Drops out when the window rolls past it |
Read down the first two rows and the family splits cleanly. OBV applies every bar's entire volume with no range weighting at all, classified by a single close-versus-prior-close test, so a bar that closed a penny up and a bar that closed four dollars up contribute identically in kind. The A/D Line and Chaikin Money Flow both weight by where the close sat inside the bar, using the same multiplier, and split only on whether the sum is windowed. And the Money Flow Index bounds its result into a 0 to 100 oscillator, built on typical price times volume rather than range position, which additionally makes it the only one of the four where the actual price level is in the formula.
Put the A/D Line's position in one sentence: it weights by range position the way CMF and MFI do, and it never forgets the way OBV never forgets. It is the range-weighted cumulative corner of the family, and that combination is both its distinguishing feature and the source of both its problems. That is the whole family. Four indicators, four weighting decisions, and no fifth variant hiding behind them.
Williams Accumulation/Distribution Is a Different Indicator
If your platform lists something called Williams Accumulation/Distribution, or WAD, that is not the indicator this page is about. Larry Williams built a separate tool that happens to share the accumulation and distribution language, and the construction has almost nothing in common with Chaikin's. Williams A/D compares today's close against yesterday's close, then measures the distance from the close to a true range boundary that itself incorporates yesterday's close, and accumulates those price distances into a running total. As the TA-Lib function reference for Williams' Accumulation/Distribution shows, the standard implementation takes high, low and close as its inputs, with no volume term in the calculation at all.
The contrast is almost exactly inverted. Chaikin's A/D Line weights by volume and never looks at the previous close. Williams' version hinges on the previous close and never looks at volume. Two lines that will diverge from each other constantly, sitting one row apart in your indicator menu under nearly the same name. When a tutorial quotes a formula that does not include the term for volume, you are reading about the Williams version, and everything on this page stops applying.
Where the A/D Line Breaks
The A/D Line is a lagging indicator by construction. Every value is built from bars that have already closed, so it describes what participation has already done and contains no predictive term anywhere. It is also not a standalone signal. A line saying closes have tended to finish high in their ranges has no view on trend, level, structure or risk, which is why every reasonable treatment pairs it with price analysis rather than trading it directly. None of the conditions below are faults in the formula. They are what a range-weighted cumulative total does when the inputs stop behaving like the ones it assumed.
| Condition | What the A/D Line does | Why it misleads |
|---|---|---|
| A hard gap down that closes strong | Adds most of the day's volume, because the close sat high in that day's own range | The line rises on a session where the stock lost real money. Mechanically correct, descriptively backwards, and the single most common reason people decide the indicator is broken |
| Years of loaded history | Carries every money flow volume figure since the first bar, forever | A slope drawn across the last three months competes with a total built over three years. The old history is not weighted down, it is simply still there |
| An earnings or news volume spike | Applies a multiple of a normal day's money flow volume in one step | A vertical step that will still be visible on the chart months later. Give the line several sessions before reading a new slope across it |
| A doji or a locked bar where high equals low | Divides by zero in the multiplier, so the term has no defined value | Whatever your platform does with that bar is a handling decision, not a reading. It is worth knowing which convention your software picked |
| A very narrow intraday range | Produces multipliers near +1 or -1 off a few cents of movement | Full-strength accumulation and distribution readings off bars where nothing happened. The multiplier has no sense of scale |
| A thin, low-volume name | Runs exactly as specified on bars where a handful of prints set the close | The multiplier is decided by whoever traded last in the final minute, and the cumulative total dresses that noise up as a smooth trend |
| Spot forex | Multiplies the range score by whatever the platform labels volume, usually tick count | No consolidated tape means no authoritative volume figure to weight with, so two brokers produce two different A/D lines for the same pair |
| A divergence that runs for months | Keeps disagreeing with price without any resolution arriving | Nothing in the construction sets a shelf life on a divergence. A long one is not stronger evidence, and plenty close by the A/D Line catching up rather than by price turning |
The narrow-range row is the one that catches intraday traders. The multiplier is a ratio, so it has no idea whether the range it divided by was four dollars or four cents. A 1-minute bar that traded between 40.02 and 40.06 and closed at 40.06 scores a perfect +1.00, identical to a session that ran four points and closed on its high. Both get their full volume added as accumulation. On a daily chart that rarely matters. On fast intraday tape it means the indicator is reporting maximum conviction off noise, which is a reason to prefer the higher timeframe version of the read and to check what the indicators actually worth having on an intraday chart are doing before you trust it.
What a screenshot read can and cannot see here
Worth being straight about, since this site sells a tool. SnapPChart has no Accumulation/Distribution field. It does not compute a Money Flow Multiplier, it does not accumulate money flow volume, and it carries no state for an A/D value the way it does for market structure and trend, the moving average stack, the VWAP relationship, MACD, support and resistance, and volume behaviour, which it reads qualitatively from the image rather than as a computed series. What it does is grade a chart screenshot you upload. So if you plot the A/D Line in a lower pane on your own platform before taking that screenshot, that line is part of what the analysis sees, as a shape drawn beneath price. That is a picture of a cumulative total, not a verified calculation: it cannot tell you the step across the pane was one earnings gap scoring backwards, it cannot know how much history your platform anchored the total on, and it cannot check whether the volume figure underneath it was shares or tick count. Which chart states a screenshot read genuinely carries and which it only infers from shape is the subject of the wider guide to how AI reads a chart, and a neutral description of what a single chart read covers sits on the AI chart analysis page. If the A/D Line is your reason for the trade, validating it stays your job.
Score each bar from +1 to -1 on where its close sat inside its own high-low range, multiply by volume, add to a total that never resets. Rising is accumulation, falling is distribution, and the number on the axis is an artefact of where counting started. Confirmation when price and the line agree, divergence when they do not, and a divergence needs a separate trigger before it is worth anything. The one thing to remember above all: the formula has no term for the previous close, so a stock can gap down 6% and still print a rising A/D Line if it closed in the upper half of that day's range.
Frequently Asked Questions
What is a good Accumulation/Distribution Line value?
There is no such number, and asking for one is the fastest way to misread the indicator. The A/D Line is a cumulative total that began at whatever bar your platform started counting from, so a reading of 8.4 million on one chart and 260,000 on another say nothing about which stock is under more accumulation. Load two years of history and you get one number. Load six months of the same symbol and you get a completely different one, with an identical shape over the overlapping stretch. There is no overbought level, no oversold level, and no threshold anybody can quote you. Slope and the line's own sequence of highs and lows are the entire readable surface.
What period setting should I use for the A/D Line?
It does not have one. That is not a defaults question people are being lazy about, it is a structural fact: the A/D Line sums every bar since the chart began, so there is no lookback to set and no settings dialog worth opening. The confusion usually comes from Chaikin Money Flow, which does take a period and ships with 20 or 21 depending on the platform. If your charting software is offering you a length field on something labelled Accumulation/Distribution, check whether you actually selected the Chaikin Oscillator or a CMF variant, because the plain A/D Line takes no arguments at all.
Is the Accumulation/Distribution Line the same as the Chaikin Oscillator?
No, though one is built out of the other. The Chaikin Oscillator is a MACD applied to the A/D Line: a 3-period EMA of the A/D Line minus a 10-period EMA of it, plotted around a zero line. So the oscillator is measuring the momentum of the cumulative total rather than the total itself, which makes it a faster and noisier read on the same underlying money flow volume series. A rising A/D Line and a Chaikin Oscillator below zero are not contradicting each other. One says the total is higher than it was; the other says the total is rising more slowly than it recently did.
Why do two platforms plot different A/D lines for the same stock?
Usually the anchor. The total starts wherever loaded history starts, so any difference in how many bars each platform pulled produces a different absolute value forever after. Extended-hours data is the second cause: if one feed folds pre-market and after-hours prints into the daily bar and the other does not, the high, the low, the close and the volume are all different inputs, so the multiplier and the money flow volume both change. Split and dividend adjustment of the historical series is the third. Compare the shape of the last few months rather than the numbers, because the numbers were never designed to agree.
Does the A/D Line work on a 5-minute chart?
The arithmetic runs unchanged and the line plots fine, but two things shift. Intraday volume is front and back loaded, so the opening and closing ranges contribute far more money flow volume than the middle of the session does, and an intraday A/D slope that looks like steady accumulation is often three bars near the bell carrying the whole thing. And the multiplier gets jumpy on small ranges: a 5-minute bar with a four-cent high-low range can produce a multiplier of +1.00 on a move nobody would call buying pressure, then flip to -1.00 on the next bar. Read the intraday version as a within-session participation tally, not as a compressed month of accumulation.
This article is for educational and informational purposes only and is not investment, financial or trading advice. The three-step Accumulation/Distribution formula, the Money Flow Multiplier definition and its +1 to -1 bounds, the attribution of the indicator to Marc Chaikin, the accumulation and distribution interpretation of a rising or falling line, the trend-confirmation and divergence conventions, and the stated limitation that the line ignores the change from one period to the next are the conventional published accounts reproduced by charting platforms and reference sources, including the StockCharts ChartSchool and Fidelity pages linked above. The Williams Accumulation/Distribution description reflects the standard published implementation referenced above and is included only to distinguish that indicator from the one this page covers. The six-bar worked example, including every high, low, close, volume figure, multiplier and running total in it, is an illustrative construction built so the arithmetic can be checked by hand; it is not a real security, a real trading session or a market observation, and the gap scenario described around bar 5 is a hypothetical illustration rather than a recorded event. Nothing here is a backtest of my own, no rule set or indicator pairing described is claimed to be profitable, and no edge is claimed or implied. Indicator readings describe what price and volume have already done and do not predict what they will do next. Day trading and active trading carry a substantial risk of loss and are not suitable for every investor. SnapPChart grades a static chart screenshot you upload and returns a setup grade, entry, stop, targets and reasoning for that single image; it does not calculate an Accumulation/Distribution Line, a Money Flow Multiplier, money flow volume, Chaikin Money Flow, a Money Flow Index value, On-Balance Volume or any other indicator value itself, does not track indicator states it has not been shown, does not scan the market, and does not track your account, positions or P&L. It can only account for indicators that are visibly drawn on the image you upload. Do your own analysis, size positions so that being wrong is survivable, and consider speaking to a licensed financial professional about your own circumstances before trading.
Writes about AI-assisted day trading, technical analysis, and the systems traders actually use to stay disciplined.
A rising A/D Line is one line agreeing with you. The trade is still ungraded.
Net buying pressure over the last stretch of bars is not an entry, a stop, or a reason the reward covers the risk. Upload the screenshot and SnapPChart reads that one image against a fixed rubric, then returns a setup grade, an entry, a stop with the reasoning behind the level, targets, and the reward-to-risk those levels imply. One skipped C-grade setup covers the cost.