Anchored VWAP: The Anchor Is the Whole Tool

Michael Whitman · Trading since 2017, full-time since 2020 · Reviewed 2026-09-06

Anchored VWAP is a volume-weighted average price calculated from a starting bar the user selects, rather than resetting each session. The choice of anchor determines the line completely, which makes it a tool for answering a specific question rather than a general indicator.

How it works

A candlestick chart of the site's shared price history. The headline on the chart reads: A volume-weighted average from a point you choose. — A volume-weighted average from a point you choose. Illustrative chart - not real market data.

Take every bar since a chosen starting point. Multiply each bar’s typical price by its volume, add those up, and divide by total volume. That is the average price at which everything since the anchor actually traded.

A calmly advancing stretch of the long price series. The headline on the chart reads: The ordinary version anchors to the session open and resets. — The ordinary version anchors to the session open and resets. Illustrative chart - not real market data.

Ordinary VWAP anchors to the session open and resets every day. The anchored version does not reset — you place the anchor wherever you want and it accumulates from there.

A gently rising stretch of the long price series. The headline on the chart reads: The anchor is the whole tool, and it is your decision. — The anchor is the whole tool, and it is your decision. Illustrative chart - not real market data.

Which makes the anchor the entire tool. The formula is fixed; the only input you supply is where to start, and that choice determines the line completely.

The question it can answer

A strongly rising stretch of the long price series. The headline on the chart reads: Anchoring to an event asks who is up since that event. — Anchoring to an event asks who is up since that event. Illustrative chart - not real market data.

Anchor to an earnings gap, and the line is the average price paid by everyone who has traded since that event. Price above it means the average participant since then is ahead; below it means they are underwater.

That is a fact about positions, not a forecast, and it is genuinely hard to get any other way. Most chart tools describe price. This one describes where money went in, which is a different kind of statement.

It is also the one honest use of the tool. Anchoring to an arbitrary swing low because it produces a line that looks like support is a different activity, and it is the common one.

A flat but volatile stretch of the long price series. The headline on the chart reads: And a different anchor gives a different line on the same chart. — And a different anchor gives a different line on the same chart. Illustrative chart - not real market data.

Two traders anchoring differently get two different lines from the same data. Anything that moves when you change your mind about the starting point is a description, not a cause — the same objection this site makes to Fibonacci levels and drawn trend lines.

The defence is naming the anchor before drawing it. “Anchored to the Q2 earnings gap” is a claim somebody else could reproduce. “Anchored to the low that makes the line work” is not.

In practice

A flat, quiet stretch of the long price series. The headline on the chart reads: The longer the anchor runs, the less each new bar moves it. — The longer the anchor runs, the less each new bar moves it. Illustrative chart - not real market data.

A long anchor becomes almost immovable. After a thousand bars, one more bar is a thousandth of the total weight — so the line flattens into a fixed reference rather than something that responds to price. That is useful for a long-horizon question and useless as a signal generator.

A candlestick chart with a volume histogram beneath it, with the volume histogram emphasised. The headline on the chart reads: It is one of the few tools that is not pure price. — It is one of the few tools that is not pure price. Illustrative chart - not real market data.

It uses volume, which almost no other chart tool does. On this site’s shared history the five common price-derived oscillators correlate with each other between 0.53 and 0.87 on bar-to-bar changes; a volume-weighted measure is not in that family and can disagree with all of them.

A declining stretch of the long price series. The headline on the chart reads: On foreign exchange the volume input is tick count, not size. — On foreign exchange the volume input is tick count, not size. Illustrative chart - not real market data.

The volume input has to be real. Spot forex has no consolidated volume, so platforms substitute tick count — activity rather than size. Crypto shows one venue’s share. Equities and futures are the clean cases, and this tool is worth much less on the others.

A candlestick series containing several opening gaps, with the largest opening gap marked. The headline on the chart reads: A gap at the anchor bar sets the whole line's starting point. — A gap at the anchor bar sets the whole line's starting point. Illustrative chart - not real market data.

Anchoring at a gap is the most common and most defensible use, and it also means the first bar in the calculation is an unusually heavy one that carries disproportionate weight for a long time.

A long-horizon candlestick view of the same price series. The headline on the chart reads: On a long anchor it becomes a nearly flat reference line. — On a long anchor it becomes a nearly flat reference line. Illustrative chart - not real market data.

A declining stretch of the long price series, with the entry price and the level at which a stop would trigger drawn as horizontal lines. The headline on the chart reads: It is a reference, not a stop level. — It is a reference, not a stop level. Illustrative chart - not real market data.

Do not hang a stop on it. The line moves with volume as well as price, so it can shift on a bar where price barely moved — a stop that reprices for reasons unrelated to your thesis is not a stop.

And trading every touch costs a share of a bar. Illustrative chart - not real market data.

Each touch traded costs 2% of a median bar’s range in round-trip costs on this history, and a line price oscillates around produces a lot of touches.

A 72-bar candlestick section of the shared price history. The headline on the chart reads: Nobody is defending a line only you anchored. — Nobody is defending a line only you anchored. Illustrative chart - not real market data.

And the self-fulfilling effect requires a shared anchor. Session VWAP is watched by a great many people and by execution algorithms benchmarked against it, which gives it real behavioural weight. A line anchored to a point only you chose has none of that.

What anchored VWAP is not

It is not VWAP. That resets each session and is an execution benchmark; this accumulates from a point you picked.

It is not support. It is an average of prices paid, not a place where orders are resting.

It is not objective. The anchor is a judgement, and the line follows from it.

And it is not a volume-weighted moving average. That is a rolling window that discards old bars; this accumulates and never forgets.

When it fails

A sideways, range-bound candlestick series. The headline on the chart reads: In a range it sits in the middle and is crossed constantly. — In a range it sits in the middle and is crossed constantly. Illustrative chart - not real market data.

In a range it sits in the middle and gets crossed on every leg. The average price of a market oscillating around a mean is that mean, and crossings of it are the oscillation rather than a signal.

The second failure is anchor shopping. Trying several anchors until one produces a line that price “respects” is fitting a tool to a chart you have already seen.

A third is trusting the volume input without checking it. Tick volume and single-venue volume are not what the formula assumes.

A fourth is expecting a private line to be defended. The mechanism behind session VWAP is that institutions are measured against it; that mechanism does not transfer to an anchor you invented.

And a fifth is using a very long anchor as a signal. After enough bars the line barely moves, and “price crossed above” becomes a statement about a nearly fixed number rather than about anything current.

The original data

On this site’s shared 576-bar history the five price-derived oscillators measured correlate with each other between 0.53 and 0.87 on bar-to-bar changes, and the round-trip cost of 0.0098 price units is 2% of the median bar range of 0.493 and 45% of the smallest bar of 0.022. The figures are in research/series-measurements.json, produced by site/measure_series.py.

A 72-bar window of the shared price history, cut short at the decision bar. The headline on the chart reads: Price is back at the earnings anchor. Buy it? — Price is back at the earnings anchor. Buy it? Illustrative chart - not real market data.

That correlation range is the case for keeping one volume-based tool on the chart. Five tools built from closes cannot disagree with each other in any way that carries information — the lowest pair in the set still moves together on more than half of all bars. A measure that weights by what traded is the cheapest genuinely independent input available, and anchored to a real, nameable event it answers a question about positioning that nothing derived from price alone can reach. Anchor it to something you can name out loud, and the tool stays honest.

VWAP is the session version and the institutional benchmark. Volume-weighted moving average is the rolling-window relative. And volume explains what the weighting input actually measures.

What I actually do

This is the one discretionary chart tool I would defend without hedging, and only because of what it asks. Anchored to an earnings gap, it answers whether the people who bought that news are ahead or behind - which is a fact about positions, not a prediction, and I cannot get it any other way.

— Michael Whitman

This page is educational, not financial advice. Test every idea on your own charts before risking money.