Anchored VWAP vs VWAP: Formula, Anchors and Settings
Anchored VWAP vs VWAP: the same formula measured from a bar you choose instead of the session open. The price input is a setting, and it moves your level.
Anchored VWAP is the volume-weighted average price measured from a bar you pick. Plain VWAP measures it from the session open and resets every session. The formula is the same in both cases, and your level moves with where you start the line and which price of each bar it weights.
What is anchored VWAP?
The average price paid since a point you choose, weighted by volume, drawn as a line forward from that point.
TradingView's help page defines it as a tool that "displays the volume-weighted average price for a specific time period, starting from a user-selected point", and describes the weighting as taking "into account the number of shares traded at each price level". That word, shares, matters once you put the tool on a currency pair.
TradingView says the tool was "developed by the late physicist and technical analyst, Paul Levine, while working on the Market Interpretation/Data Analysis system (MIDAS) between the years 1995 and 1997". It credits earlier work: "Levine developed the methodology of professional trader, Kevin Haggerty, who used volume-weighted average price curves to determine the direction of a position in the 80s." It names "Brian Shannon, Zach Hurwitz, and the CMT Association" as the people who brought it into the mainstream. StockGro dates it the same way: "Introduced by Paul Levine between 1995 and 1997".
Anchored VWAP vs VWAP
One line resets on a clock. The other starts where you put it and keeps going.
| VWAP (session) | Anchored VWAP | |
|---|---|---|
| Starting point | The session open, decided by the platform | A bar you select |
| Reset | Every new session | Never, until you move the anchor |
| What it is a benchmark for | The average price paid today | The average price paid since the event you chose |
| Who decides the window | The platform | You, and you have to be able to justify it |
| Stability | Settles as the session fills out | Unstable right after the anchor, by the same mechanism |
Searches for rolling VWAP, session VWAP and normal VWAP are the same question in different words. Session and normal VWAP are the clock-reset version in the first column. A rolling VWAP is an anchor you move on a schedule, and OptimusFutures flags the thing people forget about it: "traders often need to remember to re-anchor VWAP on a new month or week".
What is the anchored VWAP formula, and which price does it use?
The sum of price times volume from the anchor, divided by the sum of volume from the anchor. Which price goes into that sum is a setting rather than a fixed part of the formula, and the published definitions use three different ones.
- StockGro prints "Anchored VWAP = Σ(Price × Volume) / ΣVolume" and never says which price. Its worked example uses one price per bar without naming it.
- LiteFinance names the input: Pi is the "average value of a candlestick (High + Low + Close) / 3", Vi the "volume of this candlestick", with "the total product of price and volume from the anchor point" over "total volume from the anchor point".
- OptionsTradingIQ prints the close instead: "Anchored VWAP = Cumulative [(Volume x (Close)) / Cumulative Volume]". As printed, the division sits inside the cumulative bracket, which is not the quantity the page goes on to describe.
OptionsTradingIQ also explains why the definitions differ: "Almost all Anchored VWAP indicators will let you choose how to calculate the VWAP at each candle", and "A good starting point is to just use the close, but OHLC is also a solid starting point." Your own chart has a default for that setting, and it is worth looking up.
The difference between those two inputs on any one bar is
close - (High + Low + Close) / 3 = (2C - H - L) / 3
Because the close always sits inside the bar, that difference can never exceed one third of the bar's range in either direction. It is largest on bars that close at an extreme, and it is zero when a bar closes at its own midpoint.
On a bar with an 80 pip range that closes at its high, the close is 26.7 pips above the (High + Low + Close) / 3 price, one third of 80. Feed a run of bars like that into the sum and two charts of the same instrument, with the same anchor, print two different levels. Before you compare your anchored VWAP with anyone else's, check which price input each chart is using.
A worked example
Three bars, round numbers, for the arithmetic.
| Bar | Price input | Volume | Price x volume | Running sum of price x volume | Running volume | Anchored VWAP |
|---|---|---|---|---|---|---|
| 1 | 1.1000 | 1,000 | 1,100.0 | 1,100.0 | 1,000 | 1.10000 |
| 2 | 1.1030 | 3,000 | 3,309.0 | 4,409.0 | 4,000 | 1.10225 |
| 3 | 1.1060 | 2,000 | 2,212.0 | 6,621.0 | 6,000 | 1.10350 |
The level after three bars is 6,621 divided by 6,000, or 1.10350. A plain average of the three price inputs would be 1.10300. The two higher prices carry more volume between them than the first bar does, so they pull harder.
Where do you put the anchor?
On a bar you can justify in one sentence.
ChartingLens says it twice: "If the anchor has no story, the line has no meaning", and "An AVWAP from an arbitrary date is an arbitrary line." It offers a test: "could you explain to another trader in one sentence why that bar matters?" And "Every anchor needs a one-sentence justification."
Most of the published anchor advice was written for stocks. ChartingLens and TrendSpider both lean on equity events, TrendSpider listing anchors "such as earnings, breakouts, swing highs/lows, or macro events". A currency pair has no earnings release and no listing day, so those two anchors have nowhere to land. Everything else on that list still does.
The anchor list that already works in forex vocabulary is LiteFinance's five: the trading day opening, a key level breakout, a reaction to news, a reversal point, and the onset of a strong trend move. Its anchor cases include "news releases, support and resistance breakouts, or the beginning of a new trend", and it says plainly that "AVWAP helps analyze significant economic events or statements by the Fed." TradersMastermind simply does it, with a worked chart titled "Anchoring VWAP to a swing high in USDJPY".
For a currency pair, drop earnings and IPO day and keep the method. What survives is the session or week open, a swing high or low, a scheduled release or central bank decision, a key level break, and the start of a strong trend move.
What is the volume in a forex VWAP?
Whatever volume your feed supplies, and the published guides do not settle it for you.
A volume-weighted average needs a volume. On a stock, that is shares traded, which is why TradingView's definition reaches for the word shares. On a spot currency pair there is no single exchange tape to count.
MetaTrader's references define two volume fields on a price bar, printed identically in the MQL4 and MQL5 documentation: "long tick_volume; // Tick volume" and "long real_volume; // Trade volume". Neither page says which of the two a forex symbol fills.
So look up what your own symbol reports, and expect an anchored VWAP level that depends on your feed. Record the feed alongside the level, and treat a level somebody else published on a different feed as a different number.
Anchored VWAP vs a moving average
A moving average weights every bar in its window the same way and ignores volume. Anchored VWAP weights by participation and has no window at all, only a start.
LiteFinance states the mechanical difference plainly: "Anchored VWAP calculates the volume-weighted average price from the anchor point, while moving averages calculate the average price over a certain period, disregarding trading volume." ChartingLens puts the same thing from the reader's side: a moving average answers what the last N bars averaged, with N arbitrary, while an anchored VWAP answers what has been paid since an event, "weighted by real participation".
How do you read anchored VWAP?
As a benchmark. Price above the line means buyers since your anchor are, on average, in profit, and below it means the reverse. It is a reference level, and it is not a trigger.
OptimusFutures gives two caveats. The first is the line's early instability: "realize the closer you are to the anchor point, the more volatile VWAP will be and the less reliable it will be". The second is how long to wait: "You want there to be enough volume that there's an established VWAP traders would use as a reference point", and "For Anchored VWAP that extends of multiple days, you'd want at least a few days of data."
The same page describes using the line as a kind of trailing reference for a position already open. This guide gives no entries, exits, stops or targets.
What anchored VWAP cannot tell you
- Whether a level that has held will hold again. OptimusFutures names that as the common mistake on the tool: traders watch it work as support and resistance and then assume it will keep working.
- Where price is going. It is an average of what has already been paid.
- Anything at all, if the anchor has no story behind it, which is ChartingLens's point.
- What volume it is weighting by, until you have checked your own symbol.
Where is anchored VWAP in the menu?
On TradingView it is a drawing rather than an indicator, which is why searching the indicator list does not find it: "Select anchored VWAP from the drawing tools list on the chart's left-side panel", then click the bar you want to start from. Other platforms file it differently, so check both the drawing tools and the indicator list on yours.
| Tool | What it does |
|---|---|
| TradingView | Anchored VWAP as a drawing tool. Click to set the anchor; color and thickness are editable |
| LiteFinance's guide | Formula with (High + Low + Close) / 3 as the price input, plus a five-item anchor list with no equity events in it |
| OptionsTradingIQ | The close as the price input, and a note that almost every implementation lets you change it |
| OptimusFutures | Writes about what goes wrong: assuming a level keeps holding, forgetting to re-anchor, and instability near the anchor |
| ChartLabs | Anchored Volume Weighted Average Price, an indicator under Standard in the Indicators menu. Type the anchor into its Anchor (UTC) setting as a date and time, YYYY-MM-DD HH:MM; left empty, the line starts from the first bar on the chart. Source defaults to (H + L + C) / 3 and can be changed. The session version, Volume Weighted Average Price, resets every day at 00:00 UTC by default. The chart's legend shows each bar's volume. Manual backtesting bar by bar on cleaned data from Dukascopy, with every trade journaled against the chart you saw |
What has to be recorded for the reading to be reproducible?
Four things: the anchor bar, the price input, the volume field, and the instrument.
Change the price input and your level moves by up to a third of a bar's range on the bars that matter most. Change the feed and the volume weights change with it. Move the anchor one bar and, early on, the line moves a lot, because the line is least stable right where it starts.
So a test of an anchored VWAP idea is a test of one specific configuration. Write it down next to the result, then step the chart bar by bar and log what you did at the level: whether you took the trade, whether you waited, and how it resolved. That is the part a person has to do, and it is what manual backtesting is for.
Related reading: Forex Backtesting: Test Your Strategy on Real Data for the session and spread effects behind any forex test, Backtesting a Trading Strategy: A Beginner's Guide for the step-by-step method, and Manual vs Automated Backtesting for which of the two answers which question.
Next step
The anchor is a judgement call, and judgement is trained by repetition with a record of what you did. Step real historical charts bar by bar, take the trades by hand, and read every one of them back against the chart you were looking at. ChartLabs has a 7-day free trial with no credit card, then one plan at $20 a month or $200 a year: start your free trial.
Related Topics
Related Guides
Manual vs Automated Backtesting: Which One Should You Use?
Automated backtesting tells you whether a set of rules made money. Manual backtesting tells you whether you can trade them. A practical comparison of what each one tests, what it costs you in time and skill, and which to reach for first.
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.