WhitmanTrading

How to Use VWAP

To use the volume-weighted average price, treat it as a reference for where the average participant is positioned this session. Above it, buyers are in profit on average; below it, sellers are. It resets each session, which is why it says nothing about anything older than that.

The volume-weighted average price is what the average unit traded for so far this session. That is a plain fact about the session rather than a prediction, and its usefulness comes from other people using it too.

Before you start

A session anchor you understand, because the line resets and the reset point matters. Daily, weekly, or anchored to an event — each answers a different question and they are not interchangeable.

A decision about whether you are using it as a filter or as a target. As a filter it decides which direction you are allowed to trade. As a target it marks where price may return to. Pick one.

The instrument’s ordinary bar range, so distance from the line can be judged. On this site’s shared series the median bar range is 0.493 and the ninetieth percentile is 1.101.

The steps

1. Confirm the anchor before reading anything

A candlestick chart with an average line running through it.
The line resets, and the reset point is the whole context. Illustrative chart - not real market data.

A daily line knows nothing about yesterday. If the question you are asking spans several days, the daily version is answering a different one.

2. Read which side price is on

The first half of a price series above a reference line.
Above it, the average buyer this session is in profit. Illustrative chart - not real market data.

Above the line, participants who bought this session are on average in profit. Below it, they are not. That is the whole informational content of the position.

3. Use it as a directional filter

A section of the price series divided by a central reference.
One side at a time removes half the bad trades. Illustrative chart - not real market data.

Long only above, short only below. It removes half the setups and most of the ones taken against the session’s own weight.

4. Measure distance, not just side

A window of price bars stretched far from a reference.
Far from the line is a different condition from near it. Illustrative chart - not real market data.

Price 3 bar-ranges above the line is a different situation from price sitting on it. Distance is where the mean-reversion question lives, and it needs the bar range to be meaningful.

5. Watch the first test after a break

The second half of a price series returning to a level.
The first return is the informative one. Illustrative chart - not real market data.

Price crossing and then returning to the line is the event worth watching. Holding on the first test says something; failing it says something else.

6. Anchor it to an event when the session version cannot help

A range-bound section measured from a fixed origin.
An anchored line answers a longer question. Illustrative chart - not real market data.

Anchor to a high, a low, or a news event and it becomes the average price since that moment — which is the version that works on a swing horizon.

7. Take the stop from structure, not from the line

A long-horizon view with an invalidation level marked.
The line moves every bar; the level does not. Illustrative chart - not real market data.

The line recalculates continuously, so a stop on it drifts. Where the idea is wrong is fixed, and that distance sets the size.

How to tell it worked

You can state which anchor you are using without checking. If you cannot, the line’s context is unknown and its readings are unmoored.

Trades taken against the session’s side are 0 for the period. That is the filter doing what it was added to do.

Distance was expressed in bar ranges rather than in points. 2 bar ranges away means something across instruments; 4 points does not.

And the stop came from a level, so it did not move when the line did.

What it cannot do

A candlestick chart annotated with the round-trip cost of a switch.
Every reaction traded costs a round trip. Illustrative chart - not real market data.

It cannot see across the reset. On this site’s shared series a round trip measures about 2% of the median bar range of 0.493, and a strategy trading every cross of a line that resets daily pays that repeatedly for an average that is only a few hours old. The figures are in research/series-measurements.json.

A candlestick chart with a volume histogram beneath it.
And with little volume the weighting means very little. Illustrative chart - not real market data.

In a thin session the weighting is doing almost no work. The line becomes an ordinary average of a handful of bars, which is a much weaker reference than the name implies.

Choosing the anchor

Session anchoring answers “where is the average participant today”. It is the right question for an intraday trade and the wrong one for anything held overnight.

Anchoring to a significant high or low answers “what has everyone paid since that event”. That is the version that matters when a level is being retested weeks later.

Anchoring to an earnings release or a policy decision isolates the period since new information arrived, which is a genuinely different reference from either of the others.

Use one at a time. Three anchored lines on a chart guarantees one of them is near price, which is the same failure as taking all three signals from an oscillator.

The original data

Of the 24,971 unique videos in research/search-study-corpus.jsonl, 93 have an instruction-shaped title about this average, at a median of 6,568 views across 71 channels. Volume profile appears in 58 videos at 10,561 and order flow in 21 at 14,422. The counts come from site/rank_howto.py, which deduplicates by video id.

A candlestick series with several gaps, the largest of them marked.
An overnight gap starts the session away from the line. Illustrative chart - not real market data.

93 videos at a 6,568 median. It sits in the middle of the indicator set for coverage and near the top for how many of those videos are about intraday execution specifically, which is the context it actually belongs to.

A stretch of price bars cut short at a decision point.
Price is far above the line. Short it back to the average? Illustrative chart - not real market data.

The answer to the question on that chart is that distance from an average is not a reason on its own. A strong session spends the whole day above the line and the gap widens rather than closing. The line tells you where the average participant is, not that price owes them a return — and on this site’s series 54% of 566 ten-bar windows ended higher than they began.

When it fails

The failure is a strong directional session traded as mean reversion, and the line makes it look reasonable every time. Price runs above the average, the distance grows, each new extension looks like a better entry for a fade — and the average itself is dragged upward by the same volume that is driving price. The reference keeps moving toward the price rather than the price returning to it, and every short is taken against the session’s entire weight of trading.

The second failure is forgetting the reset. A daily line knows nothing about yesterday.

A third is trading it in a thin session. The weighting is doing no work.

A fourth is placing a stop on the line. It moves every bar.

A fifth is measuring distance in points. Bar ranges travel between instruments and points do not.

And a sixth is using the session version for a swing trade. That is what anchoring is for.

VWAP explains the weighting and why the reset matters so much. VWAP strategies covers the common ways it is traded. And anchored VWAP is the version for questions longer than one session.

What I actually do

The reason I keep it on an intraday chart when almost nothing else survives is that it is one of very few lines other people are also looking at. Most indicators describe price to me alone; this one describes a level that large orders are actually measured against, which makes reactions around it mean something.

— Michael Whitman

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