What it is
Volume Weighted Average Price is the average price paid over a period with every trade weighted by its size. It resets at the start of each session, so it answers a precise question: relative to everyone who has traded today, is the current price a good fill or a bad one?
How it works
It is computed cumulatively through the session. Each bar contributes a typical price, usually (high + low + close) / 3, multiplied by that bar's volume; the running total of those products is divided by the running total of volume. Bands are commonly drawn one, two and three standard deviations of price away from the running line. An anchored VWAP applies exactly the same arithmetic from a chosen starting point — an earnings gap, a swing low, the first print of the year — instead of from the session open.
How traders use it
Institutional desks use VWAP as an execution benchmark: an algorithm tasked with accumulating a large order is judged on whether it filled below the session VWAP. That mandate creates real mechanical demand near the line, which is why intraday traders treat it both as a magnet and as a bias line. Trading above a rising VWAP is a different environment from chopping around a flat one. Anchored VWAP from a major event marks the average cost of everyone who has traded since that event, a useful proxy for where a crowd of positions flips from profitable to underwater.
Where it breaks down
The limitations are structural. VWAP is cumulative, so by late afternoon it is dominated by the morning and barely reacts to new information, while in the first bars of a session it is jumpy and close to meaningless. It depends on the volume feed, which is fragmented across venues in equities and varies by exchange in crypto, and it is not usable on instruments without reliable volume such as spot forex. It is a reference, not a signal: crossing VWAP is a fact about the day, not a reason to trade.
Educational reference. This entry describes how a concept is defined and used. It is not investment advice, not a recommendation, and not a signal. Any rule you build from it should be tested with realistic costs before it is traded, and no historical result guarantees a future one.