VWMA Strategy Guide: Calculation, Signals, and Testing
A volume-weighted moving average gives more weight to bars with greater reported volume. It can help compare price with recent trading activity, but it cannot reveal who traded or guarantee a stronger signal than a simple moving average.
Calculate VWMA
For a window of N bars, VWMA equals the sum of each bar's price multiplied by volume, divided by total volume in the same window. TradingView's VWMA documentation describes the indicator and its interpretation.
This hypothetical five-bar example uses closing prices:
| Close | Volume | Close × volume |
|---|---|---|
| 50 | 1,000 | 50,000 |
| 55 | 1,500 | 82,500 |
| 58 | 2,000 | 116,000 |
| 60 | 1,800 | 108,000 |
| 62 | 1,200 | 74,400 |
Total weighted price is 430,900 and volume is 7,500, so VWMA is approximately 57.45. The close-times-volume column is a bar-based approximation, not measured transaction-level dollar turnover.
Compare VWMA with SMA
An SMA assigns equal weight to each bar. With identical volume on every bar, VWMA and SMA of the same price series and length coincide. Lower overall volume alone does not make them equal; relative volume across bars matters.
VWMA above SMA means the higher-volume observations, collectively, pull the weighted average upward. That is not proof of institutional buying. VWMA can respond faster or slower than SMA depending on the distribution of volume and prices.
Three rules worth testing separately
- Price cross: enter only after a completed candle crosses the VWMA. Specify whether the next bar opens the trade, and where an invalidation exit belongs.
- Trend filter: allow long candidates only when VWMA slopes upward. Choose the slope lookback in advance rather than adjusting it to fit losing trades.
- VWMA/SMA relationship: compare averages with the same length and source. A cross changes their relative weighting relationship; it is not a direct measurement of buying versus selling pressure.
Combining all three can reduce the number of signals without improving the remaining trades. Compare each addition against the simpler rule on a separate evaluation period.
Volume and execution limits
Exchange volume, broker tick volume, and individual crypto venue volume describe different datasets. A zero-volume window makes the ratio undefined; missing volume must not be treated as evidence of inactivity. Also distinguish a rolling VWMA window from VWAP's anchor/reset structure.
Use position size and a planned exit to constrain exposure, but account for gaps and slippage: a stop price is not a guaranteed fill or a guaranteed maximum loss. A small paper-trading sample can expose workflow problems; it cannot establish profitability. Record all signals, costs, and excluded trades before drawing conclusions.
The earlier AAPL, SPY, NVDA, and META anecdotes are not supported by a reproducible dataset. The calculation and test procedure above stand independently of those claims.