Negative Volume Index (NVI)

The Negative Volume Index (NVI) is a technical analysis indicator that builds a running total of price changes on days when trading volume is lower than the day before, and ignores days when volume rises. Its premise is that crowds of less-informed traders dominate busy, high-volume days, while well-informed investors, often called "smart money", do more of their buying and selling quietly when volume is low. Tracking prices only on those quiet days is meant to reveal what the informed side of the market is doing.

The idea is usually traced to the market analyst Paul Dysart in the 1930s and was popularised by Norman Fosback in his 1976 book on stock market indicators. The NVI has a companion, the Positive Volume Index, which changes only on days when volume rises. Both belong to the same family as other volume tools such as the Money Flow Index and the Klinger Oscillator, and they are available on many charting platforms.

Imagine judging a restaurant by how its prices change, but only paying attention on quiet weekday afternoons, when regulars and insiders are there, and ignoring busy weekend evenings full of tourists. The theory is that the quiet days show what knowledgeable customers are willing to pay, while the busy days are distorted by crowds.

The NVI applies this to markets. On a quiet trading day, it adds that day's percentage price change to its running score. On a busy day, it leaves the score unchanged. Over months, the line shows whether prices have tended to rise or fall on quiet days, which analysts read as a hint about the direction informed investors favour.

The index usually starts at an arbitrary base value, often 1,000. On each day, if volume is lower than the previous day's volume:

If volume is equal to or higher than the previous day's, the index is unchanged: $\text{NVI}_t = \text{NVI}_{t-1}$.

• $\text{NVI}_t$ is the index value today and $\text{NVI}_{t-1}$ yesterday's value.
• $C_t$ and $C_{t-1}$ are today's and yesterday's closing prices.

For example, suppose the NVI stands at 1,000. On Monday, volume falls and the price rises from 50 to 51, a gain of 2%, so the NVI becomes 1,000 + 0.02 × 1,000 = 1,020. On Tuesday, volume rises and the price drops to 49; the NVI stays at 1,020. On Wednesday, volume falls again and the price rises from 49 to 49.98, a 2% gain, so the NVI becomes 1,020 + 0.02 × 1,020 = 1,040.40.

• Comparison with a long moving average. The most common approach plots the NVI against its own long-term moving average, often a one-year average of about 255 trading days. When the NVI is above the average, conditions are read as favourable for a bull market; when it is below, as less favourable.
• Trend confirmation. As part of a wider trend analysis, a rising NVI during a price rise suggests the advance has support on quiet days, not just on bursts of heavy trading.
• Pairing with the Positive Volume Index. Analysts sometimes compare the two indices to judge whether price moves are being driven mainly by quiet accumulation or by crowd-driven, high-volume days.
• Market-wide use. The NVI is often applied to a broad index rather than single shares, as a gauge of the overall market's health.

Fosback's historical studies of US stocks reported that periods when the NVI was above its one-year average were much more often followed by rising markets than periods when it was below, which is the main reason the indicator is still used. Such results come from past data and particular markets, and later periods have not always matched them.

Some crypto traders apply the NVI to coin charts, but its core assumption fits less neatly. Crypto trades around the clock, so "daily" volume depends on an arbitrary cut-off time, and volume is spread across many exchanges, with some reported figures inflated by wash trading. Large holders often trade over the counter, which does not appear in exchange volume at all. The notion of "smart money" acting on quiet days may therefore be harder to detect in crypto than in traditional stock markets.

The NVI rests on an assumption about who trades on quiet and busy days that is difficult to prove and may change over time, especially with the growth of algorithmic and high-frequency trading. Because it uses a long moving average, its signals are slow and can lag turning points by months. It ignores price moves on high-volume days entirely, which can include important news-driven moves. Most analysts therefore treat it as a background gauge of market conditions alongside other tools, such as MACD and the Relative Strength Index, rather than as a trading signal on its own. Like other indicators it summarises historical data and does not predict prices.