On-Balance Volume
On-Balance Volume (OBV) is a technical indicator that combines price and trading volume into a single running total, intended to reveal whether volume is flowing into or out of an asset even when the price itself hasn't moved much yet. It was developed by the market analyst Joseph Granville and introduced in his 1963 book "New Key to Stock Market Profits," making it one of the older volume-based indicators still in wide use today.
How OBV Is Calculated
OBV is a cumulative figure updated one period (usually one trading day) at a time using a simple rule: if today's closing price is higher than yesterday's, add today's entire volume to the running OBV total; if today's close is lower, subtract today's volume; if the close is unchanged, OBV stays the same. Concretely: suppose OBV starts at zero, and on Day 1 the stock closes at $50 on volume of 100,000 shares with a higher close than the prior day, bringing OBV to 100,000. On Day 2 it closes at $51 (higher again) on volume of 150,000 shares, bringing OBV to 250,000. On Day 3 it closes at $49 (lower) on volume of 80,000 shares, subtracting that volume and bringing OBV down to 170,000. The absolute level of OBV isn't meaningful on its own - what matters is its trend and slope over time.
Two consequences of that starting point are worth noting before reading any chart. First, because the series begins from an arbitrary number, two charting packages that start their calculation on different dates will show different absolute values for the same stock on the same day, which is harmless as long as nobody compares the numbers themselves. Second, because each day contributes the whole of its volume, a single extraordinary session - an index rebalancing, an earnings gap, a block crossing - can shift the line by more than the preceding month of ordinary trading did, and that shift will persist in the running total forever. Analysts who use the indicator seriously tend to look at its slope over a defined window rather than at the line since inception.
Reading OBV Divergences
The main use of OBV is spotting divergence between price and volume-based momentum. If a stock's price is flat or only drifting sideways for a couple of weeks while OBV climbs steadily higher, that's read as a sign of quiet accumulation - more volume is flowing in on up days than out on down days, even though price hasn't broken out yet - often interpreted as a bullish signal that a move higher may follow. The mirror image, price holding steady while OBV steadily declines, suggests distribution or selling pressure building beneath a seemingly stable price, a bearish warning.
The word divergence does a lot of work here and deserves a caution. A divergence is only meaningful against a defined reference: the indicator making a lower high while price makes a higher high, over a span long enough that both readings are genuinely peaks rather than noise. Drawn loosely on a chart after the fact, a divergence can be found almost anywhere, because two series that are related but not identical will constantly disagree over short windows. The practical discipline is to fix the lookback before looking - comparing, say, successive swing highs at least a few weeks apart - and to accept the result when no divergence is present, which is the majority of the time.
OBV Across Markets
OBV works best wherever reported trading volume is centralized and reliable - exchange-listed stocks, futures, and most major cryptocurrency pairs on centralized exchanges all qualify. It's considerably less reliable in the spot forex market, which is decentralized and traded over-the-counter across many venues with no single official volume figure; forex platforms typically substitute "tick volume" (the number of price changes) as a rough proxy, which measures something related but not identical to actual traded volume, weakening the indicator's original logic in that context.
The rule that makes this indicator simple is also its structural weakness: the entire day's volume is assigned to the direction of the close, with no gradation. A session that opens down 5 percent, trades heavily all day and closes a fraction of a cent higher counts as a full day of accumulation, identical in the calculation to a session that rises steadily from open to close on the same volume. A session that closes a fraction lower counts as full distribution. Around a flat close the indicator therefore flips its entire daily contribution on a rounding difference, which is why it behaves erratically in rangebound markets and why a single very high-volume day can dominate weeks of the series.
The obvious fix produced a whole family of successors. Instead of assigning volume by the direction of the close, weight it by where the close sits within the day's range, so a close at the high counts fully positive, a close in the middle counts as nothing, and a close at the low counts fully negative. The accumulation and distribution line and the money flow measures built on it all work that way, and they trade the crude clarity of the original for a smoother, less jumpy reading. Granville's own record is a suitable coda. Having built a following on this indicator, he turned emphatically bearish in early 1982 and stayed so through the start of one of the longest bull markets on record, which did more damage to his reputation than to the indicator, and is a reminder that a tool for measuring whether volume confirms a move was never a tool for deciding what the move would be.