On-Balance Volume (OBV) — Does Volume Really Lead Price?

"Volume leads price" is one of the oldest slogans in technical analysis. Behind it sits a simple intuition: big players don't buy with a single order — they accumulate a position over weeks, quietly, so as not to move the market. Price is still standing still, but volume already shows someone systematically absorbing supply. In 1963, Joe Granville turned that intuition into a concrete indicator: On-Balance Volume — and promised it let you see a move before it started.

Sixty years later we can do something Granville couldn't: check it against data. The result is more interesting than a simple "works / doesn't work" — OBV tests out surprisingly decent, but too weak to trade on its own. And that gap is exactly where its real usefulness lives.

How OBV Works

The construction is simple enough to be suspicious:

• a candle closes higher than the previous one → add its entire volume to the running total;
• a candle closes lower → subtract the entire volume;
• an unchanged close → do nothing.

That's it. OBV is a single cumulative line plotted below the chart. Its absolute value means nothing (it depends on the starting point) — only the direction and whether it confirms price matter. Three classic situations:

1. Confirmation. Price rises and OBV rises, printing new highs alongside price — the move has "fuel," real turnover is behind the advance.
2. Bearish divergence. Price prints a new high, but OBV prints a clearly lower high — the rally is running on progressively less volume, demand is thinning out. A warning, not a signal.
3. Bullish divergence / silent accumulation. Price sits in consolidation or makes lower lows while OBV climbs steadily — someone is accumulating. This is the "volume leads price" scenario Granville was hunting for.

Philosophically, OBV is a crude approximation of capital flow: it treats the entire session's volume as buying or selling based purely on the direction of the close. That's obviously a brutal simplification — a candle up 0.01% "counts" its whole volume as demand. This bluntness is at once the indicator's biggest strength (it resists overfitting) and its biggest weakness (a lot of noise).

It's worth distinguishing OBV from its close cousin, the Accumulation/Distribution Line (A/D Line). A/D also accumulates volume, but weights it by where the close sits within the candle's range: a close near the highs adds most of the volume, near the lows it subtracts most, in the middle it barely moves the needle. Sounds smarter, but it has its own flaw: it ignores price gaps, so after a gap day it can show accumulation that wasn't really there. Neither indicator is "better" — they just simplify the same reality differently, which is one more reason not to treat either as an oracle.

What the Numbers Say

QuantifiedStrategies tested an OBV-based strategy on stock market data. Results: 369 trades, an average of +0.6% per trade, a 75% win rate, profit factor of 2.01 (figures from QS summaries — the source file wasn't available in our research, so confirm the numbers on the source site before citing them). The authors' own conclusion, though, matters more than the numbers: OBV alone is a weak strategy — its value shows up as confirmation of other signals.

How do you reconcile "75% win rate, PF 2.01" with "weak strategy"? Three things become visible only once you look under the hood of results like these:

• +0.6% per trade is a thin margin. On liquid, zero-commission ETFs — acceptable. On crypto, with spread, commission and slippage, a realistic cost of 0.1–0.2% per side can eat half of that edge. Thin edges don't survive a move to more expensive markets.
• A high win rate ≠ a strong system. A 75% win rate with small per-trade gains is a typical profile for mean-reversion approaches — it works until a rare, deep loss consumes a whole streak of small wins. A PF of 2.01 says the balance is positive, but without the safety margin that comes from systems with an asymmetric payoff.
• One test, one market. 369 trades is a decent sample, but it's still one rule variant on one market. The verdict "good as confirmation, weak alone" is far better grounded than the specific numbers.

That's the honest frame for OBV: a second-opinion indicator. It doesn't generate an edge on its own, but it can filter out some false signals from systems that do have an edge — especially breakout systems, where the question "is there real turnover behind this move?" is the whole question.

How to Apply It Step by Step

The most sensible use of OBV in crypto: a confirming filter for breakouts out of consolidation (D1 or H4).

1. Find a consolidation: the market has been range-bound for weeks, with a clear resistance overhead (e.g., BTC sawing between 88,000–92,000).
2. Watch OBV during the consolidation: this is the key step. A flat OBV = the market is genuinely undecided. OBV rising while price sits still = silent accumulation; a breakout to the upside will have backing. OBV falling = distribution; treat an upside breakout with suspicion.
3. Signal: a candle close above the consolidation resistance plus OBV at a new local high (confirmation). A price breakout without a new OBV high is a second-tier signal — skip it or trade half size.
4. Stop and size: stop below the breakout level or the last low in the consolidation; size from the % risk model; minimum 2:1 R:R to the nearest target.
5. Divergences as an alarm, not a signal: a bearish OBV divergence at trend highs isn't a short order — it's a reason to tighten stops and stop adding to longs.

The routine is simple and cheap in time: once a day, after the D1 candle closes, glance at two lines — price and OBV — on the markets you watch. You're looking exclusively for divergences: price stands still while OBV moves (accumulation/distribution in progress), or price makes an extreme that OBV doesn't confirm. Everything else is noise you can safely ignore.

Numerical example on ETH (illustrative): ETH has been range-bound between 3,000–3,300 for five weeks. Price is hovering around the middle of the range, but OBV has been climbing steadily for three weeks and has just broken above its level from the last price high — a textbook accumulation picture. The daily candle closes at 3,340, above resistance, on 1.8x average volume; OBV is at a new high. Entry at 3,340, stop below the last low in the range at 3,140 (−6%), target from the width of the consolidation: 3,300 + 300 = 3,600 (R:R ≈ 1.3:1 — below the threshold, so honestly: either you wait for a retest of 3,300 for a lower entry and R:R ~2.3:1, or you skip it). This example is deliberately built to show that OBV confirmation doesn't excuse you from R:R arithmetic — the filter improves signal quality, not trade geometry.

When It Doesn't Work and Common Pitfalls

• Trading OBV alone. The backtest conclusion is unambiguous: solo isn't enough. OBV without price-structure context generates dozens of "divergences," most of which lead nowhere.
• Divergences in a strong trend. The most expensive trap the indicator sets. In a strong uptrend OBV can diverge for weeks while price keeps rising anyway — anyone shorting "because divergence" is paying for the lesson that divergence is a necessary condition for a reversal, never a sufficient one.
• Thin altcoins and fake volume. OBV is only as good as the volume data underneath it. On small pairs volume is often pumped (wash trading) or concentrated on a single exchange — OBV calculated from that data is reading tea leaves. Stick to BTC/ETH and liquid altcoins.
• A single anomalous candle. The "all volume counts toward the close direction" construction means one panic session with gigantic turnover can bend the OBV line for weeks. After such sessions the indicator shows the past for a while, not the present.
• Comparing OBV values across markets. OBV is a running sum from an arbitrary starting point — "BTC's OBV vs. ETH's OBV" means nothing. Only compare the shape of the line against the shape of price on the same market.
• Minute-level timeframes. On LTFs volume noise dominates the signal; sensible OBV applications start at H4/D1. The higher the timeframe, the more real market decisions are packed into each volume reading.

The honest bottom line on the question in the title: sometimes yes, volume can lead price, and OBV is the simplest tool to see it. But the numbers show that edge is too thin to trade alone: +0.6% per trade is a margin that costs and a single bad streak can eat. Treat OBV as a smoke detector, not a navigation system: it's great at warning you something is smoldering beneath a consolidation, but direction, entry, stop and position size still have to come from a strategy with its own, measured edge.

FAQ

Rafał — Krypto Bez Ściemy

Trader and founder of Strefa Tradingu. He’s been breaking crypto down on YouTube for years — no hype, no signals, with a focus on market structure and risk management.

kryptobezsciemy.plInstagramYouTubeKlub Zero