On Balance Volume Indicator: How OBV Confirms Breakouts and Flags Divergence

Price tells you where a stock closed. Volume tells you how many people bothered to agree on that price. The on balance volume indicator, better known as OBV, is one of the oldest attempts to fold those two facts into a single line. It is simple enough to calculate with a pen. That simplicity is both its strength and the reason it misleads people who take it too literally.

We’ll go through where OBV came from, how the running total is built, what a rising or falling line actually means, the two signals traders lean on (breakout confirmation and divergence), and the weak spots you should know about before you put money behind it. There’s a worked example on a made-up NSE stock so you can see every number move.

What the on balance volume indicator measures

OBV was developed by Joe Granville, who first explained it in his 1963 book Granville’s New Key to Stock Market Profits. StockCharts’ ChartSchool describes it as an indicator that assesses buying and selling pressure through cumulative volume, and notes it was among the earliest measures to track the inflow and outflow of volume.

Granville’s core idea was that volume precedes price. If more shares change hands on up days than on down days, someone is accumulating, and the price should eventually follow. If down-day volume dominates, the opposite. Investopedia’s OBV explainer frames this as a contest between “smart money”, meaning institutions such as mutual funds and pension funds, and retail investors: institutions buying into a stock that retail is selling can push volume up while the price barely moves, until the price finally catches up.

Whether you buy that story or not, the mechanics are clean. OBV is a single line that climbs when buying pressure wins and falls when selling pressure wins.

How OBV is calculated

There are only three rules. Fidelity’s indicator guide lays them out like this:

• If today’s close is higher than yesterday’s close: OBV = yesterday’s OBV + today’s volume.
• If today’s close is lower than yesterday’s close: OBV = yesterday’s OBV − today’s volume.
• If today’s close equals yesterday’s close: OBV = yesterday’s OBV.

That’s it. Notice what is missing. The size of the price move plays no part. A stock that closes up 50 paise and a stock that closes up 8% both get the full day’s volume added. The whole session’s volume is treated as up-volume or down-volume based purely on the direction of the close.

Also notice that the running total has to start somewhere. StockCharts sets the first value equal to the first period’s volume; plenty of people simply start at zero. Either way the absolute number is arbitrary. Investopedia makes the point that because OBV is cumulative from a chosen start date, its numerical value depends on that start date. ChartSchool goes further and says the scale isn’t even relevant. What you read is the direction and shape of the line.

Worked example: ten sessions on a hypothetical NSE stock

Take an imaginary large-cap, call it Sahyadri Engineering, trading on the NSE. The closes and volumes below are invented purely to show the arithmetic. Volume is in lakh shares, and we start OBV at zero on day 1.

Walk through a few rows. Day 3 closed lower than day 2, so its 9 lakh shares come off: 15 − 9 = 6. Day 5 closed flat at 510, so OBV stays at 24 and those 7 lakh shares simply don’t count. Day 9 was a soft day, down just ₹2, but the full 7 lakh is subtracted anyway.

Now step back and look at the pattern. Between day 2 and day 9 the price drifted inside a band of roughly ₹503 to ₹514. Nothing dramatic. But the four up days carried 15 + 18 + 20 + 22 = 75 lakh shares, while the four down days carried only 9 + 8 + 6 + 7 = 30 lakh. Heavy volume on the up days, light volume on the pullbacks. OBV quietly climbed from 0 to 45 while the price went nowhere much, and on day 10 the price broke above its range high of 514 with OBV also at a fresh high.

OBV doesn’t care how far a stock moved. It only asks which way it closed, and how many shares voted on that close.

Reading the line: trend, ranges and breakouts

ChartSchool suggests a three-step routine. First, define the trend of the OBV line itself. Second, check whether that trend matches the price trend. Third, look for support and resistance levels on OBV, because breaks of those levels can generate signals of their own.

The interpretations that follow are fairly consistent across the sources:

• Price and OBV rising together. Fidelity’s guide says that when both are making higher peaks and higher troughs, the upward trend is likely to continue. The reverse holds for lower peaks and troughs in a downtrend.
• OBV rising during a sideways range. Fidelity reads this as possible accumulation and a warning of an upward breakout. A falling OBV inside a range points to possible distribution and a downside break.
• OBV confirming a breakout. ChartSchool says OBV can be used to confirm a price trend, an upside breakout or a downside break. In its examples the price breaks a trendline and OBV breaks its own resistance or support at around the same time.

That second point is where our Sahyadri example sits. A range-bound stock with OBV grinding higher is the classic setup. The breakout on day 10 came with OBV already at a new high, so the volume picture backed the price move rather than contradicting it.

If you already use other volume tools, OBV is a good cross-check rather than a replacement. Our guide to the Chaikin Money Flow indicator covers a volume measure that weighs where the close sits within the day’s range, which answers a slightly different question.

A practical filter: put a moving average on OBV

Raw OBV is jumpy. Investopedia suggests adding a moving average to the OBV line and watching for OBV breakouts; if OBV breaks out at the same time as price, that supports the price breakout. ChartSchool’s own illustration shows OBV smoothed with a 20-day EMA. The average doesn’t add information, it just makes it easier to see when the line has genuinely turned rather than wobbled for a day or two.

OBV divergence: when volume and price disagree

This is the signal most traders actually come to OBV for. ChartSchool defines the two types:

• Bullish divergence: OBV moves higher or forms a higher low while the price moves lower or makes a lower low.
• Bearish divergence: OBV moves lower or forms a lower low while the price moves higher or makes a higher high.

Fidelity uses slightly different labels, calling them positive and negative divergence, and describes the bearish case as price making higher peaks while OBV fails to make higher peaks, which suggests the uptrend is likely to stall or fail.

Here’s what a bearish divergence might look like on a hypothetical stock. In early September it peaks at ₹820, and OBV at that point reads, say, 140 lakh. A few weeks later the price pushes to a higher peak at ₹845, but OBV only gets back to 128 lakh. Price says buyers are still in charge. Volume says fewer shares turned up for the second push than the first. That gap is the warning.

Two cautions. A divergence is a warning, not a timing tool: ChartSchool’s own wording is that it should alert chartists that a price reversal could be in the making, which is a long way from saying when. And ChartSchool also points out that OBV is built on closing prices, so you should compare closing highs and lows when looking for divergences, not intraday wicks. A stock that spiked to ₹850 intraday and closed at ₹838 has a closing high of ₹838 for this purpose.

A divergence tells you the crowd behind a move is thinning out. It doesn’t tell you when the move ends.

Where OBV goes wrong

Every indicator has failure modes, and OBV’s are worth spelling out because they are baked into the formula.

1. It ignores the size of the move

We saw this in the table. Day 9’s ₹2 dip subtracted the same 7 lakh shares a ₹30 crash would have. Investopedia contrasts OBV with the Volume-Price Trend indicator, which looks at how much higher or lower the close was, not just the direction. The accumulation/distribution line takes another route, using where the close sits within the recent range and multiplying by volume.

2. One huge volume day can distort it for a long time

Investopedia warns that a large single-day volume spike can throw off the indicator for quite a while, and gives examples such as a surprise earnings announcement, a stock being added to or removed from an index, or big institutional block trades. ChartSchool similarly notes that volume spikes can cause a sharp move that needs a settling period.

Run the numbers on our hypothetical stock. Say it usually trades around 20 lakh shares a day, and on results day it trades 3 crore (300 lakh) shares and closes up just ₹1. OBV jumps by 300 lakh in one session. That’s 300 ÷ 20 = 15 normal sessions’ worth of volume, all booked as “buying” because of a one-rupee close. Indian traders will recognise the situation: results days, index rebalancing days and big block deals all produce exactly this kind of session. When you see a vertical jump in OBV, check what happened that day before reading anything into it.

3. It produces false signals

Investopedia describes OBV as a leading indicator that is prone to false signals and recommends balancing it with lagging indicators. ChartSchool’s bottom line is similar: use OBV alongside other parts of technical analysis, such as pattern analysis or momentum oscillators, and don’t treat it as a standalone signal.

4. Volume isn’t the same as delivery

OBV works off the traded volume number on your chart. That figure doesn’t tell you whether shares were taken into delivery or squared off within the same session. If you want that second layer for an Indian stock, our explainer on delivery percentage in stocks covers how to read it alongside price and volume.

A simple way to use the on balance volume indicator

Put together, a sensible OBV routine for swing traders looks something like this:

1. Start with the price structure. Mark the range, the trendline or the pattern first. OBV is there to confirm or question it, not to find it.
2. Check the OBV trend against price. Are both making higher highs? Is OBV rising inside a flat range? Or is it sagging while price holds up?
3. Draw levels on OBV too. A break of OBV’s own resistance alongside a price breakout is stronger than a price breakout on flat OBV.
4. Scan for one-off spikes. If the line jumped on a results day or a block deal, mentally discount that bar.
5. Confirm with something else. A momentum oscillator, a moving average or a candlestick pattern at the level. And decide your stop loss before you enter, not after.

Back to Sahyadri Engineering for a moment. A trader using this checklist would have seen the range (roughly ₹503 to ₹514), noticed OBV climbing through it, confirmed there was no single freak volume day driving the line, and treated the day-10 close at ₹518 as a breakout worth acting on, with a stop placed back inside the range. That’s the method, not a prediction; the numbers in the example are invented.

For how OBV fits next to RSI, MACD, moving averages and chart patterns, our pillar guide to technical analysis for Indian stocks ties the toolkit together.

The short version

OBV is a running total of volume, added on up closes and subtracted on down closes. Its absolute value is meaningless; its direction is the whole point. A rising line inside a range hints at accumulation, a matching OBV breakout strengthens a price breakout, and a divergence warns that a trend has fewer participants behind it than the price suggests. It ignores the size of moves, it gets bent out of shape by one-off volume spikes, and it throws false signals. Use it as a second opinion on the chart you’ve already read, and it earns its place.

Disclaimer: This article is for information and education only and is not investment advice. Sahyadri Engineering and every price, volume and OBV figure in the examples are hypothetical and used only to illustrate the method. Technical indicators can and do give false signals. Do your own research and consult a SEBI-registered investment adviser before making any investment decision.