Volume Analysis & Market Breadth
Price tells you what happened; volume tells you whether anyone meant it. Breadth tells you whether the rest of the market agreed. This lesson covers the toolkit for both — from per-stock volume signatures to advance-decline lines and the McClellan Oscillator — with the working rules a serious trader needs to interpret them in real time.
1.3.1Why volume matters more than most charts show
Volume is the count of shares (or contracts) that changed hands in a period. It is the only direct evidence of conviction in a price move. A +2.3% NVDA day on volume 0.5x the 20-day average is a different animal than the same +2.3% on volume 2.4x average — the first is a drift, the second is a participatory event.
Most retail charts compress volume into a tiny histogram at the bottom of the screen, giving it visual weight roughly proportional to its perceived importance. Reverse that mentally. Volume context belongs on the same intellectual footing as price context. The cheapest way to upgrade an entire trading process is to require a volume verdict on every chart you read.
Volume’s information content is not symmetric. Up-volume on rising prices is the cleanest “yes” the market issues. Up-volume on flat or falling prices is distribution — institutions feeding the market without lifting it. Quiet drifts higher with declining volume are tired rallies. Each of these has a different forward distribution.
Price is what they showed up to do. Volume is how many of them showed up. The two together are the entire question.
1.3.2Reading volume on a single name
Three measurements form the daily-bar interpretation toolkit.
1.3.2.1Relative volume (RVOL)
RVOL = today’s cumulative volume at a given time of day, divided by the average of the same time-of-day across the prior 20 sessions. RVOL of 1.0 is normal. RVOL 1.5–2.0 deserves attention. RVOL 3.0+ means something is happening — news, a flow event, or a regime move. NVDA running at 2.4x RVOL 30 minutes into the session almost always finishes the day with a |return| well above its 20-day average.
1.3.2.2Volume signature on the bar
Combine the day’s price action with the day’s volume. The combinations matter:
| Price action | Volume | Interpretation |
|---|---|---|
| Strong up day | RVOL 1.5+ | Genuine demand; trend healthy |
| Strong up day | RVOL < 0.8 | Drift / short-cover; suspect |
| Tight inside day | RVOL < 0.8 | Healthy continuation pause |
| Strong down day | RVOL 1.5+ | Real distribution |
| Strong down day | RVOL < 0.8 | Profit-taking; not a regime break |
| Wide-range outside day | RVOL 2.0+ | Reversal candidate; respect direction of close |
1.3.2.3Volume by price (VbP)
Volume-by-price is a horizontal histogram showing how much volume changed hands at each price band. It reveals high-volume nodes (HVN) where significant inventory was acquired — these become magnets and supports — and low-volume nodes (LVN) where price moved through quickly and tends to do so again. SPY’s VbP for 2026 shows a thick HVN around $701–$705 from the early-year consolidation and an LVN through $695–$700. Pullbacks tend to slice through the LVN and stabilise at the HVN.
1.3.3VWAP, OBV and the volume toolkit
1.3.3.1VWAP — the institutional anchor
Volume-weighted average price is the average fill an institutional algorithm would have achieved if it traded proportionally to volume. It is the single most-watched intraday line by professionals because most institutional execution mandates benchmark to VWAP. NVDA trading $118.40 with session VWAP at $117.85 has a long lean; algos are below their target and likely to keep buying. The opposite case — price below VWAP and the line declining — generally means continued algorithmic selling pressure into the close.
1.3.3.2OBV — the running tape
On-Balance Volume is the cumulative sum of volume signed by daily direction (+volume on green days, −volume on red days). It tracks net flow into a name. OBV making new highs while price merely consolidates is constructive accumulation; OBV rolling over while price holds is silent distribution. Joe Granville’s original 1963 formulation is a crude tool, but the direction of OBV remains a useful confirming signal.
1.3.3.3Money flow oscillators
Money Flow Index (MFI) is essentially RSI weighted by dollar volume. Chaikin Money Flow normalises the buying-pressure-versus-selling-pressure question to [-1, +1]. These add nuance over OBV and are worth a small slot on a working chart, but stacking three at once is double counting.
1.3.4Market breadth: the index lie detector
An index level is a weighted average. SPY at all-time highs can mean almost any combination of underlying participation, from 90% of names cooperating to 5 mega-caps doing all the work. Breadth metrics measure the participation directly.
1.3.4.1The advance-decline line
A daily count of (advancing issues − declining issues) on the NYSE, cumulated over time. Healthy markets see the A-D line rising in step with the index. The classic warning is the index making new highs while the A-D line trails the prior high — narrow-leadership rallies.
1.3.4.2Percent above moving average
The fraction of S&P 500 components trading above their 50-day or 200-day SMA. Above 70% on the 50-day is robust; below 30% is washed-out and contrarian-bullish historically. Below 20% on the 200-day, recovery rallies have averaged +15% over the next 12 months looking back to 1990.
1.3.4.3The McClellan Oscillator
A 19/39-day EMA spread of net advances. Bounded readings: above +50 is overbought, below −50 is oversold. Crucially, McClellan tends to lead price at turns by a few sessions, since deteriorating breadth precedes index price weakness in most non-crisis sell-offs.
1.3.4.4Market Participation Index (MPI)
A composite breadth gauge used across this Academy. It blends percent above 50-day, A-D line slope, and net new highs. Readings above 65 indicate broad participation; below 35 indicates narrow tape. An MPI of 78 in early 2026 reflected the post-Q4 rally where over 82% of S&P names were above their 50-day. Use MPI as a regime filter for trend strategies.
1.3.5Breadth divergences and regime change
The most actionable use of breadth is divergence detection.
1.3.5.1The classic warning
Index makes new high; advance-decline line, percent-above-200, or net-new-highs all fail to confirm. The market is being held up by a narrowing roster of leaders. This was textbook in late 1999, late 2007, and late 2021. None of these printed at the high; the divergences appeared 1–3 months before the eventual top. They are warnings, not timing tools.
1.3.5.2The bullish divergence
The mirror image. Index makes a new low while breadth fails to confirm. A-D line holds above its prior trough; net new lows shrink. This appeared in March 2009, December 2018, and October 2022. Selling exhaustion typically precedes price by 2–6 weeks.
1.3.6Worked example: the August 2024 unwind
On 5 August 2024 the JPY-carry-trade unwind hit US equities. SPY closed −2.99%, NDX worse. What did volume and breadth say in the run-up?
- Two weeks prior: SPY printed marginal new highs while the A-D line failed to confirm; percent-above-50 had peaked at 71% three weeks earlier and was at 61%.
- Three sessions prior: McClellan Oscillator turned negative on rising volume.
- The day itself: SPY RVOL 2.6x; the VIX printed an intraday high of 65.73; the McClellan reached −102 by the close.
Lessons. First, the deteriorating breadth gave a warning two weeks early — enough time to tighten or hedge. Second, when the unwind came, RVOL 2.6x and a McClellan −100+ are exhaustion-territory readings; selling into that print historically pays poorly. Within four weeks SPY had recovered the entire decline.
1.3.7Common mistakes and how to fix them
1.3.7.1Confusing volume with attention
Symptom: a stock trends on social media and you assume the chart’s volume spike is meaningful demand. Fix: distinguish RVOL on rising price (demand) from RVOL on falling price (panic). Headlines without price confirmation are noise.
1.3.7.2Treating breadth as timing
Symptom: “breadth diverged, so I shorted.” Fix: divergences identify regime risk, not timing. Use them to reduce size, tighten stops, raise hedges — not to flip aggressively short.
1.3.7.3One-day breadth verdicts
Symptom: “the McClellan is at −75, sell.” Fix: oscillator extremes are mean-reverting most of the time. Persistence — multiple consecutive sessions at extremes plus price confirmation — is the discipline.
1.3.7.4Volume blindness on new highs
Symptom: a name breaks to all-time highs and you ignore that RVOL was 0.5x. Fix: a new high without volume is a hollow break. Either wait for confirmation or sit out.
1.3.7.5Index-only watching
Symptom: tracking SPY without ever pulling up an A-D chart. Fix: cultivate the habit — every Friday, three breadth charts on one page (A-D line, % above 50, net new highs). Five minutes a week.
Key takeaways
- Volume is the conviction layer behind every price move. Demand a volume verdict on every chart you read.
- RVOL, volume signature on the bar, and Volume-by-Price form the three-tool toolkit for single-name volume analysis.
- VWAP is the institutional anchor; OBV tracks net flow; money-flow oscillators add nuance but don’t stack three at once.
- Breadth (A-D line, percent above 50/200, McClellan, MPI) reveals what the index level hides about participation.
- Breadth divergences identify regime risk on a multi-week horizon — they are warnings to manage exposure, not timing signals.
- The August 2024 unwind illustrates the typical sequence: deteriorating breadth weeks early, exhaustion readings on the day, and a fast recovery from extremes.