03 · Volume-Price Analysis

Moving averages, MACD, and KDJ are all processed from "price", but volume is the only raw data collected directly from the market without any mathematical processing. That is why technical analysis has an old saying: "Price can lie; volume hardly can." The core task of volume-price analysis: use volume to judge whether the current rise/fall in price is "real".

1. Volume Basics

1.1 What Volume Is

Volume = the total quantity (or value) traded within a period. It is a direct measure of "participation":

• A market with expanding volume = many participants, wide disagreement, thorough turnover;
• A market with shrinking volume = few participants, much watching, narrow disagreement.

1.2 Expansion / Contraction / Stacking

Stacked volume sketch:
volume │    █
      │   ██
      │  ███      ← consecutive expansion, stacking like stairs
      │ ████
      │█████
      └──────────→ time

Core understandings:

• Volume always needs "a comparison target": against the recent average (expansion/contraction) or against historical extremes (blow-off/ground volume) — an isolated absolute number is meaningless.
• Expansion itself carries no direction — an expanding rise and an expanding fall are both expansion; price decides direction, volume decides "how real this move is".

2. The Eight Basic Volume-Price Relationships

Combining "price direction (up/down/flat)" with "volume change (up/down)" yields the eight basic volume-price relationships. This is the core working table of volume-price analysis — memorize it and review your charts against it.

Three iron rules of the eight relationships:

1. A rising price needs volume running alongside: only expanding rises are real; shrinking rises don't travel far.
2. Avoid expanding declines: heavy-volume falls mean selling isn't finished — don't rush to bottom-fish; the first shrinking-volume stabilization after an expanding decline is the observation point.
3. Every reversal presumes "a break on volume": reversals at bottoms and tops almost always come with violent volume changes — volume is the reversal's "blood test report".

3. Volume Confirmation on Breakouts

3.1 The Three Elements of an Effective Breakout

Effective breakout = price breaks a key level (neckline/box/prior high/low)
                   + volume expansion (usually ≥ 1.5× recent average volume)
                   + holding the break (no fall back inside the key level for 2–3 days)

• Breakout on expansion: large real buy/sell orders flood out; a retest that holds after the break = effective breakout, trend launched.
• Breakout on contraction: the break lacks volume support; most likely a false breakout or bull/bear trap, followed by a fall back into the range.
• Shrinking-volume retest after the breakout: contraction on the retest is good (only profit-takers cashing out, no panic sellers); a retest that holds near the key level = second chance to get on board.

3.2 Typical Signs of a False Breakout

4. Top and Bottom Divergence

Divergence = price direction and volume direction disagree. When price and volume "disagree", it is usually price that is lying.

4.1 Top Divergence (Volume-Price, Bearish)

Pattern: Price makes a new high, but volume (or OBV) fails to make a new high in step — instead it keeps shrinking.

price     ↗  ↗
        ╱╲ ╱╲   ← each peak higher than the last
volume  ╱  ╲╱
      ╱╱╱╱    ← each peak lower (volume not following) → top divergence

Meaning: the higher it climbs, the fewer followers; the rise is "hollow" and can flip at any time. Common at the end of a markup wave after consecutive rallies, or in the spike phase when good news cashes out.

4.2 Bottom Divergence (Bullish)

Pattern: Price makes a new low, but volume (or OBV) fails to make a new low — it may even expand.

price    ╲  ╲
        ╲╲ ╲╲   ← each trough lower than the last
volume  ╲
       ╱╱╲      ← expansion at the bottom / contraction to the limit (can't fall further) → bottom divergence

Meaning: the lower it falls, the more money steps in; the decline is "hollow" and can reverse at any time. More reliable when combined with bottom patterns like long lower wicks or dojis.

4.3 Three Commandments for Using Divergence

1. Divergence is a segment, not a point: divergence only shows "momentum exhaustion" — it cannot tell you "the exact day of reversal". Price continuing to new highs/lows after a divergence is the norm.
2. The bigger the timeframe, the more reliable: daily divergence > minute-chart divergence. Minute-chart divergences fire a dozen times a day and are worthless.
3. Wait for confirmation: the final confirmation of a divergence is the break of the price trendline/key level. Otherwise, bottom-fishing or top-picking on divergence kills just the same.

5. Volume Profile (VPVR) and High-Volume Nodes

5.1 What the Volume Profile Is

Volume profile (VPVR, Volume Profile Visible Range): stacking the volume traded at each price level horizontally produces a "price–volume" horizontal distribution chart that answers "at which price level does the market's majority hold its cost basis".

price
50 ┤█████████      ← high-volume node (large amounts changed hands here)
45 ┤███
40 ┤█████████████  ← even denser (the primary node)
35 ┤█████
30 ┤██
    └────────── volume

5.2 What High-Volume Nodes Do

5.3 Usage and Pitfalls

• Confirming support/resistance: price finding support at the top edge of a high-volume node = reliable support; conversely, breaking through a node requires volume (freeing trapped positions needs real buying).
• Bottom accumulation shape: after a long decline, the high-level peak gradually migrates down while bottom-side volume concentrates = thorough turnover, a solid bottom.
• Pitfall: the volume profile is estimated from historical trade data (different software uses different estimation algorithms); it is not real position data. Wash trading can fabricate fake peaks. Use it as a "relative reference", never as precise position statistics.

6. Turnover Rate and Activity

6.1 Turnover Rate

turnover rate = today's volume / free float × 100%

It expresses "what fraction of the float changed hands in one day" — the core gauge of activity.

6.2 Key Points for Using Turnover

• High turnover + stalling = distribution: turnover stays high but price can't rise — someone is unloading heavily.
• Low turnover + slow bleed = no buyers: sellers can't even get out; the bleed drags on.
• Sudden turnover jump during sideways action: possibly accumulation or a prelude to a regime change — watch direction.
• Pitfall: the absolute value depends on float size (small caps naturally run high turnover); compare against the instrument's own history, never across instruments.

7. "Smart-Money Flow": Concept and Reality

7.1 What "Smart-Money Flow" Is

Mainstream software classifies executed orders by size (e.g., large orders > 500K, medium, small), defines net large-order inflow (buys − sells) as "smart-money net inflow/outflow", and dresses it up as "the smart money is buying/selling".

7.2 Why You Must Be Wary of This Data

1. "Large order" ≠ "smart money": to hide intent, institutional money splits large orders into small ones for execution (iceberg orders); hot money and quants can deliberately post large orders to create appearances. Classifying "smart money" by order size is essentially guessing.
2. Directional distortion: software counts aggressive large buys as "smart-money inflow" and aggressive large sells as "outflow". But whether one large trade is an "aggressive buy" or "aggressive sell" depends on where the counterparty's order was sitting — the attribution under exchange matching rules is not precise.
3. Easily manipulated by wash trading: the same capital buying and selling to itself (wash trades) can fabricate "smart-money net inflow of X hundred million", and combined with a pump-and-dump script the data becomes a bull-trap tool.
4. Statistical caliber differs by vendor: for the same stock on the same day, different software can show completely opposite "smart-money net inflow" numbers.

7.3 How to Use It Without Getting Burned

• Use it as a sentiment gauge, not as truth: "smart-money net inflow" only reflects "statistics of large-order behavior" — never treat it as "the operator's intent".
• Cross-verify: combine it with price location (net inflow at lows means something completely different from net inflow at highs), turnover rate, and the real volume trend.
• The most reliable "smart-money fingerprints" are volume-price themselves: prolonged low-level stacked accumulation (volume without price rise), shrinking-volume shakeouts before markup (contracting pullbacks) — these shapes are the "fingerprints" of money's behavior, harder than any money-flow number.

8. Trade Distribution and Multi-Timeframe Volume

8.1 The Three Dimensions of Trade Distribution

8.2 Multi-Timeframe Volume Coordination

Multi-timeframe principle: the big timeframe sets direction, the small one finds rhythm

Weekly: does volume support the large-degree trend (expanding rise = big trend healthy)
Daily: is the expand/contract rhythm of this rise/fall healthy
60m / 15m: volume confirmation for short-term breakouts and retests

Typical coordination cases:

• Weekly expansion + daily contracting pullback + minute-chart expanding breakout = three-level resonance, highest reliability.
• Weekly contraction + daily expansion with stalling = large-degree momentum insufficient; the intraday expansion is just a bounce — don't chase.

8.3 Pitfalls of Multi-Timeframe Volume

• When timeframes contradict (daily rising on contraction while weekly prints blow-off volume), defer to the bigger timeframe — small-timeframe volume is just one segment of the big one.
• Don't call 15-minute blow-off volume "blow-off": wrong degree, wrong conclusion all the way down.

Limitations and Pitfalls

1. Volume can lie too: wash trades, matched prints, fake orders, and order-splitting by quants can distort volume. "Price lies, volume hardly does" gets discounted in institution- and quant-dominated markets — volume data itself can be manipulated.
2. "Expansion must rise / contraction must fall" is amateur talk: expansion can break out upward or downward; contraction can be a shakeout or a no-bid void. Volume must be read together with price location.
3. Divergence and volume-price shapes lag the turn: by the time you "see" a top divergence, the operator has usually unloaded the first batch. The value of volume-price analysis is "being fooled less", not "being one step earlier".
4. Turnover rate and volume profiles rely on estimation: float definitions and profile-estimation algorithms both introduce bias; pledges and locked shares blur what "free float" even means.
5. On illiquid instruments every volume-price signal distorts: the volume itself is too small (e.g., dead coins, obscure stocks) — a few trades can fabricate blow-off or ground volume; volume-price relationships are entirely unreliable on such instruments.
6. Volume-price analysis still only describes price behavior: news and fundamental shifts can invalidate volume-price rules instantly (a good-news limit-up on shrinking volume can still chain limit-ups). Volume-price is an "observation tool", not a "cause" — every signal ultimately submits to risk control and trading discipline (see the 07 · Trading Systems↗ chapter).