SPY climbed steadily from $589.22 on 3 February 2025 to $604.17 on 19 February, a clean uptrend on the price chart. But look at the volume structure. On 3 February, the close of $589.22 sat well above the midpoint of the day’s range ($582.05 to $591.71, midpoint $586.88). Buying pressure dominated. By 20 February, the close of $601.65 sat below the midpoint of the $598.34-$602.93 range (midpoint $600.64). The close was drifting toward the lower half of the range even as price pushed higher. Chaikin Money Flow would have picked up this shift. The indicator was weakening while price was strengthening. Two days later, SPY began its drop from $601.57 to $576.68 over the next five sessions.
Chaikin Money Flow (CMF) measures buying and selling pressure by examining where the close falls within the high-low range of each session, then weighting that position by volume. If the close is near the high, the session’s volume is classified as buying pressure. If the close is near the low, it is selling pressure. CMF sums this over a lookback period (typically 20 or 21 days) and normalizes the result. The output oscillates between -1 and +1, though readings beyond -0.25 or +0.25 are rare.
I use CMF as a confirmation tool. It tells me whether volume is supporting the price move or contradicting it. When price makes a new high and CMF confirms with a positive reading, the trend has volume behind it. When price makes a new high and CMF is declining or negative, something is wrong under the surface.
The Formula
CMF has three components. The first is the Money Flow Multiplier, which measures where the close sits within the high-low range:
\text{MFM} = \frac{(\text{Close} - \text{Low}) - (\text{High} - \text{Close})}{\text{High} - \text{Low}}This simplifies to:
\text{MFM} = \frac{2 \times \text{Close} - \text{High} - \text{Low}}{\text{High} - \text{Low}}The MFM ranges from -1 (close equals the low) to +1 (close equals the high). If the close is at the midpoint of the range, the MFM is 0.
The second component is Money Flow Volume, which weights the MFM by the session’s volume:
\text{MFV} = \text{MFM} \times \text{Volume}The third component is CMF itself, which sums MFV over a lookback period and divides by total volume:
\text{CMF}(n) = \frac{\sum_{i=1}^{n} \text{MFV}_i}{\sum_{i=1}^{n} \text{Volume}_i}The default lookback is 20 or 21 periods. The division by total volume normalizes the result, so CMF is not biased by absolute volume levels. A high-volume stock and a low-volume stock can have the same CMF reading if their close positions within the daily range are similar.
What CMF Tells You
CMF above zero means buying pressure has dominated over the lookback period. More volume occurred on sessions where the close was in the upper half of the range. Buyers were in control, pushing price toward the high before the close.
CMF below zero means selling pressure has dominated. More volume occurred on sessions where the close was in the lower half of the range. Sellers were winning, pushing price toward the low.
The magnitude matters. CMF at +0.05 is technically positive but barely. CMF at +0.20 is strongly positive, meaning closes have consistently landed near the highs on heavy volume. I treat readings between -0.05 and +0.05 as neutral. The money flow is balanced. Neither buyers nor sellers are in clear control.
Zero-Line Crossovers
The simplest CMF signal is the zero-line crossover. When CMF crosses above zero, buying pressure is taking over from selling pressure. When it crosses below zero, the reverse.
NVDA from late February into March 2025 showed this dynamic. After the earnings-driven selloff, NVDA closed at $130.24 on 24 February, then fell to $120.11 on 27 February and $114.02 on 3 March. Volume was enormous: 443 million shares on 27 February, 411 million on 3 March. And the closes were near the lows. On 27 February, the close of $120.11 was near the bottom of the $119.97-$134.97 range. That is a deeply negative MFM for that session (roughly -0.98), multiplied by massive volume. CMF would have been deeply negative during this period.
The turn came gradually. On 7 March, NVDA closed at $112.65 in a $107.52-$113.44 range. The close was in the upper half. On 12 March, it closed at $115.71 in a $112.85-$116.73 range. Again, upper half. On 14 March, it closed at $121.64 in a $118.12-$121.85 range. Nearly at the high. CMF would have been improving through this sequence as closes shifted from the lower to the upper part of the daily range, even though price was still well below the February highs. The zero-line crossover from negative to positive, if it occurred during this recovery, would have confirmed that buying pressure was genuinely replacing selling pressure.
I do not trade zero-line crossovers in isolation. They are confirmation, not triggers. If price is breaking above resistance and CMF crosses above zero at the same time, the breakout has volume support. If price breaks above resistance but CMF stays below zero, the breakout is suspect.
Divergence Signals
CMF divergence is the most valuable signal the indicator produces. It occurs when price moves in one direction but CMF moves in the other, indicating that volume is not supporting the price trend.
Bearish divergence: price makes a higher high, but CMF makes a lower high or turns negative. This means the new high in price was not accompanied by the same quality of buying pressure as the previous high. Closes are drifting toward the lower half of the range on the rally, even though the highs are increasing. This was the pattern in the SPY example from the opening. Price was making higher highs into 19 February, but the close position within the daily range was deteriorating.
Bullish divergence: price makes a lower low, but CMF makes a higher low or turns positive. This means the selloff is losing its grip. Closes are starting to land in the upper half of the range even though prices are still falling. The selling pressure is exhausting itself.
AAPL from late February into March 2025 showed potential bullish divergence forming. Price dropped from $246.03 on 24 February to $208.77 on 13 March. But look at the close positions. On 10 March, AAPL closed at $226.49 in a $223.25-$235.14 range. The close was in the lower third (MFM approximately -0.45). On 12 March, it closed at $216.04 in a $213.98-$220.79 range. The close was slightly below the midpoint (MFM approximately -0.40). On 14 March, it closed at $212.56 in a $208.67-$213.02 range. The close was near the high (MFM approximately +0.79). Even as price kept making lower lows ($226.49, $216.04, $212.56), the close position within the daily range was shifting toward the highs. That shift would produce improving CMF readings. The price was saying “lower,” but the volume structure was saying “sellers are losing conviction.”
CMF vs OBV
On-Balance Volume (OBV) and CMF both measure the relationship between volume and price, but they use different mechanics.
OBV is binary. If the close is higher than the previous close, the entire session’s volume is added to the cumulative OBV line. If the close is lower, the entire volume is subtracted. There is no middle ground. A session where the close is $0.01 higher than the previous close gets the same volume credit as a session where the close is $5 higher.
CMF is proportional. It does not care about close-to-close direction. It cares about where the close sits within the high-low range of each individual session. A session with a higher close than yesterday but a close near the low of its own range (a long upper wick, for instance) would be positive for OBV but negative for CMF.
This makes CMF more granular. It captures the intra-bar struggle between buyers and sellers. OBV only sees the outcome (up or down). CMF sees the quality of the outcome (did buyers push the close to the high, or did it close near the low despite finishing up?). I use both, but when they disagree, I give more weight to CMF for identifying divergences and to OBV for confirming trends.
CMF and the Accumulation/Distribution Line
CMF is directly derived from the Accumulation/Distribution Line (A/D Line). The A/D Line is a running cumulative sum of Money Flow Volume. CMF takes that same MFV calculation but sums it over a fixed window (20 periods) and normalizes by volume. This makes CMF a bounded oscillator rather than a cumulative line.
The A/D Line grows without bound over time, making it difficult to compare across stocks or across timeframes. CMF solves this by producing a reading between -1 and +1. You can compare CMF readings on AAPL and NVDA directly. You cannot do this with the A/D Line because the absolute values depend on the stock’s entire trading history.
When CMF and the A/D Line agree (both rising or both falling), the volume signal is strong. When they diverge (CMF declining while the A/D Line is rising), it usually means recent sessions are showing weaker buying pressure than the longer-term trend. The A/D Line carries history; CMF reflects the current window.
Settings and Lookback Period
The standard CMF lookback is 20 or 21 periods. This aligns with roughly one month of trading on a daily chart. Shorter lookbacks (10-14) make CMF more responsive but noisier. Longer lookbacks (30-40) smooth the signal but can delay zero-line crossovers by several sessions.
I use 21 periods on the daily chart for swing trading. It captures about a month of volume data, which is enough to identify sustained buying or selling pressure without reacting to every two-day fluctuation. For shorter-term trading, a 10-period CMF picks up shifts faster but produces more false crossovers.
The lookback also determines how sensitive CMF is to single high-volume days. With a 21-period lookback, one massive volume day still contributes only one-twenty-first of the window. With a 10-period lookback, it contributes one-tenth. If you trade around earnings or other high-volume events, a shorter lookback will reflect those events more prominently in the CMF reading.
Where CMF Falls Short
CMF uses the high-low range, which means it is sensitive to wicks. A session with a very long upper wick (high far above the close) will produce a negative MFM even if the close was up on the day. This can be misleading in volatile markets where intraday ranges are driven by short-lived spikes rather than sustained directional pressure.
CMF also struggles in low-volume environments. When volume is thin, the MFV calculation is dominated by small numbers, and the resulting CMF can swing erratically. Stick to liquid instruments where daily volume is consistent. Penny stocks and thinly traded ETFs produce unreliable CMF readings.
In strongly trending markets, CMF can stay positive (or negative) for extended periods without providing actionable signals. The indicator simply confirms what you already know: the trend is intact. It adds value primarily at inflection points (divergences, zero-line crossovers) and breakouts (confirmation), not during the middle of a trend.
CMF does not distinguish between different types of volume. A large block trade by a single institution and thousands of small retail orders produce the same volume figure. The MFM weights them identically. Volume Profile and Anchored VWAP provide a more structural view of where volume concentrated at specific price levels, which CMF cannot do.
Putting CMF on Your Chart
CMF is a subpanel indicator that oscillates around a zero line. Most platforms display it as a histogram or a line. I prefer the line version because it makes divergences easier to spot visually. Plot it below the price chart with horizontal reference lines at +0.10, -0.10, and the zero line.
The primary use case: when you see a breakout from a consolidation pattern, check CMF. If it is above zero and rising, volume is supporting the breakout. If it is below zero or declining, the breakout lacks volume conviction and may fail. This one check has saved me from more false breakouts than any other confirmation tool.
Combine CMF with a momentum indicator for timing and a price structure tool for levels. CMF tells you whether money is flowing into or out of the stock. The price chart tells you where to act on that information. Together, they answer two different questions: “Is the move real?” (CMF) and “Where do I trade it?” (price structure).
Educational content only. Not investment advice. Trading involves risk. You are responsible for your decisions.