TL;DR: Chaikin money flow indicator readings weight volume by where each bar closes in its range, so they describe pressure, not future price. On ES and NQ daily bars in the year to November 3, 2026, the 20-day reading sat inside plus or minus 0.10 on most sessions. Most raw zero line crosses reversed within five sessions, and a 0.05 buffer cut the signal count by more than half. Use it as context beside price structure, and treat the Chaikin Oscillator as a separate, faster momentum study.
Price just broke out, and the money flow line under the chart reads minus 0.03. Is that a warning, or noise?
You can only answer that if you know three things: what the number is built from, how often it normally sits there, and what has followed it before. This guide covers all three, with the arithmetic rebuilt from real E-mini S&P 500 (ES) and E-mini Nasdaq-100 (NQ) futures bars.
What the Chaikin Money Flow Indicator Measures
The Chaikin money flow indicator, usually shortened to CMF, is a volume-weighted average of accumulation and distribution over a set number of bars. Marc Chaikin built it on one idea: where a bar closes inside its own high-low range says something about who controlled that bar. A close near the high suggests buyers held the session. A close near the low suggests sellers did.
Each bar gets a score between minus 1 and plus 1 for its close location. That score is multiplied by the bar's volume, and the sum over the look-back window is divided by the total volume in the same window. The result is an oscillator bounded between minus 1 and plus 1, centered on zero.
Those bounds are almost never touched. ChartSchool notes that a 20-day CMF would need 20 consecutive closes on the high to reach plus 1, and that the reading typically moves between minus 0.50 and plus 0.50.
CMF belongs to a family of volume studies that share raw material but answer different questions:
| Study | What it uses | Window | Scale |
|---|---|---|---|
| On-Balance Volume | Whether the close rose or fell against the prior close | None, running total | Unbounded |
| Accumulation/Distribution Line | Close location in the range times volume | None, running total | Unbounded |
| Chaikin Money Flow | Close location times volume, divided by window volume | Usually 20 or 21 bars | Minus 1 to plus 1 |
| Money Flow Index | Typical price change times volume, in an RSI formula | 14 bars by default | 0 to 100 |
| Chaikin Oscillator | 3-day EMA minus 10-day EMA of the Accumulation/Distribution Line | 3 and 10 bars | Unbounded |
The key difference from the Accumulation/Distribution Line is the reset. The line is a running total, so its level depends on where your data starts. CMF divides by the volume inside its own window, so old history washes out and the reading can be compared from one month to the next.
Related lesson: Accumulation distribution line
Chaikin Money Flow Formula Step by Step

The Chaikin money flow formula has three steps, and only the last one has a window in it.
- Money flow multiplier = ((close minus low) minus (high minus close)) divided by (high minus low).
- Money flow volume = money flow multiplier times the bar's volume.
- 20-period CMF = the 20-period sum of money flow volume divided by the 20-period sum of volume.
There is a shortcut for step 1. The multiplier equals twice the close's position in the range, minus 1. A close dead center scores zero, a close at the high scores plus 1, and a close at the low scores minus 1.
A worked ES example from the first ten sessions
The two tables in this worked example rebuild the 20-day reading for ES on November 3, 2026, from daily continuous-contract bars (Yahoo Finance ES=F data, read November 4, 2026). Every column after volume comes from the formula. This first table holds October 7 to October 20, 2026.
| Date | High | Low | Close | Volume | Multiplier | Money flow volume | 20-day CMF |
|---|---|---|---|---|---|---|---|
| 2026-10-07 | 7884.50 | 7815.75 | 7852.75 | 1,399,667 | 0.0764 | 106,884 | 0.0468 |
| 2026-10-08 | 7858.25 | 7783.00 | 7816.25 | 1,700,014 | -0.1163 | -197,676 | 0.0789 |
| 2026-10-09 | 7870.50 | 7817.00 | 7859.75 | 1,130,030 | 0.5981 | 675,906 | 0.0716 |
| 2026-10-12 | 7887.00 | 7846.25 | 7858.00 | 927,661 | -0.4233 | -392,691 | 0.0552 |
| 2026-10-13 | 7864.00 | 7834.50 | 7855.75 | 1,192,792 | 0.4407 | 525,637 | 0.0917 |
| 2026-10-14 | 7879.75 | 7741.00 | 7770.25 | 2,063,500 | -0.5784 | -1,193,484 | 0.0508 |
| 2026-10-15 | 7801.75 | 7717.25 | 7791.00 | 1,909,800 | 0.7456 | 1,423,875 | 0.0868 |
| 2026-10-16 | 7809.50 | 7676.00 | 7704.25 | 2,380,698 | -0.5768 | -1,373,137 | 0.0228 |
| 2026-10-19 | 7773.25 | 7677.00 | 7734.50 | 1,740,657 | 0.1948 | 339,089 | -0.0004 |
| 2026-10-20 | 7791.50 | 7704.50 | 7755.25 | 1,718,580 | 0.1667 | 286,430 | 0.0034 |
The last ten ES sessions and the final reading
The second half of the ES window runs from October 21 to November 3, 2026, with the same columns. The reading on the last row is the one a chart showed for November 3.
| Session | Session high | Session low | Session close | Contracts traded | MF multiplier | MF volume | CMF 20 |
|---|---|---|---|---|---|---|---|
| 2026-10-21 | 7812.25 | 7751.00 | 7800.25 | 1,680,253 | 0.6082 | 1,021,868 | 0.0655 |
| 2026-10-22 | 7798.75 | 7676.75 | 7704.50 | 2,274,209 | -0.5451 | -1,239,630 | -0.0012 |
| 2026-10-23 | 7714.25 | 7648.50 | 7659.75 | 1,791,568 | -0.6578 | -1,178,484 | -0.0681 |
| 2026-10-26 | 7739.50 | 7651.75 | 7714.50 | 1,677,784 | 0.4302 | 721,782 | -0.0230 |
| 2026-10-27 | 7772.75 | 7699.75 | 7758.25 | 1,556,188 | 0.6027 | 937,976 | 0.0204 |
| 2026-10-28 | 7819.50 | 7750.25 | 7783.25 | 1,576,563 | -0.0469 | -73,990 | 0.0631 |
| 2026-10-29 | 7773.75 | 7709.50 | 7754.25 | 1,998,904 | 0.3930 | 785,561 | 0.0819 |
| 2026-10-30 | 7713.00 | 7558.25 | 7614.00 | 2,539,896 | -0.2795 | -709,858 | 0.0460 |
| 2026-11-02 | 7693.75 | 7570.75 | 7574.50 | 2,002,583 | -0.9390 | -1,880,474 | -0.0295 |
| 2026-11-03 | 7623.75 | 7546.00 | 7596.75 | 1,645,013 | 0.3055 | 502,496 | -0.0261 |
Add the money flow volume column across all 20 ES sessions (both tables) and you get minus 911,920. Add the volume column and you get 34,906,360. Divide the first by the second and the reading is minus 0.0261, the bottom-right cell.
What one bar can do to the reading
Look at November 2, 2026. ES ranged 123 points and closed 3.75 points above the low, so the multiplier was minus 0.9390 and almost the whole session's 2,002,583 contracts went in as selling pressure. The reading dropped from 0.0460 to minus 0.0295 that day, crossing zero on one bar (the bar that left the window changed the sum too, but far less).
That is the core mechanic to keep in mind. CMF is a net figure, and a single high-volume bar that closes at an extreme can flip its sign. Before reading meaning into a cross, check which bar did the work.
How to Read Chaikin Money Flow Readings
The basic interpretation of a Chaikin money flow reading is simple. Above zero, more volume landed in the upper half of the bars in the window than the lower half, weighted by how far up. Below zero, the reverse. ChartSchool frames a positive reading as confirming an uptrend and a negative reading during an uptrend as a question mark over its strength.
The absolute level matters too. A reading of 0.25 says something different from 0.02, even though both are positive. But there is no fixed overbought or oversold level for CMF the way there is for an RSI-style study, because the bound of plus or minus 1 is almost never reached. You have to know what normal looks like on the market you trade.
What ordinary Chaikin money flow looks like on ES and NQ
Here is the 20-day reading's own distribution over the 252 completed sessions from November 3, 2025, to November 3, 2026, on daily ES and NQ continuous-contract bars:
| Measure of the 20-day reading | ES | NQ |
|---|---|---|
| Lowest reading | -0.2703 | -0.1600 |
| 10th percentile | -0.0786 | -0.0612 |
| 25th percentile | -0.0069 | -0.0184 |
| Median | 0.0561 | 0.0607 |
| 75th percentile | 0.1219 | 0.1660 |
| 90th percentile | 0.2417 | 0.2912 |
| Highest reading | 0.5134 | 0.5058 |
| Median absolute reading | 0.0763 | 0.0841 |
| Sessions above zero | 179 of 252 (71.03%) | 171 of 252 (67.86%) |
| Sessions inside plus or minus 0.05 | 81 of 252 (32.14%) | 89 of 252 (35.32%) |
| Sessions inside plus or minus 0.10 | 155 of 252 (61.51%) | 143 of 252 (56.75%) |
| Sessions beyond plus or minus 0.25 | 26 of 252 (10.32%) | 39 of 252 (15.48%) |
| Sessions beyond plus or minus 0.50 | 2 of 252 | 2 of 252 |
| Longest run on one side of zero | 69 sessions, positive | 59 sessions, positive |
Three things stand out. First, a reading of 0.10 is close to the middle of the range on both contracts, not a strong signal. Second, the line sat above zero on roughly seven sessions in ten during a year when both indices rose, so "positive" on its own told you little. Third, readings past plus or minus 0.50 happened twice a year, so the usual "typical range" is really the outer edge.
It also helps to know how much cancelling goes on under the reading. Over those 252 ES sessions, the money flow volume added up to 192,357,074 in absolute terms but only 23,714,491 net, about 12.33% of the gross. Most days offset each other. The reading is the small residue.
A practical reading habit follows from this. Compare the current value with your instrument's own percentiles rather than a textbook threshold. A reading in the top tenth of the past year is unusual. A reading of 0.06 is just Tuesday.
Related lesson: Money flow index
Chaikin Money Flow Settings and Period Choice
The two standard Chaikin money flow settings are 20 and 21 periods. ChartSchool describes the look-back as typically 20 or 21 days and uses 20 as its chart default. Fidelity names 21 as the standard period, and TradingView calls 20 and 21 the most popular choices.
One bar of difference sounds cosmetic. It is not always. On ES, the 20-day and 21-day readings disagreed on the sign of the reading on 21 of the 252 sessions measured. On NQ it was also 21. On November 3, 2026, NQ read minus 0.0579 at 20 periods but plus 0.0100 at 14 periods and plus 0.0172 at 60. Same bars, three different settings, opposite answers.
| Period | ES zero line crosses in 252 sessions | NQ zero line crosses in 252 sessions | ES reading on November 3, 2026 | NQ reading on November 3, 2026 |
|---|---|---|---|---|
| 14 | 39 | 44 | -0.0165 | 0.0100 |
| 20 | 35 | 27 | -0.0261 | -0.0579 |
| 21 | 27 | 33 | -0.0143 | -0.0515 |
| 60 | 3 | 6 | -0.0617 | 0.0172 |
Shorter windows generally cross zero more often, and the 60-day line crossed only a handful of times. The relationship is not perfectly neat (NQ's 21-day line crossed more than its 20-day line), which is a good reminder that small samples wobble.
How to choose: pick the period that matches the holding horizon of the decision you are using it for, then keep it fixed. ChartSchool shows a 20-day and a 60-day CMF on one chart to read short and longer money flow together, a reasonable way to see whether the two horizons agree. The trap is changing the period until the chart says what you hoped. If you catch yourself doing that, the setting is not the problem.
Chaikin Money Flow Trading Strategies Tested on Zero Line Crosses
Most Chaikin money flow trading strategies fall into four buckets:
- Zero line cross. Treat a move above zero as buying pressure taking over, and below zero as selling pressure taking over.
- Buffered cross. ChartSchool suggests a bullish threshold slightly above zero (plus 0.05) and a bearish threshold slightly below (minus 0.05) to cut whipsaws.
- Trend confirmation. Stay with an uptrend while CMF holds positive, and question it when CMF turns negative.
- Divergence. Price makes a new high while CMF makes a lower high (or the reverse at lows), hinting that pressure behind the move is fading.
The zero line cross is the most repeated instruction, so it is worth testing. These counts are this guide's own calculations on Yahoo Finance daily ES=F and NQ=F continuous-contract data, 252 sessions to November 3, 2026. The audit below counts every cross of the 20-day reading on ES and NQ over the same 252 sessions, then checks whether price was higher (after a bullish cross) or lower (after a bearish cross) 10 and 20 sessions later. A cross counts as reversed if the next cross came within five sessions.
| Zero line rule | ES signals | ES agreeing after 10 sessions | ES agreeing after 20 sessions | ES reversed within 5 sessions | NQ signals | NQ agreeing after 10 sessions | NQ agreeing after 20 sessions | NQ reversed within 5 sessions |
|---|---|---|---|---|---|---|---|---|
| Raw cross of zero | 35 | 13 of 32 (40.62%) | 12 of 30 (40.00%) | 24 of 34 (70.59%) | 27 | 8 of 24 (33.33%) | 11 of 24 (45.83%) | 15 of 26 (57.69%) |
| Cross of plus or minus 0.05 | 10 | 2 of 8 (25.00%) | 3 of 7 (42.86%) | 2 of 9 (22.22%) | 11 | 3 of 10 (30.00%) | 3 of 10 (30.00%) | 1 of 10 (10.00%) |
| Cross of plus or minus 0.10 | 5 | 2 of 5 (40.00%) | 2 of 5 (40.00%) | 0 of 4 | 6 | 2 of 6 (33.33%) | 2 of 6 (33.33%) | 0 of 5 |
Zero line cross results with and without the buffer
In this guide's own count of 20-day CMF zero line crosses on Yahoo Finance daily ES and NQ data, the 0.05 buffer did exactly what ChartSchool says it does: it cut whipsaws. Raw crosses on ES reversed within five sessions 70.59% of the time. With the 0.05 buffer, the reversal rate fell to 22.22%, and the number of signals dropped from 35 to 10.
What the buffer did not do is make the signals right more often. None of the rules reached even half agreement with the direction of price 10 or 20 sessions later, on either contract. Part of that is the year itself: both indices trended up, so bearish crosses kept failing. Part of it is that a cross describes the past 20 bars, and the market does not owe the next 20 anything.
The counts are small (10 and 11 buffered signals), so read this as a warning about sample size and whipsaw cost, not as proof that CMF crosses can never work. The honest takeaway is narrower. A zero line cross tells you recent closes shifted. It does not tell you the next move has started, and treating those as the same thing is where most of the whipsaw losses come from.
Divergence needs a confirmation step
Divergence is the other popular CMF strategy. Fidelity describes a sell setup when price makes a higher high while CMF makes a lower high and starts falling, and the mirror image for buys. ChartSchool adds a useful caution: a bullish divergence while CMF is still negative only shows less selling pressure. It takes a move into positive territory to show actual buying pressure.
That turns divergence into a two-step rule. The divergence is the observation. The cross back through zero (or through your buffer) is the confirmation. Acting on the first step alone means acting on fading pressure that may keep fading.
Using Money Flow to Confirm Breakouts and Time Exits

Breakout confirmation with Chaikin money flow
Fidelity's guidance is to wait for CMF to confirm a breakout: if price breaks up through resistance, wait for a positive CMF reading before trusting the move. It is a sensible idea. The question is whether it filters much.
To test it, this guide counted (on Yahoo Finance daily ES=F and NQ=F data) every session that closed above the highest close of the prior 20 sessions (a simple breakout), then check price 10 sessions later:
| Breakout test, 252 sessions to November 3, 2026 | ES | NQ |
|---|---|---|
| Closing breakouts above the prior 20-session high close | 40 | 38 |
| Breakouts with CMF above zero | 35 | 35 |
| All breakouts higher 10 sessions later | 24 of 40 (60.00%) | 24 of 38 (63.16%) |
| Breakouts with positive CMF higher 10 sessions later | 22 of 35 (62.86%) | 23 of 35 (65.71%) |
| Breakouts with negative CMF higher 10 sessions later | 2 of 5 | 1 of 3 |
| Any session higher 10 sessions later (base rate) | 133 of 242 (54.96%) | 125 of 242 (51.65%) |
The confirmation rule passed almost every breakout, because a close at a 20-session high usually comes after a run of strong closes, which is exactly what pushes CMF positive. On ES, 35 of 40 breakouts already had positive money flow. The filter removed five breakouts, and the remaining group did only a few points better than the unfiltered one.
So the useful version of the rule is the exception, not the confirmation. A breakout with negative CMF is the rare case worth a second look, because price made a new high while the closes inside the window leaned toward the lows. With only 5 and 3 such cases here, that is a prompt for caution, not a statistic.
Chaikin money flow as an exit warning
CMF can also serve as an exit prompt on a position that is already working. The logic: if you are long in an uptrend and CMF slides from a high reading toward zero, the closes have started landing lower in their ranges. That does not mean the trend is over. It means the pressure that carried it has eased.
The temptation is to bail the moment the line dips, then watch the trend resume without you. A steadier approach is to tighten an existing plan rather than invent a new one mid-trade. For example, a trader might move a stop to the most recent swing low when CMF falls below its own 25th percentile while long, instead of exiting on the cross itself. The stop still decides the exit. CMF only decides when to look harder.
Pairing Chaikin money flow with RSI and trend filters
ChartSchool recommends using CMF with a pure price oscillator such as MACD or RSI, and its sample scans pair CMF turning positive with RSI moving above 50. The logic is that CMF measures where closes land, while RSI measures how far price has moved. When both agree, two different inputs say the same thing.
The simplest pairing is a trend filter. Use price structure or a moving average to decide direction, and use CMF only to ask whether recent closes support that direction. Do not stack three volume studies built from the same multiplier and count them as three confirmations. They are one input wearing three labels.
Related lesson: Trend following
Why Price Gaps Fool the Money Flow Multiplier

The money flow multiplier only looks inside one bar. It never checks the previous close. TradingView flags this as the shortfall in the calculation, and ChartSchool shows how a stock can gap down, close sharply lower and still add buying pressure if the close lands above the middle of the day's range.
On index futures this is not an edge case. These counts are this guide's own calculations on Yahoo Finance daily ES=F and NQ=F continuous-contract data, 252 sessions to November 3, 2026.
| Multiplier against the close-to-close change | ES | NQ |
|---|---|---|
| Sessions where the multiplier's sign contradicted the close change | 41 of 252 (16.27%) | 43 of 252 (17.06%) |
| Sessions that opened below the prior close | 116 | 116 |
| Of those, sessions still scored as buying pressure | 71 (61.21%) | 72 (62.07%) |
| Sessions that opened above the prior close | 132 | 133 |
| Of those, sessions still scored as selling pressure | 67 (50.76%) | 60 (45.11%) |
A real ES gap example
March 3, 2026, is a clean ES case. The prior session closed at 6888.25. ES opened at 6883.50, fell to 6718.75 and closed at 6824.75, down 63.50 points (0.92%). Because that close sat above the middle of the 167.25-point range, the multiplier was plus 0.2676, and the session's 2,259,359 contracts added about 604,522 of positive money flow. A down day went into the reading as buying pressure.
That is not a bug in your platform. It is the formula working as designed. CMF asks where the bar finished inside its own range, not how far price traveled from yesterday. Once you read it that way, the contradiction stops being confusing and starts being information: sellers drove price well below the open at some point in the session, and buyers took part of that back before the close.
Roll gaps in continuous futures data
Futures add a second gap problem. A continuous chart switches from the expiring contract to the next one, and the two trade at different prices. On NQ, the continuous series jumped from a 25,132.79 close on December 19, 2025, to a 25,635.25 open on December 22, 2025. December 19 was the third Friday of December, when the December equity index contracts expire, so most of that jump reflects the switch to the March contract rather than a real overnight move.
The volume shows the same seam. ES volume in the continuous series fell to 590,558 on December 17 and 416,394 on December 18, 2025, from 1,070,240 on December 16, consistent with a series still tracking the expiring contract while traders had already rolled. CME treats the Monday before the third Friday as the equity index roll date, when the next month becomes the lead month. Bars from that week understate the real activity, which drags on the CMF denominator and numerator alike.
Chaikin Oscillator vs Chaikin Money Flow
The Chaikin Oscillator is a different study built from the same raw material. It takes the Accumulation/Distribution Line (the running total of money flow volume) and subtracts its 10-day exponential moving average from its 3-day exponential moving average. ChartSchool describes it as applying the MACD formula to the Accumulation/Distribution Line, which makes it an indicator of an indicator, at least three steps removed from price.
So CMF asks, "Over the last 20 bars, did volume lean toward the highs or the lows?" The Chaikin Oscillator asks, "Is the accumulation line speeding up or slowing down?" One is a level. The other is momentum.
| Chaikin Oscillator measure, 252 sessions to November 3, 2026 | ES | NQ |
|---|---|---|
| Highest value at 3 and 10 | 2,960,028 | 1,102,214 |
| Lowest value at 3 and 10 | -2,064,127 | -634,478 |
| Median absolute value at 3 and 10 | 642,511 | 224,867 |
| Zero line crosses at 3 and 10 | 31 | 33 |
| Zero line crosses at 6 and 20 | 15 | 23 |
| Same sign as 20-day CMF at 3 and 10 | 195 of 252 (77.38%) | 179 of 252 (71.03%) |
| Same sign as 20-day CMF at 6 and 20 | 219 of 252 (86.90%) | 212 of 252 (84.13%) |
Chaikin oscillator settings and scale
ChartSchool notes that the default 3 and 10 settings produce a line that frequently crosses zero, and suggests doubling both to 6 and 20 to smooth it. The measurements agree: on ES the crosses fell from 31 to 15, and agreement with the 20-day CMF rose from 77.38% to 86.90%. Smoothing buys agreement and costs speed.
The bigger trap is scale. The oscillator has no bounds, so its values follow the instrument's volume. ChartSchool's sample scans use plus 1000 and minus 1000 as thresholds just above and below zero for stocks. On ES, the oscillator closed above plus 1000 on 162 sessions and below minus 1000 on the other 90. It was never inside that band. A threshold copied from a stock scan filtered nothing on an index future.
If you want a comparable scale across markets, use CMF. If you want an early read on shifts in accumulation, use the oscillator, and judge its levels only against its own history on the same contract.
Related lesson: On balance volume
Limits of the Chaikin Money Flow Indicator as a Standalone Signal
The Chaikin money flow indicator describes the past accurately, because that is what the formula does. The harder question is whether it tells you anything about what comes next. TradingView's own write-up says CMF should not necessarily be used by itself as a stand-alone signal, and ChartSchool says the same about all indicators.
A simple test: sort every session by its 20-day reading into five equal groups, from lowest to highest, and check where price was 20 sessions later. These counts are this guide's own calculations on Yahoo Finance daily ES=F and NQ=F continuous-contract data, 252 sessions to November 3, 2026.
| Fifth by 20-day reading | ES reading range | ES sessions | ES higher 20 sessions later | ES average 20-session change in points | NQ reading range | NQ sessions | NQ higher 20 sessions later | NQ average 20-session change in points |
|---|---|---|---|---|---|---|---|---|
| Lowest fifth | -0.2703 to -0.0278 | 46 | 37 (80.43%) | 187.91 | -0.1600 to -0.0329 | 46 | 35 (76.09%) | 1153.50 |
| Second fifth | -0.0274 to 0.0284 | 46 | 27 (58.70%) | 56.52 | -0.0319 to 0.0218 | 46 | 24 (52.17%) | 204.95 |
| Middle fifth | 0.0286 to 0.0772 | 47 | 31 (65.96%) | 85.30 | 0.0218 to 0.1057 | 47 | 16 (34.04%) | -141.34 |
| Fourth fifth | 0.0778 to 0.1508 | 46 | 20 (43.48%) | -35.77 | 0.1068 to 0.2093 | 46 | 24 (52.17%) | -105.60 |
| Highest fifth | 0.1538 to 0.5134 | 47 | 31 (65.96%) | 128.33 | 0.2141 to 0.5058 | 47 | 28 (59.57%) | 882.49 |
| All sessions | All readings | 232 | 146 (62.93%) | 84.65 | All readings | 232 | 127 (54.74%) | 398.56 |
In this guide's forward test on Yahoo Finance ES and NQ data, if high money flow predicted higher prices, the right-hand column would climb from the lowest fifth to the highest. It does not. The lowest fifth actually did best on both contracts, because in a rising year the weakest readings tended to mark pullbacks that later recovered. The rank correlation between the reading and the next 20 sessions of price was minus 0.0983 on ES and minus 0.0934 on NQ, and a 2,000-shuffle permutation test put the chance of seeing a link that strong by luck at 0.136 and 0.1645. In plain terms: no reliable relationship.
Structural limits of Chaikin money flow
Chaikin money flow has five built-in limits that no setting removes:
- It lags. Every input is a finished bar, so the reading reports pressure already spent.
- It ignores gaps. The multiplier never sees the prior close, so gap days can score against their own direction.
- It depends on the volume source. Continuous futures data, session boundaries and partial-day volume all change the denominator.
- It struggles in chop. ChartSchool notes CMF is less reliable when a trend flattens or volatility rises, because the line flips repeatedly without follow-through.
- Sample limits apply to any test. The ES and NQ tests in this guide cover 252 sessions on two correlated index futures. A different year, market or timeframe can look different, which is a reason to run the same check on what you trade.
None of this makes the study useless. It makes it a context tool, and that is a fair job for it. CMF is good at answering "have recent closes leaned toward the highs or the lows, on how much volume?" It is poor at answering "where is price going?"
Chaikin Money Flow on Futures Contracts and Continuous Data
The Chaikin money flow arithmetic is the same on futures as on stocks, but three inputs behave differently.
- Continuous contracts. A continuous chart stitches contract months together. Volume around roll week can belong to a contract most traders have already left, and the price seam can create a false gap.
- Session boundaries. CME equity index futures trade nearly around the clock, so a daily bar includes overnight trading. A chart built from regular trading hours only will produce different highs, lows and closes, and a different CMF.
- Bar size. CMF works on any bar length, but intraday volume swings with the clock. A 20-bar reading on 5-minute bars mixes the quiet lunch period with the open, so compare it only with the same time of day.
The contract you trade matters for sizing, not for the indicator. ES is $50 per index point and the Micro E-mini S&P 500 (MES) is $5 per index point, so the same CMF reading says nothing about how many contracts fit your risk. The stop distance and the contract's point value decide that.
On our accounts, you can chart CMF on the platform tied to your account: Tradovate, NinjaTrader and TradingView through a Tradovate connection, Tradesea, Quantower, Sierra Chart and R Trader Pro through Rithmic, or WealthCharts. Every position has to be flat by 4:45 PM ET, so a daily money flow reading is background for a trade that opens and closes in the same session, never a reason to hold overnight.
Related lesson: What is volume
Chaikin Money Flow Indicator FAQ
What does CMF stand for in banking and trading?
In trading and charting, CMF stands for Chaikin Money Flow, the volume study created by Marc Chaikin. In banking, the same three letters are used for unrelated terms that differ by institution, so check the context. On a price chart, CMF almost always means this indicator.
What is a good Chaikin money flow reading?
There is no universal good number. Above zero means recent closes leaned toward bar highs, and below zero means they leaned toward lows. Judge the size of the reading against your instrument's own history, since 0.10 was close to the middle of the range on ES and NQ over the past year.
What is Chaikin's volatility?
Chaikin Volatility is a separate Marc Chaikin study that measures range, not money flow. Fidelity describes it as a 10-day exponential moving average of the daily high minus low, then the percent change in that average over 10 days. It rises when daily ranges widen and says nothing about direction.
How do you read the MFI indicator?
The Money Flow Index runs from 0 to 100 and uses typical price and volume in an RSI-style formula, with 14 periods as the default. Readings above 80 are typically treated as overbought and below 20 as oversold. In strong trends it can stay at those levels while price keeps going.
What is Marc Chaikin's prediction?
Marc Chaikin's indicators make no prediction; Chaikin Money Flow summarizes where recent bars closed in their ranges, weighted by volume. He is also behind the Chaikin Oscillator and Accumulation/Distribution studies that share its input. Any market forecast attributed to Marc Chaikin is separate commentary and should be judged on its own evidence.
Is the Chaikin oscillator better than Chaikin money flow?
Neither is better; they answer different questions. CMF is a bounded level that compares well across markets, while the Chaikin Oscillator measures the momentum of the Accumulation/Distribution Line and reacts faster. The oscillator also crosses zero more often, so it needs more filtering.
Chaikin Money Flow Risk Note
Trading futures involves substantial risk of loss, and leverage means losses can exceed what you expect on a single trade. Indicator readings, including the historical measurements in this guide, describe the past and do not predict future results. Evaluation and funded accounts are simulated trading accounts. This article is educational and is not investment advice or a recommendation to trade any product.

