Volume & money flow

Money Flow Index · MFI

A volume-weighted RSI on a 0–100 scale — overbought and oversold with participation baked in.

Works best in ranging marketsEngine-computed on a fixed sample series
145120962080Above 80 = overboughtBelow 20 = oversold
MFI 92.06How to read MFI on the chart — the callouts mark what to look for.

The formula

Take the typical price (high, low and close averaged) times volume to get each bar's raw money flow. Sum the flow on up-days (positive) and down-days (negative) over the lookback (14), divide them for the money-flow ratio, then squash it onto a 0–100 scale exactly like RSI.

MFI = 100 − 100 ÷ (1 + Positive Flow ÷ Negative Flow)
Worked example
Positive flowNegative flowRatioMFI
60,00030,0002.0066.7

MFI = 100 − 100 ÷ (1 + 2.00) = 66.7

What it is

The Money Flow Index, or MFI, is often described as a volume-weighted RSI: it is a momentum oscillator bounded between 0 and 100 that measures buying and selling pressure while giving weight to how much volume accompanied each move. Where RSI looks only at price changes, MFI folds volume into the calculation, so a rally on heavy volume registers more strongly than the same rally on thin volume. This makes MFI a gauge of conviction as well as direction — it answers not just whether price rose, but whether money genuinely flowed in to lift it. Because volume amplifies the swings, MFI tends to reach its overbought and oversold extremes more decisively than RSI, and its divergences from price carry extra weight. Traders use it to spot overextension, to confirm the strength behind a trend, and to catch tops and bottoms where the volume behind price quietly dries up.

How it is calculated

The calculation starts with the typical price of each bar, the average of the high, low, and close. That typical price is multiplied by the bar's volume to produce the raw money flow, a dollar-weighted measure of the period's activity. Each bar's money flow is then classified as positive if its typical price is higher than the prior bar's typical price, or negative if lower. Over a lookback period, conventionally 14 bars, the indicator sums all the positive money flow and all the negative money flow separately, then forms the money flow ratio by dividing positive by negative. Finally that ratio is squeezed into a 0-to-100 scale using the same formula RSI uses — 100 minus 100 divided by one plus the ratio — so the output shares RSI's familiar bounded structure but is driven by volume-weighted flows rather than pure price change.

Reading it, step by step

MFI is read against its 0-to-100 scale with tighter extremes than RSI: readings above 80 mark an overbought condition where buying pressure has been intense, and readings below 20 mark an oversold condition where selling has dominated. Because volume amplifies the calculation, these extremes are reached more emphatically than on RSI, so an 80-plus print signals a genuinely stretched market. The midline near 50 acts as a rough divide between net buying and net selling pressure and can serve as a pullback reference in a trend. The most valuable read, though, is divergence: when price makes a new high but MFI makes a lower high, the advance is running on shrinking money flow and lacks the volume to continue, a bearish warning; the reverse at a low is bullish. Because it blends volume, MFI divergence frequently appears before the same divergence shows up on a price-only oscillator like RSI.

Reading the signals on the chart

145120962080
MFI 92.06The ▲/▼ marks flag the most recent crossings of the 20 and 80 lines — the classic oversold / overbought signals.

Best timeframes

  • Scalping1m – 5mperiod 9–14
  • Day trading5m – 15mperiod 14
  • Swing1H – daily
  • PositionDaily – weekly

MFI needs reliable volume, so it is strongest on liquid instruments with clean volume data.

MFI vs related tools

MFIRSIChaikin Money Flow
Uses volumeYesNoYes
Bounded 0–100YesYesNo
Overbought / oversold80 / 2070 / 30None
Best signalDivergenceDivergenceZero cross

Common price-action setups

How the signal typically plays out on the chart.

Oversold reclaim

In a range MFI turns up from below 20 and reclaims it — buy the reclaim, stopping under the swing low that printed the oversold reading.

Buy 20 reclaim
Mean-reversion bounce
Overbought rollover

MFI rolls back under 80 in a range — sell or trim, with a stop above the swing high that made the overbought print.

Sell 80 rollover
Pullback lower
Bearish divergence

Price makes a new high while MFI makes a lower high — the advance lacks the volume to continue; exit as MFI turns down. MFI divergence often leads plain RSI.

Exit divergence
Distribution warning

Best timeframes and settings

The default 14-period setting is the standard across daily, weekly, and intraday charts, mirroring RSI's convention and offering a balanced view of the recent flow of money. Shorter lookbacks such as 9 or 10 make MFI more sensitive and cause it to hit the 80 and 20 zones more often, which suits active intraday trading but produces more false extremes. Longer lookbacks like 21 smooth the line and reserve the extremes for more significant moves, favoring swing and position trading. Some traders adjust the overbought and oversold thresholds to 90 and 10 in strongly trending markets to avoid premature signals, or to 70 and 30 in quiet ranges. The essential requirement is reliable volume data, so MFI performs best on centralized, high-volume instruments and degrades where volume is fragmented, synthetic, or simply unavailable.

When and where to use it

MFI is at its best in ranging or mean-reverting markets, where fading its moves out of the 80 and 20 zones lines up with price rotating between support and resistance. In trending markets its role shifts from a fade signal to a warning system, where divergences flag when a trend is losing its volume support. It is especially useful on liquid stocks, index ETFs, and futures where volume is meaningful and centrally reported. It should be used cautiously or avoided on instruments with poor or decentralized volume data, such as some spot forex feeds, because the volume weighting that gives MFI its edge becomes unreliable. As with any oscillator, overbought does not mean sell in a powerful uptrend, so MFI is best deployed as a confirmation and divergence tool rather than a mechanical reversal trigger.

Strategies that use it

In a range-bound market, the classic strategy fades the extremes: sell or short when MFI turns down out of the overbought zone above 80 near resistance, and buy when it turns up out of the oversold zone below 20 near support, targeting the opposite side of the range. In a trending market, the divergence strategy is more appropriate — you hold a trend position but treat a clear MFI divergence against a new price extreme as a cue to tighten stops or scale out ahead of a likely reversal. A third strategy uses failure swings and midline behavior: an uptrend that keeps finding support near the 50 line on MFI pullbacks confirms healthy demand, and a break well below 50 warns the flow is turning. Each approach benefits from waiting for MFI to actually turn out of an extreme rather than shorting a rising overbought reading or buying a falling oversold one.

Combining it with other indicators

MFI is naturally compared with RSI, and running the two together is instructive: because MFI leads at tops and bottoms when volume is the tell, an MFI divergence that RSI has not yet shown can be an early edge, while agreement between the two strengthens confidence. Pairing MFI with a trend filter such as a moving average or ADX tells you whether to fade its extremes in a range or respect its divergences in a trend. Support and resistance or Bollinger Bands give price context, so an MFI extreme that coincides with a band tag or a key level is far more actionable. Other volume tools like OBV or Chaikin Money Flow corroborate the accumulation-or-distribution story MFI is telling. Candlestick reversal patterns appearing at an MFI extreme add a precise, bar-level trigger to the oscillator's broader warning.

Where it fails

The cardinal mistake with MFI is treating overbought as an automatic sell signal in a strong uptrend, where the indicator can pin near 80 or 90 for a long stretch while price keeps climbing, punishing anyone who shorts each extreme. Its dependence on volume is a double-edged sword: on instruments with fragmented, synthetic, or missing volume, the calculation is corrupted and the readings mislead. Divergences can persist far longer than expected, so an early MFI divergence is a warning to manage risk, not a precise timing signal to reverse. In very quiet or illiquid conditions, a single large-volume bar can swing the money flow ratio disproportionately and produce a distorted reading. As with every oscillator, it can whipsaw in a directionless market, so its signals should be filtered by trend context and confirmed by price action rather than traded blindly at the thresholds.

A worked example

Imagine an index ETF that has rallied for weeks and now pushes to a fresh high at 420, with the 14-period MFI reading 78 — strong but just shy of overbought. A week later price edges to a new high of 424, yet MFI now reads only 68, forming a distinctly lower high even as price is higher; the rally is being made on lighter money flow, a textbook bearish divergence. Because you are long, you take this as your cue to tighten your trailing stop and stop adding. Over the next few sessions price stalls, then breaks lower, and MFI slides through 50 and heads toward the oversold zone, confirming that selling pressure has taken over. Your tightened stop exits the position near 419, capturing most of the advance, and you note that the MFI divergence appeared a full session before the same signal became visible on the price-only RSI, illustrating why the volume weighting can give an earlier read at a top.

Common mistakes

  • Shorting a strong uptrend just because MFI is above 80 — it can pin at the extreme while price runs.
  • Trading every 20 / 80 touch with no regard for trend or structure.
  • Using it on instruments with fragmented or synthetic volume, where the reading is unreliable.
  • Treating overbought as an automatic sell instead of context.
  • Ignoring divergence, which is MFI's highest-value read.