reference · Technical analysis
Money Flow Index: Price and Volume Classification
Money Flow Index classifies price-times-volume by whether typical price rose or fell, then transforms the positive-to-negative flow ratio onto a zero-to-100 scale.

Define typical price and raw flow
Typical price is (high + low + close) / 3. Raw money flow is that price multiplied by the bar's volume. A rise in typical price classifies the current raw flow as positive; a fall classifies it as negative. Sum each category over the chosen window, divide positive by negative, then calculate MFI = 100 − 100 / (1 + ratio). MetaQuotes documents these stages.
The word money can be misleading. The output is an indicator derived from prices and a volume field, not a cash-flow statement tracking deposits, withdrawals or participant portfolios. With quote-event counts instead of transaction quantity, raw flow does not even represent a traded cash amount.
Compute an original short example
Suppose the prior typical price was 9. The next three typical prices and volumes are 10 with 100, 11 with 100, and 10 with 50. The first two observations rise relative to their predecessors and contribute positive raw flows of 1,000 and 1,100. The third falls and contributes negative raw flow of 500.
The positive sum is 2,100 and negative sum is 500, so the ratio is 4.2. MFI is 100 − 100 / 5.2, approximately 80.7692. This three-observation calculation is deliberately small enough to inspect; it is not a suggested strategy parameter.
Now imagine the falling observation had volume 200 instead of 50. Its negative raw flow would be 2,000, the ratio would be 1.05 and MFI approximately 51.2195. The typical-price sequence is unchanged, but different recorded volume changes the oscillator substantially.
Handle ties and limiting cases
When typical price is unchanged, specify whether the implementation assigns zero to both directional categories. If negative flow is zero while positive flow exists, the limiting transformed value is 100; the reverse case approaches zero. If both sums are zero, the calculation needs an explicit missing or neutral convention.
Ensure the lookback has enough earlier price data to classify its first flow observation. Comparing its typical price only with the first price inside the window can accidentally discard an otherwise valid directional observation. This is a different warm-up issue from the indicator's displayed range.
Use the result as a description
Upper and lower threshold labels are historical conventions. A high MFI can persist and does not prove that identifiable investors are about to sell. A divergence between price and MFI also depends on the chosen peaks, window and confirmation timing; those choices need to be fixed before evaluating later outcomes.
Review OBV's simpler signed-volume rule and the source of the volume field. Then use an unseen evaluation period if testing an MFI-based condition. A bounded score should not be mistaken for an estimated probability of a winning trade.
Questions and answers
Is MFI a probability?
No. Its zero-to-100 scale is a mathematical transformation of classified price-volume sums, not a calibrated chance of a future price move.
Does MFI prove institutions are buying?
No. The inputs do not identify participants or their holdings. It classifies activity using changes in typical price.
Sources and further checks
Use the current source for your exact instrument, account and platform. Referencing a general specification does not establish support for every TradeCopier workflow.
- MetaQuotes: Money Flow Index · Checked September 19, 2026
Found an error? Send a correction with this page's address and a primary source. See our editorial standards for how we handle examples, claims and revisions.

