Learn · The Twelve Measurements — Lesson 3 of 12
Money Flow Index (MFI), explained
A price can rise because buyers are committing real money, or because almost nobody is trading and the few who are happen to be paying up. MFI exists to tell those apart.
What it measures
Like RSI but weighted by volume, so a rise on heavy trading counts for more than a rise on thin trading. It answers whether money is actually moving in, rather than whether price merely drifted.
Mechanically, MFI multiplies each day's typical price by its volume to get a money flow, sorts recent days into positive flow (price rose) and negative flow (price fell), and expresses the balance on a 0–100 scale. The result reads like RSI — but where RSI counts price changes and treats a thin session the same as a heavy one, MFI weighs each session by how much money actually moved through it.
Why it moves prices
Participation is what separates moves that hold from moves that fade. A rally on heavy dollar volume means many participants repriced the asset and hold it at the new level; a rally on thin volume means the price moved through empty space and can move back just as easily. Because MFI is built from exactly this distinction, it tends to diverge from price before drifts exhaust — price grinding higher while MFI falls describes a move running out of buyers even as it makes new highs.
How to read it
High values mean buying volume has dominated recently and the move may be stretched; low values mean selling volume has dominated. Extremes in either direction tend to mean-revert.
The conventional bands are 80 (money has been flooding in; the move is stretched) and 20 (money has been flooding out). As with every oscillator, the bands describe how far a move has travelled — extremes tend to mean-revert, but strong markets can pin the gauge at an extreme for longer than seems reasonable.
How Farlens uses it
Money Flow Index carries 10.3% of the Farlens composite (weight 1.2 of 11.6 across all twelve measurements). Coverage: Needs reliable volume data, so it is unavailable where volume is not reported.
Two conventions matter here. First, a measurement that cannot be observed for an instrument is skipped, never counted as zero — a zero would read as “neutral” and quietly dilute the composite. Second, on any given day a contribution is the measurement's score times its weight, divided by the total weight of the measurements present that day — so the published breakdown always sums to the score beside it. The full stack, with every weight published, is on the signals page.
A worked example
Farlens publishes one frozen example reading — AAPL on 22 July 2026, deliberately weeks old — so the shape of a reading can be studied without opening the live layer. That day the composite printed +14.7, “Mixed”, and this measurement read −63.3.
| Measurement present that day | Reading |
|---|---|
| Trend | +100.0 |
| Advance/decline pressure | +100.0 |
| On-Balance Volume | +97.4 |
| Momentum (RSI) | −75.7 |
| Money Flow Index | −63.3 |
| Sector rotation | −45.3 |
| Volume surge | −29.7 |
| Short volume share | −23.9 |
MFI read −63.3 that day while trend read +100.0 — recent sessions had seen money leaving on the pullback even though the year-scale uptrend was intact. When the fast, participation-weighted gauges disagree with the slow structural ones, the composite prints a genuinely mixed reading, and that disagreement is information in itself.
A real reading, kept deliberately stale. Descriptive of one historical day — not a prediction, and not representative of current conditions.
What it cannot tell you
MFI needs trustworthy volume, so it degrades exactly where volume reporting does — thin listings, some foreign markets, and any instrument where the real trading happens off-exchange. It also cannot say whose money moved: one institution rebalancing and ten thousand retail orders can produce the same flow number with very different implications. And like every oscillator it is bounded — in a relentless trend it saturates and stops adding information until the trend pauses.
Frequently asked
MFI vs RSI — which is better?
They answer different questions: RSI asks how one-sided recent price changes have been, MFI asks how one-sided the money behind them has been. Farlens computes both, at different weights, precisely because their disagreements are informative.
What does an MFI divergence mean?
Price making new highs while MFI declines means each new high is being bought with less money than the last — historically a description of tiring moves, though divergences can persist through long trends.
Does MFI work for crypto?
Yes, where exchange volume is reliable. Farlens computes it for major crypto assets alongside the derivatives measurements.
Continue the course
- Previous — Lesson 2: Derivatives positioning
- Next — Lesson 4: On-Balance Volume
- Course index — all twelve measurements