RSI vs Money Flow Index
The relative strength index compares recent gains with recent losses using price alone. The money flow index performs the same comparison with each bar weighted by its volume, so it is the only difference between them and it is entirely a data question.
One of these is the other with volume in the formula. That makes the choice unusually clean: it is decided by your data rather than by your preferences.
What each one is
The relative strength index compares recent gains with recent losses using price alone, on a bounded scale. The relative strength index covers it.
The money flow index runs the same comparison with volume weighting. Each bar’s contribution is scaled by how much traded, so a large move on heavy volume counts for more than the same move on light volume. The money flow index covers it.
Everything else about them matches. Same bounded scale, same thresholds, same readings, same failure modes in a trend.
Where they differ
Whether volume enters the calculation. That is the entire difference, and it means the two disagree exactly on bars where volume was unusual relative to the move.
What a disagreement between them means. A price move on light volume reads weaker on the weighted version, which is a genuine second piece of information — provided the volume figure is trustworthy.
Where each is usable. On a centrally traded instrument with published volume, both work. On a spot currency pair, where the reported figure covers one venue out of many, only one of them is reading something real.
How much can go wrong. A price-only tool cannot be misled by a bad volume feed. A weighted one can, and the failure is invisible because the output still looks like a normal oscillator.
Where they agree
They agree most of the time. On bars where volume is unremarkable, the weighting changes very little and the two lines sit close together.
Both are bounded and both pin in a trend. A sustained move holds either at an extreme, and reading that as exhaustion is the standard error with both.
Both fail in a range. In sideways trade the swings are brief, and short runs cross any threshold repeatedly on either, whether the input is price alone or price weighted by volume.
And neither contains a stop. A single busy session can print a bar several times wider than usual, and no oscillator level anticipates that. The invalidation belongs at structure rather than at an oscillator level.
Which one to use: RSI or Money Flow Index?
Run the money flow index when your volume feed is real. A centrally cleared instrument publishes its volume, and on those the weighting is a genuine addition rather than a decoration.
Run the money flow index when you care whether a move had participation. A rise on light volume is the exact situation the weighting exists to flag.
Run the relative strength index on anything with a partial volume feed. A weighting computed from a fraction of the trading is worse than no weighting, because it looks authoritative and is not.
And run one, not both. They share the price input entirely, so agreement is close to automatic and tells you nothing you did not have.
Why the data question decides it
Because a bad input produces a confident wrong reading. The weighted oscillator does not know its volume figure is partial; it weights by whatever it is given and prints a normal-looking line.
And because the error is silent. There is no marker on the chart distinguishing a reading built from complete volume from one built from a slice of it.
How to check your volume feed
Compare the figure across two data providers. If they disagree materially on the same instrument and the same bar, you are looking at venue-specific volume rather than total volume.
Ask what is actually being counted. On an exchange-traded instrument it is the exchange’s own trades. On a spot currency pair it is whatever your broker saw, which is a small share of the market.
Watch what happens around a known event. Volume that barely moves through a major release is a sign the feed is not capturing the trading that actually occurred.
And if you cannot answer any of that, use the price-only version. Open, high, low and close are recorded the same way by every data vendor, so price data of consistent quality is available everywhere. Volume data of matching quality is not, particularly for currencies and for instruments that trade across many venues.
The original data
The money flow index appears in 51 titles at a median of 3,514 across 43 channels, and the relative strength index in 594 at a median of 5,377.
51 videos on the weighted version at 3,514 against 594 on the price-only one at 5,377. A twelfth of the coverage and a smaller audience per video — the version with more information in it is the one almost nobody makes, which is usually a sign the extra input is harder to rely on than it sounds.
The answer to the question on that chart depends entirely on your feed. On a centrally traded instrument that is a real observation about participation; on a spot pair it is an artifact of partial data — and the chart looks identical either way.
How the videos on each are made
In length there is little between them, with medians of 8.9 minutes (RSI) and 8.9 minutes (money flow index). Of the money flow index videos, 88% remain under 50,000 views, against 77% of the RSI videos. Titles shaped as a lesson, meaning how to, tutorial, explained and the like, account for 57% of the money flow index set and 57% of the RSI set. 10 titles name the two together; their median is 5,877 views.
When it fails
The failure is running the weighted version on an instrument with a partial volume feed, and nothing tells you it has happened. The oscillator prints a normal line, crosses thresholds, and produces divergences that look exactly like the ones on a properly fed chart. Every one of them is weighted by a figure describing a fraction of the trading. Decisions are made with apparent extra information, and the extra information was noise from one venue.
The second failure is running both together. They share the price input.
A third is reading an extreme as exhaustion. It means recent bars were one-sided.
A fourth is comparing readings across instruments with different feeds. Not comparable.
A fifth is tuning the length after a losing run. That is fitting.
And a sixth is assuming volume is total volume. On most instruments it is not.
Related
The relative strength index covers the price-only version. The money flow index covers the volume-weighted one. And volume analysis covers whether the input can be trusted at all.
This is the rare comparison where one tool is straightforwardly the other plus something. Whether the something helps is not a question about indicators at all — it is a question about whether the volume figure on your instrument describes the whole market or a slice of it.
— Michael Whitman
This page is educational, not financial advice. Test every idea on your own charts before risking money.