Money Flow Index (MFI)

When a stock is falling hard, brave buyers keep trying to catch the bottom and keep getting cut. When a stock is rocketing up, others keep shorting the top and get squeezed. Both are guessing at the same thing: the moment the pressure runs out - the point where the last frightened seller has sold, or the last excited buyer has bought.

The famous tool for spotting that stretch is RSI, but it has a hole: it looks only at price and ignores how much trading carried the move. So how do you measure exhaustion in a way that demands real volume before it cries "stretched", instead of being fooled by a quiet drift?

The Money Flow Index (MFI) exists to patch the blind spot in RSI. RSI (the Relative Strength Index) is a momentum oscillator scaled from 0 to 100 that measures how fast and how far price has moved recently - high readings are called overbought, low readings oversold. Its weakness is that a stock can inch higher on almost no trading and still trip an "overbought" RSI, generating a signal with nothing behind it.

MFI is, in one line, a volume-weighted RSI. It runs the same 0-to-100 idea but feeds volume into the calculation, usually over 14 days. Because it needs real trading to reach an extreme, MFI produces far fewer empty signals. When MFI says a stock is stretched, it is telling you that a meaningful amount of money - not just a thin drift of price - pushed it there. This is volume as the footprint applied to momentum: the reading only goes to an extreme when a crowd, not a handful of trades, drove the move.

The scale reads like RSI's:

• Above 80: overbought - heavy buying has stretched the stock.
• Below 20: oversold - heavy selling has stretched it the other way.
• The middle: no strong pressure either way.

A quick example makes the difference from RSI concrete. Suppose two stocks each rise 8% over a fortnight. The first rises on volume running well above its own average - buyers pressing every day. The second rises on volume drifting below average - a slow float upward while most participants sit out. Price-only RSI might read both as similarly overbought, because it only sees the 8%. MFI pulls them apart: the first pushes toward 80, the second barely leaves the middle, because the second move had almost no real trading behind it. Same price change, two very different footprints - and MFI is the one that notices.

There are three ways MFI earns its keep, and one way it constantly fools beginners.

1. The zones. Above 80 is overbought; below 20 is oversold. The deeper the reading, the more stretched the move - an MFI of 90 or of 10 is an extreme, not a routine number.
2. The exhaustion curl (a bottom). When a stock is falling and MFI drops below 20 - especially below 15 - the selling volume has reached an extreme. The signal is not the low number itself; it is when MFI then curls back up through 20 while price shows a sign of life. That curl says the heavy selling has eased.
3. The euphoria extreme (a top). When a stock goes vertical and MFI pushes past 80 or 90, the buying is maxed out. There may be very little fresh money left to keep it going, and a sharp pullback often follows.
4. Divergence. As with the other volume tools: price makes a new low but MFI makes a higher low (refusing to go as deep as before), suggesting the selling pressure is drying up even as price slips.

The figure sets a price move above its MFI, with the 20 and 80 lines marked, so you can see stretch as distance into the shaded zones.

Walk one composite chart. illustrative

A mid-cap has gone vertical on hype, running from about ₹100 to ₹180 in three weeks. The last few candles below are near-parabolic. A latecomer feels the fear of missing out and buys near ₹178. A reader checks MFI instead: it reads about 92 - an extreme that says almost every buyer who was going to pile in already has. There is little fresh money left to lift it further. The reader does not "know" it will fall, and does not short into a screaming uptrend on the number alone. But they refuse to be the last buyer, and they watch for the first sign the buying has stalled.

On the very next candle the stock stops making progress and closes weak; the one after, it drops sharply toward ₹150, then ₹134. The MFI extreme did not schedule the fall - the top candle even poked slightly higher first - but it correctly marked a stock with no room left in the tank. The reader who treated 92 as "the tank is full", rather than "buy" or "short blindly", was reading it honestly.

MFI cannot tell you when a stretched move will turn - only that it is stretched. In a strong trend, a stock can stay overbought for a long time (that is what strong trends look like), and in a real decline it can stay oversold while bleeding lower. An extreme reading shifts the odds of a turn; it never sets the date.

It is especially dangerous in the hands of anyone hunting a single magic setting. Change the lookback from 14 days and the "reliable" 20/80 triggers shift; test enough thresholds on old charts and you can always find one that would have called the turns. A curve of MFI rules tuned to fit the past is describing history, not predicting the future. There is no threshold that works everywhere.

And, again, it is drawn from price and volume, not from the company. A deeply oversold MFI tells you selling was heavy, not that the business is worth buying. So the discipline holds: never buy the low number by itself. Wait for the curl and a price pivot, place your floor at the low of the exhaustion, and if price breaks that floor on fresh heavy selling, the exhaustion was a false alarm - exit.

The classic trap is trusting a price-only oscillator and ignoring what the volume-weighted one is saying.

Two more slips: treating "below 20" as an instant buy (it is a zone to watch, not a trigger), and forgetting that in a strong trend an overbought reading can persist for weeks - shorting every 80 print in a bull run is a fast way to lose money.