OptiNod Academy

Money Flow Index (MFI) — When Volume Fails to Confirm Price

MFI weighs momentum by volume to measure the weight of money behind a move. When price rises but MFI fails to keep up, it can show capital leaving before RSI does.

The Money Flow Index (MFI) measures the amount of money behind each candle by first calculating its typical price, then multiplying that value by volume. Typical price is the average of the high, low, and close: (high + low + close) / 3. If the typical price is higher than the previous candle's typical price, that candle's money flow, typical price multiplied by volume, is classified as positive money flow. If it is lower, it is classified as negative money flow. MFI then divides the 14-period sum of positive money flow by the 14-period sum of negative money flow and converts the result into a value between 0 and 100. If the denominator is zero, MFI is 100. If positive and negative money flow are balanced, it sits near 50.

Most traders think of MFI as RSI with volume attached. RSI measures only changes in closing price, while MFI multiplies that movement by volume, so it is often treated as a slightly more refined version of RSI. Traders also tend to view the 80/20 thresholds as the only real difference from RSI's 70/30 levels, then use both as essentially the same type of momentum indicator.

That reading misses the point of MFI. Multiplying by volume means momentum is being weighed by the amount of money behind it. If price rises on weak volume, MFI values that same rise much lower than RSI does. This is why MFI divergence often appears earlier and more forcefully than RSI divergence when price makes a new high but money flow fails to confirm it.

Multiplying by volume means measuring the weight of money

The single difference between RSI and MFI is volume weighting. RSI only asks how much the close rose. Whether a 1% up candle traded three times normal volume or one-third normal volume, RSI sees the same 1% gain. MFI multiplies that 1% rise by volume, so a low-volume 1% rise counts as small positive money flow, while a high-volume 1% rise counts as large positive money flow.

The intuition becomes clear when you break the calculation down candle by candle. On November 21, 2024, BTC's daily candle had a high of $98,988, a low of $94,040, and a close of $98,317. Its typical price was the average of those three values, $97,115. Multiplying that by the day's volume of 69,228 gives money flow of roughly $6.7 billion for that candle. Because the typical price rose from the previous candle, that $6.7 billion is classified as positive money flow. MFI sums positive and negative money flow over 14 periods and converts their ratio into a single number showing which side the money is leaning toward, and by how much. The larger the volume on a candle, the greater its impact on that ratio.

This difference matters most when price rises but volume does not follow. When fewer buyers are pushing price higher, RSI still reads the move as upward momentum, but MFI gives that rise less weight because the money behind it is smaller. A rise without volume is a rise without meaningful capital inflow, and MFI is designed to show exactly that.

The difference was clear when BTC moved above $90,000 in November 2024. On November 21 and 22, BTC closed at $98,317 and $98,892, respectively, with strong volume behind the advance. MFI climbed to 89.2, well above the 80 threshold. Price and money flow were moving in the same direction. It was an overheated move, but one supported by capital.

MFI divergence appears earlier and carries more weight than RSI divergence

One month later, when BTC made an even higher price high, MFI and RSI told very different stories. On December 16 and 17, BTC pushed its high to $108,353 and marked a new all-time high. That was more than 9% above the November high near $99,000. Looking at price alone, the trend appeared even stronger.

But at that new high, the two indicators diverged. RSI was still near its overbought threshold at 70.1, so the chart still looked like a strong advance. MFI, however, was only 60.3 at the same new high. That was 29 points below its November high of 89.2. Over the same span, RSI fell only 12 points, from 82.5 to 70.1. Price made a higher high, but MFI cooled by more than twice as much as RSI.

The difference came from volume weighting. The candles that produced the December high had noticeably lower volume than the November advance. RSI does not look at volume, so it registered only the price rise and stayed near 70. MFI reflected the weaker money behind the move and dropped below 60. Traders watching only RSI divergence were slow to notice the capital outflow because the 70 overbought reading masked it. Traders watching MFI saw earlier that money was not confirming the new high. BTC failed to reclaim that December high, moved sideways through January, and fell below $90,000 in February.

80 and 20 are not hard thresholds in trending markets

MFI's 80/20 levels are commonly treated as overbought and oversold thresholds: above 80 is overbought, below 20 is oversold. In range-bound markets, that interpretation often works. Money flow shifts to one side, then reverts toward balance as MFI moves between 80 and 20.

In strong trends, the story changes. When money keeps flowing in one direction, MFI can stay pinned above 80 for a long time. Selling simply because MFI crossed 80 can make traders miss the largest part of an early trend and take losses on the wrong side of the market.

BTC's move from $47,000 to $52,000 in February 2024 was one of those periods. After MFI crossed above 80 and reached 90.9 on February 14, it stayed above 80 for nine trading days through February 22, holding between 82 and 97 the entire time. During that same period, BTC's close rose further from $51,795 to $52,259. Treating 80 as an overbought sell signal would have meant fighting the advance for nine straight trading days. In a strong capital inflow phase, 80 is not a reversal threshold. It is a sign that the trend is still active.

SOL showed the same pattern. From October 22 to October 29, 2024, SOL's daily MFI stayed above 80 for eight consecutive trading days and reached 94.2 on October 24. Over the same period, SOL's close rose from $167 to $179. MFI can stay pinned above 80 during sustained capital inflows across assets. Using 80 as a trading threshold only makes sense in ranges. In trending markets, the more important question is how long MFI remains above 80, because that shows the strength of money flowing in.

Volume can make MFI more sensitive than RSI, or less sensitive

Volume weighting cuts both ways. On high-volume candles, MFI reacts more sharply than RSI. On low-volume candles, it reacts more slowly. The same indicator can behave in two different ways depending on the volume profile.

When volume explodes, MFI can move to extremes faster than RSI. On December 5, 2024, BTC touched the area near $109,000 but sold off into a $96,946 close. Volume reached 109,922, more than twice the prior average. That high-volume decline entered the calculation as negative money flow, dropping MFI from 52.4 to 36.3 in a single day. RSI fell only 5 points, from 67.6 to 62.8. On a high-volume reversal candle, MFI cooled much faster.

In a thin, low-volume consolidation, the opposite happens: MFI becomes less responsive. Both positive and negative money flow shrink, so the ratio itself becomes less meaningful. This is why MFI can drift around 50 without much signal during thin year-end or early-year trading. Even with the same 14-period setting, MFI's response speed changes with the distribution of volume.

MFI is unreliable in low-volume assets

Because MFI uses volume as an input, distorted volume produces distorted MFI. The most common trap is a small altcoin with thin trading.

In low-volume assets, one or two large orders can account for a large share of the day's volume. Depending on the direction of that single order, MFI can swing sharply. Despite the name money flow, the move often comes from one random large order inflating positive money flow and pushing MFI above 80. If the asset is exposed to exchange wash trading, fake volume is counted as money flow as well, making the signal even less reliable.

Assets immediately after listing or after a major event have the same problem. First-day trading volume can be dozens of times higher than normal, then fall back within a few days. While that abnormal first-day volume remains inside the 14-period window, MFI can lean heavily to one side. An MFI reading before the volume distribution has stabilized does not show the true direction of money. It only reflects a single volume spike.

The filter is volume stability. Use MFI divergence as a signal only in assets with sufficiently large average daily volume and moderate day-to-day variation. If 24-hour volume is abnormally high relative to market cap, or if volume swings tenfold over a few days, treat MFI as secondary context, not as a reason to enter a trade.

MFI divergence sell setup

When price makes a new high late in a trend but MFI does not confirm it, it warns that price is rising while money is leaving. Still, do not sell immediately on divergence alone. Enter only when price structure also breaks down.

• [ ] Trend condition: BTC's daily chart has been in an uptrend for at least the previous month, and price closes above the prior high to mark a new high.
• [ ] Divergence condition: MFI at the new high is at least 15 points lower than MFI at the prior high, for example a prior high near 89 versus a new high in the 60s.
• [ ] Volume check: The candle that makes the new high has lower volume than the candle that made the prior high.
• [ ] Entry: Sell at the close of the candle that closes below the swing low immediately before the new high.
• [ ] Stop: Place the stop above the new high.
• [ ] Invalidation: If price closes back above the new high after the divergence, assume money has returned and drop the short thesis.

The key is to identify the divergence, then wait for price structure to break. Selling immediately at a new high just because MFI diverges can lead to losses when money returns and the trend continues, as it did in February 2024. Divergence is a warning that money is leaving. The downside break in price structure is the point where that warning is confirmed.

Use RSI alongside MFI to gauge the strength of a divergence

It is hard to gauge the strength of MFI divergence without comparing it with RSI divergence. Put both indicators on the same screen, and the type of divergence becomes clearer.

When MFI and RSI diverge at the same time, both price momentum and money flow are cooling, making the signal stronger. When RSI does not diverge but MFI does, price momentum is still holding, but the volume behind the advance is fading. This is where MFI's volume weighting matters most. The December 2024 BTC all-time high was exactly this case. RSI stayed near 70, so its divergence was mild, but MFI dropped below 60 and showed capital leaving first.

The interpretation is straightforward when the two indicators disagree. If only MFI cools, the weakness is in volume. Price may still have enough momentum to attempt one or two more new highs, but the money needed to support those attempts is shrinking. In that position, it is safer to hold off on new longs and watch whether price structure breaks. MFI adds the layer RSI cannot see: volume. When the two indicators point to different readings at the same new high, the gap between them shows how much money has left the move.