Momentum indicator (MOM): price change over n bars

The Momentum indicator, usually abbreviated MOM, is the simplest oscillator there is: today's close minus the close n bars ago. With the default of 10, it tells you how many dollars the price has gained or lost over the last two trading weeks. Above zero, price is higher than it was 10 bars ago; below zero, lower. Its slope shows whether that gain is growing or shrinking.

Almost every other momentum tool is a refinement of this idea, which makes MOM a good place to see both what momentum measures and the traps built into it. This guide works through a real Oracle example that shows the "echo" effect in a single pair of numbers, examines what the zero line did during Oracle's 46% decline, and explains how Finabase scores Momentum.

What Momentum measures

Momentum is speed. If a stock has risen $8 over 10 days, MOM is +8. If it then keeps rising but only by $5 over the next 10 days, MOM falls to +5 even though the price is still going up. That is the essential insight behind every momentum indicator: the rate of change slows before the direction changes. A falling MOM above zero says the advance is decelerating.

Because it compares two individual closes, MOM ignores everything in between. Those two closes alone decide the value.

The formula

A worked example on Oracle: the echo effect

Oracle (ORCL) had a violent two weeks in late May and early June 2026. Here are four MOM (10) values from Finabase's chart:

On June 2 MOM hit its high for the year, +$57.97, because the stock had soared from the mid-$180s. Ten sessions later, those same high closes became the comparison point. By June 16 the stock was back at $188.33, and MOM hit its low for the year, −$56.25, measured against June 2's $244.58.

The spike shows up in MOM twice: once going in, once coming out. Whenever MOM makes a large move, look at what happened n bars ago as well as today. Part of the reading may be an echo.

Momentum on a real chart: Oracle 2025–2026

Oracle fell from $256.89 at the end of October 2025 to $139.56 a year later, about 46%, with a low of $114.50 on July 28, 2026, and the brief May–June spike in between. MOM was positive on only 89 of 251 sessions.

Zero-line crossings. MOM crossed zero 29 times in the year. Taking each upward crossing as a buy and the next downward crossing as a sell produced 15 trades. Five made money; the returns summed to about −26%, before costs. The best trade, from the July 30 upward crossing to August 18, gained 11.9% off the summer low. Buy-and-hold over the same year lost 46%, so the rule's habit of stepping aside helped, but it was still the wrong side of a falling market most of the time.

Brief rallies. MOM spent 162 of the 251 sessions below zero, and the excursions above it were short: ten of the fifteen positive stretches lasted fewer than ten sessions before MOM turned negative again. A persistently negative MOM with only brief positive spells is the signature of a steady downtrend, and a better guide to the year than any single crossing.

How traders use Momentum

Zero-line crossings. The classic signal: buy when MOM crosses above zero, sell when it crosses below. It is the same as price crossing above or below its close of n bars ago. Use it with a trend filter, or the whipsaws seen on Oracle will dominate.

Turns from extremes. A MOM that peaks and turns down from a high level shows an advance losing speed, often before price itself turns. Remember the echo, though: a turn can be caused by an old big day leaving the window.

Divergence. A new price high with a lower MOM high (or a new low with a higher MOM low) is the classic momentum divergence.

Comparing periods. A short MOM (5) and a long one (20) on the same chart separate short bursts from sustained trends.

Pitfalls and limitations

Price units. A $10 MOM is large for a $50 stock and small for a $500 one. On Oracle, the same 10-day move in percentage terms produced a much smaller MOM at $120 than at $250. Use ROC to compare.

Two-point sensitivity. Only two closes determine each value. A single unusual close, now or n bars ago, can swing the reading.

No bounds. There are no fixed overbought or oversold levels; extremes are relative to each stock's history.

Comparison table

See Rate of Change (ROC): percentage momentum and how to read it, Chande Momentum Oscillator (CMO): formula and Finabase’s version and MACD indicator: signal line, histogram and divergence for refinements of the same idea.

How Finabase calculates and scores Momentum

Finabase computes Momentum with TA-Lib's MOM function on closing prices. A new MOM study ("Momentum") defaults to 10 periods and is drawn in its own panel around a zero line. The ready-made "Bollinger Bands" chart template pairs it with Bollinger Bands and MACD.

In the technical gauge on the symbol page and in the Scanner, MOM (10) is scored on its zero line: zero or above counts as Bullish (a buy signal), below zero as Bearish (a sell signal), with no neutral reading. The Scanner lists the stocks whose MOM crossed zero, with the date. On Oracle's chart that would have meant 29 signals in a year, so read them as a description of the last two weeks, not a forecast.

Frequently asked questions

What does the Momentum indicator tell you?

How far price has moved over the last n bars, in dollars, and from the slope, whether that movement is speeding up or slowing down.

What is the difference between Momentum and Rate of Change?

MOM is the dollar change over n bars; ROC is the percentage change. ROC can be compared across stocks and price levels; MOM cannot.

What is a good Momentum setting?

Ten periods is Finabase's default and a common choice for daily charts. Twelve, 14 and 20 are also used. Shorter periods react faster and cross zero more often.

Summary

Momentum is the plain difference between today's close and the close n bars ago. It shows speed and deceleration before direction, but it is in dollars, depends on just two closes, and echoes every big move a second time when that move leaves the window, as Oracle's swing from +$57.97 to −$56.25 showed. Read its slope and divergences, use ROC for comparisons, and filter its zero-line crossings with the trend.