Momentum (MTM): the raw price change over a lookback, how to read it and where it misleads
Momentum, often shortened to MTM, is the most direct momentum reading a chart can show: how far the latest close sits above or below the close a chosen number of bars back, drawn as a line around zero. Its output is in the contract's own price points rather than in percent, and that single property explains both why futures traders like it and where it goes wrong. This page covers the calculation in plain words, how to pick the lookback, the misreadings that catch people, and what you can adjust when the strategy generator tests it on your own contracts.
What it measures
Momentum answers one question: how much ground has price covered over the last stretch of bars? A positive reading says the market closed higher than it did at the start of the window, a negative reading says lower, and the distance from the zero line is the size of that move in the contract's own price points. On an index future the line reads in index points; on a gold chart it reads in dollars per ounce. It is a widely used public tool that appears under the same name in most charting packages, and it is the raw-difference sibling of the percentage-based Rate of Change.
The calculation fits in a sentence. Take the latest close, take the close from a set number of bars earlier, and subtract the older price from the newer one. Nothing is averaged, scaled or smoothed. That has a practical consequence for reading the line: how much it moves from one bar to the next comes down to the newest bar's own move, offset by the bar-to-bar move that has just slid out of the far end of the window. A few charting packages show a ratio version instead, dividing the newer close by the older one so the line centres on a fixed baseline rather than on zero; this page means the plain difference version, the one measured in points.
Because the output is in points, the price level is built into every reading. When a contract trades at twice the price it did years ago, an ordinary swing covers roughly twice as many points, so the same Momentum line swings about twice as wide even if the market's character has not changed at all. The percentage view divides by the older close and removes that effect. Neither scale is better in general: points match the way futures traders already think about stops, targets and tick value on one contract at today's prices, while percentages are the right ruler for comparing eras or markets.
How it behaves in real markets
In a clean trend the line holds on one side of zero and its height tracks how fast the market is travelling. As a move matures, the line usually flattens or slips back toward zero while price is still printing new extremes, because each fresh bar adds less ground than the bar leaving the far end of the window. That gap between a new price extreme and a lower Momentum peak is the classic divergence reading. It describes slowing pace; on its own it does not announce a reversal.
Choosing the lookback is the main decision, and it is a trade between speed and stability. A short lookback compares price with the very recent past, so the line reacts quickly and crosses zero often, which suits timing and exits but produces more false turns in choppy sessions. A long lookback compares with a more distant close, so the line moves slowly, stays on one side of zero for long stretches and works better as a direction filter, at the cost of turning late. A sound habit is to match the window to the swings you actually trade, and to set any level you compare against in points that make sense for the contract's current price and volatility rather than copying a figure from another chart.
The most common misuse comes straight from the point units. Traders mark a reading that once lined up with important tops and treat it as an overbought level years later, or they compare a Momentum value on one contract with a value on another. Neither comparison holds. An index future that has climbed a long way makes much larger point moves now than it did then, and two contracts quoted on different price scales, such as an equity index future and a bond future, produce readings that cannot sit side by side. Keep fixed levels to a recent stretch of the same contract, or switch to the percentage view when the comparison crosses eras or markets. The second misuse is reading every zero-line cross as a new trend. In a sideways market price drifts above and below where it stood a few bars earlier, so the line crosses zero again and again, and a rule that acts on each cross without a trend-strength or volatility filter ends up trading the noise.
Common ways to use it (concepts)
Trade only with the window's net move
With a longer lookback, the sign of Momentum is a plain statement of whether the market is higher than it was at the start of the window. As a filter it lets long entries through only above zero and short entries only below, while another family decides the exact timing.
Demanding a move of real size
Because the reading is in points, a level can state a concrete requirement: the market must have gained a meaningful number of points over the window, sized against the contract's recent range. That turns a vague idea of positive momentum into a move large enough to matter after costs.
Divergence as an exit cue
When price sets a new high but Momentum peaks lower than it did at the previous high, the trend is still up but covering less ground per window. Many traders use that as a cue to tighten exits or stop adding to the position, rather than as a reason to go short.
Pairing it with other indicator families
The point units make ATR a natural partner. ATR is also measured in points, so setting a Momentum reading against the current average range tells you whether the move over the window is large or ordinary for this contract right now. That comparison adjusts by itself as price level and volatility change over the years, which deals with the long-history problem without giving up point units.
For the zero-line problem, a trend-strength gauge such as ADX or the Choppiness Index can decide whether a cross deserves attention at all. Requiring the market to show directional persistence before a cross counts removes many of the back-and-forth trades that a bare zero-line rule collects inside a range. A filter that stays in points works too: count a cross only once Momentum has moved far enough from zero to be large against the current ATR, so a drift of a few ticks either side of the line is ignored.
It also helps to know what Momentum shares with its neighbours. The MACD compares two averages instead of two closes, so it is smoother and slower to turn, while the RSI folds up-moves and down-moves into a bounded scale. Giving Momentum one role, such as the direction filter, and a bounded oscillator another, such as entry timing, keeps each tool on its own job instead of asking several of them the same question.
In AlphaPrime
In AlphaPrime, Momentum is a built-in group in the strategy generator, listed simply as Momentum, and it is also available in the trial. You can adjust the Indicator period N, which sets how many bars back the comparison close sits, and the Threshold, the level the indicator is compared with. A threshold at zero gives the plain zero-line rules; a threshold away from zero asks for a move of a certain size in points.
The conditions cover the basic readings and several shapes specific to Momentum. Basic ones include Momentum crosses up through the level, Momentum is above the level, Momentum is rising and Momentum turns up, each with its mirror on the downside. The shape conditions go further: Momentum positive thrust to a recent high, Momentum upward divergence (price new low, Momentum not), Momentum snaps positive from a recent low, and Momentum drop-off step rising, which watches the move about to leave the far end of the window. Momentum position in recent range rising and Momentum gap to recent mid-range rising judge the line against its own recent highs and lows, a practical answer to the point-size problem over long histories.
The generator combines these conditions with rules from other families and backtests the candidates on the GPU in parallel, and you can set out-of-sample validation thresholds, so you keep only combinations that still hold up on data they have never seen. Those strategies can be exported as PowerLanguage strategy code, so a Momentum filter found in research can move straight into your usual simulation and live workflow.
FAQ
Why do some charts show Momentum around a fixed baseline instead of zero?
Some charting packages use a ratio version that divides the latest close by the older one, so the line centres on a fixed baseline rather than on zero. It tells you the same direction in proportional terms, much like Rate of Change. This page, and the Momentum group in AlphaPrime, use the difference version, measured in the contract's price points.
Why did my Momentum levels stop working over a long backtest?
A fixed level in points assumes the contract's moves stay the same size throughout the test. When price has risen or fallen a long way, typical moves grow or shrink with it, so a level that was rare early in the history can be routine later. Relating the reading to the current ATR or to its own recent range keeps the threshold in proportion.
Is a zero-line cross enough for a trend entry?
Usually not on its own. A cross only says price is now above or below where it stood at the start of the window, and in a range that flips often. Most systematic traders add a trend-strength or volatility filter and let a backtest on their own contract show whether the cross adds anything.