What Is the Impulse MACD?
Impulse MACD is a variant of the MACD that uses a smoothed moving-average band and stays flat while price remains inside it, only producing a coloured reading once price moves outside the band. It reduces the number of signals in ranging conditions at the cost of reacting later.
Covered on this page: TradingView.
The Impulse MACD is an answer to one specific complaint about the standard MACD: that it fires constantly in a sideways market. The answer works, and it costs something.
How it works
It builds a band rather than a line. A smoothed moving average defines a zone around price, using a high and a low average rather than a single value.
While price sits inside the band, the output is neutral. The histogram is drawn in a neutral colour and no directional reading is offered at all.
When price closes outside the band, the output colours. Above the upper boundary it reads bullish, below the lower one bearish, and the colour is the signal.
What the band is actually doing
The band is a definition of sideways. Anything inside it counts as a range, and the width of the band decides how much movement qualifies as no movement.
That width comes from a smoothing length you chose. A wide band ignores more and signals less; a narrow one behaves closer to a standard MACD, which is the thing it was built to improve on.
So the improvement is adjustable in both directions. There is a setting at which this indicator becomes the indicator it replaced, and nothing warns you when you are near it.
A worked example
On this site’s shared series, direction runs average 2.01 bars. A standard MACD crossing on those turns produces a signal roughly every two bars in ranging conditions.
A band wide enough to contain a typical two-bar move suppresses almost all of them. The neutral colour holds, and the screen is quiet for long stretches.
The cost arrives on the move that matters. The longest run in the same series is 11 bars, and price has to clear the band before anything colours - so several of those bars are spent waiting.
Both effects come from the same parameter. You cannot widen the band to remove the noise without also widening how far a real move must travel before it is acknowledged.
The original data
On this site’s shared series: median bar range 0.493, ninetieth percentile 1.101, largest bar 2.338. Direction runs average 2.01 bars with a longest of 11. A round trip costs 0.0098, about 2% of the median bar range.
Those two run figures are the whole argument. Most movement is two bars long and worth ignoring; the rare 11-bar run is the one worth catching, and any filter that suppresses the first delays the second.
The round trip cost is what makes the trade worth making. At 2% of a median bar per entry and exit, suppressing a stream of two-bar signals saves real money, which is the strongest case for the variant and it is a cost argument rather than an accuracy one.
The settings that change its behaviour
The band length is the main control. It sets how many bars the smoothed average is built from, and therefore how much movement has to happen before price is considered outside the range.
The signal length smooths the output. A longer one produces fewer colour changes and holds a colour through minor reversals that a shorter one would flip on.
The moving-average type matters more here than usual. The band is defined by an average, so switching between a simple and an exponential calculation changes where the boundary sits, not just how smooth it looks.
And none of them has a correct value. Each is a position on the same trade-off - fewer signals later, or more signals sooner - so the honest way to pick is to decide which of those two costs you would rather pay and set all three to agree with that answer.
Who wrote it and why that matters
It is a community script, published free on TradingView by the author who goes by LazyBear, and it is one of the most copied indicators on the platform.
Community scripts are open. The source is readable, which means the band definition and the colouring rule can be checked rather than guessed at - an advantage over any closed tool.
It also means there is no support and no specification. Forks circulate under similar names with different smoothing, so two people running “Impulse MACD” may not be running the same calculation.
Check the script you loaded. Open the source, find the band length and the moving-average type, and know which version you are looking at before comparing your chart to anybody else’s.
When it fails
The characteristic failure is trusting the neutral colour as an all-clear. Neutral means price is inside a band, which is a statement about a smoothed average, not a statement that nothing is happening.
A slow, sustained drift can stay inside the band for a long time. The indicator reports nothing throughout, and a position held on the strength of that silence is being held on the absence of a signal rather than the presence of one.
A second failure is narrowing the band to get more signals, which rebuilds the problem it solved.
A third is treating the colour change as an entry price. It confirms on a close, so the move is already partly done.
A fourth is combining it with a standard MACD for confirmation, which is one calculation agreeing with a modified version of itself.
And a fifth is assuming a filtered signal is a better signal. Fewer signals with the same hit rate is just fewer signals, and whether the filter improved anything is a question only testing answers.
Related
MACD covers the indicator this modifies. Moving average covers the band it is built from. And trading range covers the conditions it is designed to sit out.
Standard MACD’s real problem is that it produces crossovers in conditions where nothing is happening. This variant addresses exactly that, and the way it addresses it is by waiting longer. You get fewer bad signals and you also get in later on the good ones - and which of those matters more depends on how you trade, not on the indicator.
— Michael Whitman, from this video
This page is educational, not financial advice. Test every idea on your own charts before risking money. Some links on this page earn a commission if you buy through them. It costs you nothing and it does not decide what appears here or in what order — how these pages are made is set out in our methodology.