Trend & direction

Kaufman Adaptive Moving Average · KAMA

Perry Kaufman's average that speeds up in clean trends and flattens out in noise.

Works in most conditionsEngine-computed on a fixed sample series
14512096Price above EMA 12 = strengthPrice below EMA 12 = weaknessEMA 12 acts as support / resistance
EMA 12EMA 26How to read KAMA on the chart — the callouts mark what to look for.

The formula

The efficiency ratio (ER) divides the net move over the window by the total distance travelled bar by bar — near 1 for a clean trend, near 0 for chop. It scales the smoothing constant (SC) between a fast 2-period and a slow 30-period EMA, so KAMA races in trends and stalls in noise.

ER = |net change over n| ÷ Σ|each bar's change| SC = [ ER × (fast − slow) + slow ] squared KAMA = prior KAMA + SC × (price − prior KAMA)
Worked example
Net change (10 bars)Sum of bar movesEfficiency ratio
8200.40
8400.20

Same 8-point move, but ER halves as the path gets noisier → KAMA smooths more

What it is

The Kaufman Adaptive Moving Average, or KAMA, is a moving average designed by Perry Kaufman that changes its own speed based on how cleanly the market is trending. A normal moving average uses a fixed smoothing, so it reacts the same way whether price is marching in a straight line or thrashing sideways. KAMA is smarter: when price travels in a clean, efficient direction it accelerates and hugs price tightly, but when price chops back and forth in noise it slows down and flattens out, refusing to chase the whipsaws. This built-in noise filter is the entire point, because it keeps you engaged during real trends and out of trouble during meaningless churn. For a beginner, KAMA is a moving average with a volatility brake that automatically eases off the gas when the road gets bumpy.

How it is calculated

KAMA's intelligence comes from a value called the efficiency ratio, which measures how much of the market's movement was directional rather than wasted. It is calculated as the absolute net change in price over the lookback, typically ten periods, divided by the sum of the absolute bar-to-bar changes over that same window. When price moves in a straight line the net change nearly equals the total movement, so the ratio approaches one; when price ends up where it started after thrashing around, the net change is tiny relative to the total movement and the ratio approaches zero. That ratio is then scaled between two smoothing constants, a fast one derived from a two-period average and a slow one from a thirty-period average, and squared to produce the final smoothing factor. KAMA then updates like an exponential average, adding that adaptive smoothing factor times the gap between price and the previous KAMA value.

Reading it, step by step

The first thing to read is whether KAMA is flat or sloping, because that alone tells you the regime. A flat KAMA means the efficiency ratio has collapsed, price is directionless, and the correct action is usually to stand aside rather than trade. When the line begins to slope and pull away from a recent flat stretch, the efficiency ratio has risen and a real trend is taking hold, which is your cue to engage. The steepness of the slope reflects how efficient and strong the move is. Price crossing above a rising KAMA is bullish and crossing below a falling KAMA is bearish, and because KAMA stays flat during noise, these crossings are far less prone to whipsaw than crossings of a fixed-speed average. The genius of the tool is that its stillness is itself a signal.

Reading the signals on the chart

14512096
EMA 12EMA 26The ▲/▼ marks flag where price most recently crossed the line — the cues a trend-follower would act on.

Best timeframes

  • Scalping1m – 5m
  • Day trading5m – 15m
  • Swing1h – daily
  • Positiondaily +

The efficiency ratio filters chop on any timeframe, but tune the lookback to the instrument.

KAMA vs plain averages

KAMAEMASMA
Adapts to noiseYesNoNo
Speeds up in trendsYesNoNo
Goes flat in chopYesNoNo
WeightingEfficiency ratioRecent barsEqual

Common price-action setups

How the signal typically plays out on the chart.

Trend wakes up

After a long flat stretch, KAMA starts sloping up and pulling away from price — trade in the direction of the new slope with a stop below the last flat base.

Buy the slope
Trend takes hold
Pullback to KAMA

In an uptrend price pulls back to a rising KAMA and holds — buy the bounce off the line with a stop just beneath it.

Buy the bounce
Trend continuation
Slope rolls over

KAMA flattens then tilts down as price closes below it — step aside or short, since the efficiency ratio is confirming a real move rather than noise.

Sell the break
Downtrend begins

Best timeframes and settings

Kaufman's defaults are a ten-period efficiency-ratio window with fast and slow constants of two and thirty, and these are a robust starting point on daily and swing charts. KAMA adapts to any timeframe because its efficiency ratio recalibrates to whatever volatility it sees, so intraday traders and position traders alike use it. Shortening the efficiency-ratio lookback makes it more sensitive and quicker to accelerate, at the cost of reacting to smaller, less significant moves, while lengthening it makes it more deliberate. Widening the gap between the fast and slow constants increases the difference between KAMA's fastest and slowest states, making its adaptation more dramatic. The efficiency ratio is genuinely sensitive to the chosen lookback, so it rewards testing on the specific instrument rather than blind acceptance of defaults.

When and where to use it

KAMA is billed as an all-regime tool because its adaptivity is meant to handle both trends and ranges, but its real edge is as a filter that keeps you out of chop. It works across asset classes and is especially valued on noisy instruments where a fixed moving average would whipsaw constantly. It shines when a market alternates between quiet consolidation and sharp trends, because it sleeps through the former and wakes for the latter. Its main weakness is at the very start of a new move after a long, quiet consolidation: because it deliberately slowed down during the calm, it can be late to accelerate when the breakout finally comes. Use it where noise is the enemy and a little late entry is an acceptable price for avoiding whipsaws.

Strategies that use it

The purest KAMA strategy trades its slope while treating long flat patches as explicit no-trade zones: go long only when KAMA is clearly rising and short only when it is clearly falling, ignoring the market entirely when the line is horizontal. A crossover approach uses price crossing the KAMA as the trigger, buying when price closes above a flattening-then-rising KAMA and exiting when it closes back below. A dual-KAMA method runs a fast and a slow KAMA and trades their crossings, gaining an extra layer of noise filtering because both lines flatten during chop. In each case the KAMA line itself often serves as a trailing stop, since a close back through it signals the efficient move has ended.

Combining it with other indicators

KAMA pairs well with a dedicated trend-strength gauge such as the ADX, which confirms independently whether the market has enough directional force to trust KAMA's slope. Because KAMA is essentially a regime filter, layering it beneath a momentum oscillator like the RSI or the MACD lets you take oscillator signals only when KAMA agrees on direction. Volatility tools such as the ATR help size stops appropriately given that KAMA's behavior already reflects volatility. Some traders combine KAMA with a breakout channel like Donchian or Keltner, using the channel for the entry trigger and KAMA to confirm the move is efficient rather than a false poke. The recurring idea is that KAMA answers is this move real, while a companion indicator answers how strong and in which direction to act.

Where it fails

The defining failure mode is lateness after long consolidations, because KAMA's noise brake, so helpful during chop, means it can lag the very first leg of a powerful breakout while it waits for the efficiency ratio to rise. Traders who expect it to be an early-warning system are repeatedly disappointed at these transitions. The efficiency ratio's sensitivity to the lookback means a poorly chosen window can leave KAMA either too sluggish to catch trends or too jumpy to filter noise. In a market that trends smoothly with little noise, KAMA offers little advantage over a plain EMA and merely adds complexity. The way to work with it is to accept slightly late trend entries as the cost of superior whipsaw protection, and to pair it with a faster trigger when catching the exact turn matters.

A worked example

A stock consolidates for three weeks between forty-nine and fifty-one, and throughout this range the ten-period efficiency ratio hovers near zero-point-one because price keeps reversing, so KAMA sits nearly flat around fifty and issues no signals, correctly keeping you out. Then earnings spark a decisive move: price runs from fifty to fifty-six over four sessions in a nearly straight line, the efficiency ratio jumps toward zero-point-eight, and KAMA's smoothing constant leaps, dragging the line up steeply behind price. As KAMA turns from flat to clearly rising and price holds above it, you enter long near fifty-four, a couple of dollars into the move because KAMA waited for confirmation. You place your stop just under the now-rising KAMA at fifty-two and ride the trend to fifty-nine, exiting when price finally closes back beneath the flattening KAMA line.

Common mistakes

  • Expecting it to be quick off the mark right after a long consolidation — it deliberately starts slow.
  • Trading during the long flat stretches instead of standing aside as the line intends.
  • Leaving the fast/slow and lookback settings on default without testing them on your instrument.
  • Treating a barely-sloping KAMA as a genuine trend signal.
  • Assuming the efficiency ratio makes it immune to whipsaws in every market.