Every moving average forces the same bad trade-off on you. Set the period short and it reacts quickly, but it also flips up and down through every quiet, sideways stretch and hands you losing signal after losing signal. Set it long and the chop stops fooling it, but now it drags so far behind a real trend that half the move is gone before the line agrees anything happened. You end up re-tuning the period by hand, one setting for trending days, another for ranges, guessing which one today will be. The Kaufman adaptive moving average was built so you stop guessing.
By the end of this guide you will know how KAMA changes its own speed, what its main settings actually do, whether it repaints, and where it earns its place against a plain EMA or a Hull average.
Key Findings
- KAMA adjusts its own smoothing: it tracks price closely in a strong trend and flattens out when the market moves sideways, so one line handles both conditions.
- It reads direction, not just price: it measures how much ground price covered against how much it wobbled to get there, and uses that to slide between a fast and a slow average.
- Its edge is fewer false flips in a range, not earlier entries than a fast average in a clean trend.
- It does not repaint history: closed-candle values stay fixed; only the point on the live bar keeps moving until that candle closes.
What is the Kaufman adaptive moving average?
The Kaufman adaptive moving average is a moving average that changes how fast it reacts depending on what the market is doing. Perry Kaufman published it in his 1995 book Smarter Trading, and the problem he set out to solve is the one at the top of this page: a fixed-speed average is always wrong for half the market.
Here is the core idea in one image. Picture a dial with “fast” on one end and “slow” on the other. An ordinary average bolts that dial in place and leaves it there. KAMA keeps a hand on the dial and turns it every single bar, toward fast when price is moving with purpose and toward slow when price is just shuffling around.
So on a chart, KAMA looks like one smooth line laid over price, same as any moving average. The difference is in its temperament. During a strong directional run it tightens up against price like a fast average. When the move dies and price starts chopping, the same line goes quiet and nearly horizontal, refusing to be dragged around by noise that means nothing.
How does KAMA decide when to speed up or slow down?
It compares how far price actually got to how hard it worked to get there. Think of two walkers crossing a field. One marches straight across; the other zig-zags the whole way and ends up in the same spot. Both covered ground, but only the first one was really going somewhere. KAMA rewards the straight-line walker with speed and treats the zig-zagger as noise to be smoothed away.
In practical terms, a clean trend produces a lot of net travel with little wasted motion, and KAMA reads that as a genuine move worth tracking closely. A sideways range produces almost no net travel despite constant back-and-forth, and KAMA reads that as chop worth ignoring. That single measurement is the dial. The stronger the directional reading, the closer KAMA leans to its fast setting; the choppier the reading, the closer it leans to its slow setting.
Kaufman’s original defaults set the direction reading over a lookback of 10 candles, with the speed sliding between a fast bound of 2 and a slow bound of 30. You do not need to memorise those numbers to use the tool, but they are the knobs: a shorter lookback makes KAMA declare trends sooner and chase more noise, while a longer one makes it more patient and slower to engage.
KAMA vs EMA vs Hull: which moving average, when?
The honest answer is that they are built for different jobs, and picking one is really about naming your own worst failure mode. If you bleed money getting whipsawed in ranges, you want the average that goes quiet in chop. If you lose the most by entering trends late, you want the one that reacts hardest and earliest.
| Moving average | Speed behaviour | Strongest in | Main weakness |
|---|---|---|---|
| Exponential (EMA) | One fixed speed, always the same | Predictable trend following | Flips repeatedly in a sideways range |
| Hull (HMA) | Fast at all times | Earliest read in a clean trend | Can overshoot and flip on shallow pullbacks |
| Kaufman (KAMA) | Adapts, fast in trends and slow in chop | Markets that switch between trend and range | Slightly late to engage the very start of a new trend |
That last column is the real decision. KAMA trades a little bit of trend-start reaction for a lot less noise in a range, which is a good deal for most swing and intraday traders because markets spend more time ranging than trending. If you want the opposite bias, the earliest possible turn regardless of chop, the Hull moving average leans that way instead. And if you are still deciding how much weight to give any single line, our breakdown of leading versus lagging indicators is worth a read before you build a system around one.
RelicusRoad Pro
Have you been trading for a while but have never made consistent profits or are you new to FOREX trading and want to get a head start? Try RelicusRoad and you'll never look back.
Get RelicusRoad ProDoes the Kaufman adaptive moving average repaint?
A KAMA built on closed candles does not repaint its history. When a candle closes, the value it printed is locked, and scrolling back a week later shows you the same line you watched in real time. On that basic honesty it behaves like any properly built moving average.
The confusion, as always, is the live candle. While the current bar is still forming, its price keeps changing, so KAMA recalculates its newest point on every tick. Because KAMA can be moving at its fast setting during a strong push, that last point sometimes swings quite a bit before the candle finally closes. That is not repainting. The line can keep adjusting its signal until the candle closes; a signal you can trust is one that locks at the close and stays where it printed.
This matters more with an adaptive average than a plain one, precisely because KAMA is quiet most of the time. When it finally springs to life mid-candle, the turn feels significant, and it is tempting to act before the bar settles. If you want a repeatable way to prove any tool actually holds its signal after the close, the non-repaint indicator screenshot test walks through it step by step.
How do you trade with KAMA?
The cleanest use is as a regime filter, and it plays to exactly what KAMA is good at. When the line is sloping and price holds on one side of it, you have a trend, and you take signals only in that direction. When the line goes flat and price crosses back and forth over it, KAMA is telling you the market is ranging, and the right move is often to stand down rather than trade every cross.
That second half is the part traders skip. A flat KAMA is not a broken signal, it is a signal that says wait. Reading the flatness as information, instead of hunting for entries anyway, is where the tool actually saves you money.
Be clear-eyed about the limit, though. An adaptive average sharpens when you engage and when you wait, but it does nothing for the trade once you are in it. Position size, a stop you set before you click, and the patience to skip a marginal setup are still what keep an account alive. KAMA points at the right conditions; your risk rules decide whether being right pays.
Where RelicusRoad Pro fits
The frustration underneath all of this is trust in the read. A line that swings mid-candle, a turn that looks decisive and then unwinds by the close, a signal that behaved one way in replay and another way live. RelicusRoad Pro is built to lock its signals at the candle close, so the read you act on is the read that stays on the chart, whether you trade MT4, MT5, or TradingView. It does not replace an adaptive average like KAMA; it removes the doubt about whether the signal in front of you is final. To see how a filtered average pairs with a full entry system, our moving average crossover strategy guide covers the confirmation rules that keep a smooth line from trading you into every range.
Frequently asked questions
What is the Kaufman adaptive moving average?
The Kaufman adaptive moving average, or KAMA, is a moving average that adjusts its own responsiveness based on how directional price is. Perry Kaufman introduced it in his 1995 book Smarter Trading. When price is moving cleanly in one direction, KAMA speeds up and hugs the move. When price is chopping sideways with lots of back-and-forth, it slows down and goes nearly flat, which keeps it from flipping on every small wobble. That single behaviour is the whole idea: one line that reacts fast when reacting fast is worth it, and stays calm when it is not.
How does KAMA decide when to speed up or slow down?
KAMA looks at how much price has actually travelled over a lookback window versus how much it bounced around to get there. A strong one-way move covers a lot of ground with little wasted motion, so KAMA reads that as trending and turns up its speed. A sideways market covers almost no net ground despite plenty of movement, so KAMA reads that as noise and slows down. That ratio of net travel to total movement is the dial that slides the average between a fast setting and a slow one on every bar.
What are the best KAMA settings?
Kaufman’s original defaults are a lookback of 10 for the direction reading, a fast bound of 2, and a slow bound of 30. Those are a sensible starting point, not a rule. A shorter lookback makes KAMA quicker to declare a trend and quicker to chase noise; a longer one makes it more patient but slower to engage a real move. The practical method is to load the defaults, watch how the line behaves on your pair and timeframe through both a trend and a range, and adjust only if it is engaging chop too eagerly or missing clean trends.
Does the Kaufman adaptive moving average repaint?
A KAMA calculated on closed candles does not repaint its history. Once a candle closes, the value it printed for that bar is fixed and stays put when you scroll back later. The part that moves is the point on the current, unfinished candle, because its price is still changing and every moving average recalculates its newest value as that happens. That live movement is normal, not a repaint. The trap is treating a fresh KAMA turn mid-candle as final when the bar can still close somewhere else.
Is KAMA better than the EMA?
Neither is better in every situation; they answer different questions. An exponential moving average reacts at one fixed speed whether the market is trending or ranging, so it is predictable but flips repeatedly in chop. KAMA changes speed with the market, so it flips far less in a range at the cost of being slightly less eager at the very start of a new trend. If your losses come mostly from getting whipsawed sideways, KAMA is the better fit. If you need the earliest possible read in a clean trend, a fast EMA or a Hull average may suit you more.
Tired of second-guessing whether a turn is final? RelicusRoad Pro fixes its signal at the candle close, on MT4, MT5, and TradingView alike, so you act on a read that will still be there tomorrow.