How to Use KAMA (Kaufman's Adaptive Moving Average)? Efficiency Ratio, Parameters, and Choppy Market Filtering

Trading Strategies

08 October 2026

The biggest difference between KAMA and a regular moving average is this: its speed is determined by the market itself. When the trend is clean, it follows closely. When the market chops back and forth, it slows down on its own, sometimes becoming almost flat. This "adaptive" switch is the Efficiency Ratio (ER). You don't need to manually switch moving average periods, but you do need to understand what ER is telling you — and when it still fails.

OKX Exchange
A leading global cryptocurrency platform,suitable for both beginners and experienced traders.
New user benefit: 20% off trading fees upon registration!！

Efficiency Ratio: KAMA's Gas Pedal and Brake

The Efficiency Ratio measures how directly price is moving. The formula is simple: take the net price change over a period and divide it by the sum of the absolute value of each bar's change over that same period.

Let's use the default 10-period setting as an example: price moves from 100 to 110, so the net change is 10. If each of those 10 bars went up by 1, the sum is also 10, so ER = 1. If price bounced around — up 3 one day, down 2 the next — the net change might still be 10, but the sum becomes 30, so ER = 0.33. The closer ER is to 1, the more efficiently price is moving. The closer it is to 0, the more it's just thrashing around.

This value directly determines how fast or slow KAMA moves. When ER is high, KAMA's smoothing constant is pushed to its maximum, making it behave like a 2-period EMA — it moves as soon as price moves. When ER is low, the smoothing constant is pushed to its minimum, making it behave like a 30-period EMA — price moves but KAMA barely reacts. The final smoothing constant is then squared, which means in a choppy market, KAMA's "brake" is pressed even harder than a linear adjustment would suggest, and the flattening becomes more pronounced.

Why the Default Parameters Are 10, 2, and 30

KAMA's three parameters are not chosen at random. The 10 is the ER lookback period — it determines how long a window is used to judge trend efficiency. The 2 and 30 are the "fastest" and "slowest" boundaries: when the trend is at its cleanest, KAMA degrades into a 2-period moving average; when the market is at its most chaotic, it degrades into a 30-period moving average.

Perry Kaufman recommended KAMA(10,2,30) as the default when he introduced this indicator, and most platforms and libraries use this set as their default implementation. Professional traders rarely adjust these three parameters significantly, because KAMA's design intent is to let the indicator adapt on its own rather than relying on manual parameter tuning.

If you want it to be more sensitive to short-term efficiency changes, you can lower the 10 to 6 or 8. The trade-off is more signals that are also more fragmented. If you trade daily or 4-hour swing setups, the default values are usually sufficient. It's not recommended to mess with the 2 and 30, because the squaring step already creates a wide gap between the fast and slow modes. Changing those boundaries further tends to make the indicator's switching between trend and chop less decisive.

How to Read KAMA Signals

The most direct way to use KAMA is to look at its slope. A consistently rising KAMA indicates an uptrend, a consistently falling KAMA indicates a downtrend, and a flat KAMA means the market has no direction. But judging the slope requires a bit of patience: when ER hovers around 0.3, KAMA's slope may be very weak, and the "rising" or "falling" at that point has little reference value.

A more practical approach is to look for breakouts after KAMA goes flat. When KAMA has been moving sideways at a low level for more than 10 bars, it means ER has dropped into a zone where the market is almost completely inefficient. This compressed state doesn't last forever. The moment price picks a direction and KAMA starts to tilt is usually more meaningful than a crossover when ER is hovering in the middle range. Some KAMA strategy scripts on TradingView use "KAMA rising or falling for N consecutive bars" as an entry condition rather than a single-bar slope flip — this filters out the hesitation period right after ER leaves the low zone.

If you use crossovers between KAMA and price as signals, be aware that KAMA flattens in choppy markets, and price will repeatedly cross back and forth through it. Crossover signals in this situation have a low win rate. Some scripts wrap KAMA with an ATR channel and use a price breakout of 1.5x or 2.5x ATR as the trigger condition — this is specifically to avoid the meaningless crosses that happen while KAMA is flat.

The Limits of Choppy Market Filtering

KAMA can filter out choppy conditions, but that doesn't mean it can make you money in a range-bound market. Its job is to help you make fewer mistakes in a choppy market, not to turn a choppy market into a trending one.

There's a common misuse: seeing KAMA go flat and assuming "a breakout must be coming." A flat KAMA only tells you that the current Efficiency Ratio is low. It says nothing about the direction or timing of a breakout. Price can keep chopping for a long time while KAMA stays flat, and it can also fake a downside breakout before moving up. A flat KAMA is a preparation signal, not an entry signal.

Another limitation is that KAMA can still lag in extreme market conditions. When price suddenly breaks out with heavy volume after a prolonged period of consolidation, ER will spike within a few bars, but KAMA still needs several bars to accelerate from a nearly flat state and catch up to price. If you chase the breakout on the very first bar, KAMA hasn't reacted yet, and the moving average level you're referencing is still too low. Waiting for KAMA to tilt before entering means giving up some profit in exchange for confirmation.

The Practical Impact of Changing Parameters

If you lower the ER period from 10 to 5, KAMA reacts faster to efficiency changes. In the early stages of a trend, it tilts sooner and gives direction faster. The trade-off is that the "flat" assessment in choppy markets will also flip-flop more frequently: ER might jump from 0.1 to 0.4 and back to 0.2 within one or two bars, making KAMA's slope jitter. It looks like it's about to move, but it hasn't.

If you raise the slow period from 30 to 50, KAMA will go flatter and duller in choppy markets. The benefit is stronger noise filtering. The downside is that it takes even longer to catch up after a trend starts, and you'll wait longer for it to tilt.

There's no absolute right or wrong with these adjustments, but there is one principle: change only one parameter at a time, and compare across at least two different market environments (one clearly trending period + one clearly choppy period). "Optimal parameters" found by testing on a single instrument and a single price move will most likely fail in a different environment.

How KAMA Relates to Other Moving Averages

KAMA doesn't replace EMA or SMA — it suits different scenarios. If you trade an instrument you know very well with a relatively stable noise level, a fixed-period EMA may be simpler and more direct, and it has higher market consensus (for example, the 20 EMA on the SPY daily chart is watched by many people, which gives it reference value on its own).

KAMA is better suited for instruments or timeframes where the noise level changes dramatically, such as the crypto market on the 4-hour chart. Bitcoin's switches between trending and ranging are abrupt. A fixed-period moving average gets repeatedly crossed during sideways periods and then fails to keep up when a trend starts. KAMA's Efficiency Ratio is designed precisely for this kind of scenario.

OKX Exchange
A leading global cryptocurrency platform,suitable for both beginners and experienced traders.
New user benefit: 20% off trading fees upon registration!！