Kaufman Adaptive Moving Average (KAMA): Settings, Efficiency Ratio & Strategy

In the early 1990s, Perry Kaufman was already a veteran of quantitative trading. He had traded through the Russian wheat deal of 1973, survived 20% interest rates in 1980, and watched the 1987 crash unfold in real time. By the time he sat down to write his book Smarter Trading: Improving Performance in Changing Markets (published in 1995 by McGraw-Hill, with a foreword by Jack Schwager of Market Wizards fame), he had one persistent frustration with moving averages.

The problem was simple but stubborn: every existing moving average forced you to pick a speed. Use a short-period EMA and you get fast signals — but also an avalanche of whipsaws during sideways markets. Use a long-period SMA and you filter out the noise — but you also filter out half your profit because you enter late and exit late. There was no middle ground.

Kaufman's insight was that the market itself should decide how fast the moving average responds. In a clean, directional trend, the MA should hug price closely (like a fast EMA). In a choppy, range-bound market, it should barely move at all (like a very slow SMA). The key was finding a reliable way to measure whether the market was trending or chopping.

His solution was the Efficiency Ratio — a single number between 0 and 1 that captures how "efficient" price movement has been over the last N periods. If price moved 100 pips in one direction over 10 bars, and the sum of all individual bar movements was also about 100 pips, that is perfectly efficient movement (ER near 1). But if price moved only 10 pips net while the sum of individual movements was 200 pips, price was basically running in circles (ER near 0).

Kaufman then used this Efficiency Ratio to blend between a fast smoothing constant (default: 2-period EMA speed) and a slow smoothing constant (default: 30-period EMA speed). When ER is high, KAMA behaves like a 2-period EMA. When ER is low, it behaves like a 30-period EMA. And everything in between is a smooth gradient.

The default parameters Kaufman recommended were KAMA(10, 2, 30): 10 periods for the Efficiency Ratio lookback, 2 for the fast constant, and 30 for the slow constant. These remain the standard to this day, and for good reason — they work across forex, stocks, and futures without much tweaking.

What made KAMA genuinely different from other moving averages at the time was not just reduced lag. It was the idea that a moving average should have awareness of market conditions built into its formula. Before KAMA, MAs were passive — they smoothed data at a fixed rate regardless of context. After KAMA, the concept of adaptive indicators became a recognized category in technical analysis.

In MetaTrader 5, you will find KAMA built right into the platform under Insert > Indicators > Trend > Adaptive Moving Average. The default period is typically set to 14, but you can (and should) adjust it to Kaufman's original 10 for the classic behavior he intended.

The Efficiency Ratio is the brain behind KAMA, and understanding it will change how you think about every other indicator on your chart. It answers one deceptively simple question: "How much of the recent price movement was directional versus noise?"

Here is the formula in plain language:

ER = Net Price Change / Total Path Traveled

More precisely: take the absolute difference between today's close and the close N periods ago (the "signal" or direction). Then take the sum of all absolute bar-to-bar price changes over those same N periods (the "noise" or volatility). Divide the first by the second.

If EUR/USD closed at 1.0800 ten bars ago and closes at 1.0900 today, the net change (signal) is 100 pips. If you add up the absolute move of each individual bar over those 10 periods and get 120 pips total, the Efficiency Ratio is 100/120 = 0.83. That is a very efficient, trending market.

But imagine a different scenario: same start and end prices (1.0800 to 1.0810, only 10 pips net), but the individual bars moved up 30 pips, down 25, up 40, down 35... summing to 400 pips of total movement. Now your ER is 10/400 = 0.025. Price went nowhere despite massive movement. That is pure chop.

The ER ranges from 0 to 1:

• ER near 1.0: Strong, clean trend. Price moved efficiently in one direction.
• ER near 0.5: Mixed conditions. Some trend, some noise.
• ER near 0.0: Pure range or chop. Lots of movement, no net direction.

Once KAMA has the ER value, it uses it to build a Smoothing Constant (SC). The formula squares the result to create a nonlinear response — this is deliberate. Squaring means that moderate ER values (say 0.4-0.6) produce disproportionately small smoothing constants, making KAMA very reluctant to react unless the trend is genuinely strong. This built-in skepticism is what makes KAMA so effective at filtering out the "maybe" moves that wreck other MAs.

The smoothing constant formula works like this:

SC = [ER x (Fast SC - Slow SC) + Slow SC]²

Where Fast SC = 2/(2+1) = 0.6667 and Slow SC = 2/(30+1) = 0.0645 using the default parameters.

Then KAMA updates itself: Current KAMA = Previous KAMA + SC x (Price - Previous KAMA)

When SC is large (trending market), KAMA moves aggressively toward price. When SC is tiny (choppy market), KAMA barely budges. You actually see this on the chart — during a range, KAMA goes nearly flat, almost horizontal, like it is waiting. Then when a real breakout occurs and the ER spikes, KAMA suddenly starts tracking price closely.

One practical insight many traders miss: you can plot the Efficiency Ratio itself as a separate indicator. Some traders use an ER threshold of 0.30 as a trend filter — only taking trades when the ER is above 0.30, which signals that the market has enough directional momentum to justify a trend-following approach. Below 0.30, they switch to range strategies or simply stay out. This dual use of the ER — both inside KAMA and as a standalone filter — is something Kaufman himself advocated.

Here is where KAMA earns its keep, and it is the single biggest reason traders switch from EMA to KAMA: behavior during sideways markets.

Pull up any chart with both a 20-period EMA and a KAMA(10,2,30) during a ranging phase. The EMA will weave up and down with every price swing, crossing above and below price repeatedly. Each cross looks like a potential signal. Each one is a trap. If you traded every EMA crossover during a 50-bar consolidation on EUR/USD H4, you would have taken maybe 8-12 trades, most of them losers.

Now look at KAMA during that same range. It is nearly flat. It barely moves. Price crosses above and below it, sure, but KAMA itself is not generating directional signals because it recognizes — through the Efficiency Ratio — that the market is not going anywhere. The ER during a range typically sits between 0.05 and 0.20, which produces a smoothing constant so small that KAMA essentially ignores minor price fluctuations.

This "going flat" behavior is not just cosmetically different — it fundamentally changes your trading outcomes. Think about it: the number one profit killer for trend-following strategies is not missing big moves. It is death by a thousand cuts during ranges. Every false entry costs you the spread, maybe some slippage, and the psychological toll of another stopped-out trade. KAMA eliminates the majority of these false entries by refusing to respond to non-trending price action.

Here is how to use KAMA's flatness as an active signal:

The Flat KAMA Filter:

1. Monitor the slope of KAMA over the last 3-5 bars.
2. If the KAMA line has changed by less than 5 pips (on a major forex pair, H4 timeframe) over 5 bars, classify the market as ranging.
3. During flat KAMA conditions, avoid all trend-following entries.
4. When KAMA begins to slope noticeably (more than 10 pips change over 3 bars), re-engage trend strategies.

This simple filter alone can reduce your losing trades by 40-60% depending on the pair and timeframe, because it keeps you out of the market precisely when trend strategies perform worst.

What to do during flat KAMA periods:

Just because KAMA says "no trend" does not mean you cannot trade. It means you should not trade trends. During flat KAMA conditions, consider:

• Range trading: buy at support, sell at resistance within the consolidation.
• Mean-reversion setups: use RSI or Stochastic to trade bounces between extremes.
• Simply waiting. Sometimes the best trade is no trade. KAMA gives you permission to be patient.

The flat-to-slope transition:

The most powerful signal KAMA produces is not a crossover — it is the transition from flat to sloping. When KAMA has been horizontal for 10+ bars and suddenly begins curving upward or downward, it means the Efficiency Ratio has spiked, which means price has started moving with genuine directional conviction. This transition often marks the beginning of a new trend leg and provides entries that other moving averages miss entirely because they were already zigzagging during the range.

On the daily timeframe, this flat-to-slope transition is particularly valuable for swing traders. You might see KAMA flat for two weeks during consolidation, then it begins rising. That initial slope change, combined with price closing above KAMA, is a high-probability entry signal because you are catching the trend at its very inception — not three bars late like with an SMA crossover.

One caveat: KAMA can occasionally go flat during very slow, grinding trends where price makes progress but individual bars are small and choppy. In these cases, zooming out to a higher timeframe (from H4 to Daily, for example) can help you see the bigger trend that KAMA on the lower timeframe is filtering out.

Now let us get practical. Here is a concrete H4/D1 strategy using KAMA that exploits its adaptive nature for entries you would never get with a standard moving average.

The KAMA Slope-Break Strategy (H4 and Daily)

This strategy uses KAMA not as a crossover tool (price above/below the line) but as a slope-based trend indicator. The logic: KAMA's slope is more meaningful than its position relative to price, because the slope directly reflects the Efficiency Ratio's assessment of market conditions.

Setup:

• Chart: H4 or D1 (works on major forex pairs, gold, and indices)
• Indicator: KAMA(10, 2, 30)
• Confirmation: RSI(14)
• Risk management: 1.5x ATR(14) stop loss

Long Entry Rules:

1. KAMA has been flat (less than 10 pips movement over 5 bars on H4) for at least 8 bars — this confirms a consolidation phase.
2. KAMA begins sloping upward — at least 10 pips of upward change over the last 3 bars.
3. Price closes above KAMA on the bar where the slope change is confirmed.
4. RSI is above 50 but below 70 (confirms momentum without being overbought).
5. Enter long at the close of the confirmation bar.
6. Stop loss: 1.5x ATR(14) below the entry price, or below the most recent swing low — whichever is wider.
7. Take profit: 2x the stop distance (2:1 reward-to-risk minimum), or trail the stop along the KAMA line.

Short Entry Rules: Mirror the above — flat KAMA transitioning to downward slope, price closing below KAMA, RSI below 50 but above 30.

The KAMA Pullback Entry (Daily Timeframe)

This is a bread-and-butter trend continuation strategy. Once a trend is established, KAMA acts as dynamic support (in uptrends) or resistance (in downtrends).

1. Identify a clear trend: KAMA is sloping upward for at least 15 bars on the daily chart.
2. Wait for a pullback: price retraces toward the KAMA line.
3. Entry trigger: a bullish candlestick pattern (engulfing, hammer, or strong close above KAMA) forms at or near the KAMA line.
4. Stop loss: below the KAMA line by 1x ATR.
5. Target: the prior swing high, or trail using KAMA as a dynamic stop.

This works because KAMA in a trend sits closer to price than an SMA would, giving you tighter pullback entries. And because KAMA flattens when the trend stalls, you naturally stop getting pullback signals when the trend is exhausted — a built-in trade management feature.

Dual KAMA System:

For traders who like crossover systems but hate the whipsaws, try this Kaufman-approved approach:

• Trend filter: KAMA(10, 5, 30) — slightly slower, defines the big picture.
• Entry signal: KAMA(10, 2, 30) — standard speed, generates entries.
• Go long when the fast KAMA crosses above the slow KAMA, but only if the slow KAMA is also rising.
• Go short when the fast KAMA crosses below the slow KAMA, but only if the slow KAMA is also falling.

The dual system eliminates crossover signals during flat markets because both KAMAs will be flat and overlapping, producing no meaningful crossover.

Stop Loss Management with KAMA:

KAMA makes an excellent trailing stop. In a long trade, trail your stop 1-2 pips below the current KAMA value. Because KAMA tightens toward price during strong trends, your stop stays relatively close — protecting profits. And if the trend suddenly becomes choppy, KAMA slows down and stops trailing, giving price room to breathe before you get stopped out prematurely.

Compare this to trailing a stop along a 20 EMA, which will keep moving and tightening even during choppy phases, often stopping you out of perfectly good trades. KAMA's adaptive trailing gives you the best of both worlds: tight stops in strong trends, loose stops during consolidation.

Every trader eventually asks: "Why should I use KAMA when I already have my trusty EMA?" Fair question. Let us put KAMA head-to-head against the three most popular moving averages and give you an honest verdict.

KAMA vs SMA (Simple Moving Average)

The SMA treats all prices equally over its lookback period. A 20 SMA gives the same weight to today's close as to the close 20 bars ago. This creates significant lag — by the time an SMA confirms a trend, the move is often well underway.

KAMA, by contrast, adapts its responsiveness based on market conditions. During trends, KAMA tracks price as closely as a fast EMA. During ranges, it goes flat. The SMA cannot do either — it just plods along at the same speed regardless of what the market is doing.

Winner: KAMA, decisively. SMA has its place for identifying long-term trends on weekly charts, but for active trading, KAMA outperforms in almost every scenario.

KAMA vs EMA (Exponential Moving Average)

The EMA weights recent prices more heavily, making it faster than the SMA. Many traders love the EMA for this responsiveness. But here is the problem: the EMA is always responsive. It does not know the difference between a trending market and a ranging one. During consolidation, the EMA weaves back and forth, generating crossover after crossover — and most of them are false signals.

KAMA's adaptive mechanism eliminates this weakness. Where a 20 EMA might generate 10 crossover signals during a 50-bar range (8 of them losers), KAMA would generate 1-2 at most, because it flattens during low-efficiency periods.

However, the EMA does have one advantage: simplicity. It is available on every platform, every charting package, and every timeframe without special configuration. KAMA requires understanding its three parameters (ER period, fast constant, slow constant), which adds complexity.

Winner: KAMA for accuracy. EMA for simplicity. If you are willing to spend 10 minutes understanding KAMA's parameters, it is the better tool.

KAMA vs HMA (Hull Moving Average)

Now this is the interesting matchup. The HMA, developed by Alan Hull, uses weighted moving averages and a square root period to virtually eliminate lag. It is blazing fast — often the first MA to signal a trend change.

But speed comes at a cost. The HMA is so responsive that it whipsaws in choppy markets, sometimes worse than the EMA. It was designed for one purpose — minimum lag — and it achieves that brilliantly. What it does not do is adapt to market conditions. The HMA is equally fast in trends and ranges, which means it generates plenty of noise during consolidation.

KAMA takes the opposite philosophy. It sacrifices some speed in trends (it will never be as fast as HMA at catching reversals) in exchange for near-total noise elimination during ranges. In a fast-moving trending market, HMA enters first. In a choppy market, KAMA keeps you out while HMA gets you whipsawed.

Winner: Depends on your style. Scalpers and day traders may prefer HMA for its speed. Swing traders and position traders will find KAMA more profitable overall because it avoids the losses that HMA's speed inevitably causes during ranges.

The Full Comparison Table:

The Verdict:

If you could only pick one moving average for all market conditions, KAMA would be the rational choice. Not because it is the fastest or the simplest, but because it is the only one that adjusts its behavior to match what the market is actually doing. It will not catch the first pip of a move (that goes to HMA), and it will not give you a perfectly smooth line for long-term analysis (that goes to SMA). But it will consistently keep you on the right side of the trend while protecting you from the choppy periods that destroy accounts.

That said, there is no rule against using multiple MAs. A practical setup many professional traders use: KAMA for entry signals and trend direction, with an HMA or EMA for timing within established trends. Let KAMA tell you whether to trade, and let the faster MA tell you when.