Short answer: the hull moving average (HMA) is a moving average built from weighted moving averages. It turns almost in step with price yet keeps a smooth curve. It gets there by stretching a fast weighted average beyond a slow one, then smoothing that raw line over the square root of the period. The result is an indicator with far less lag than an SMA or EMA of the same length. The trade-off is a line that overshoots on sharp turns and flips needlessly in a range.
The method takes its name from the Australian analyst who designed it, and it sits on the trend analysis side of technical analysis. Most traders use it as a fast trend filter that changes colour when its slope turns. Others use it as a punctual replacement for the EMA in trading strategies built around a single line.
How the HMA calculation works, step by step
Everything starts with the weighted moving average (WMA). A WMA gives the newest close the largest weight and the oldest the smallest, with weights running 1, 2, 3 and so on. Take four closes of 2,418, 2,420, 2,421 and 2,425. The weights sum to 10. The WMA is (2,418 x 1 + 2,420 x 2 + 2,421 x 3 + 2,425 x 4) / 10, or 2,422.1. The plain SMA of those closes is 2,421, so the WMA already leans toward the latest price.
The hull calculation then stacks three of these:
- Choose a period, say 16.
- Compute a WMA over half that period, so 8 bars.
- Compute a WMA over the full 16 bars.
- Double the short average and subtract the long one. This raw line deliberately overshoots.
- Smooth the raw line with a WMA over the square root of the period, which is 4 bars.
Suppose gold is rallying, the 8-bar WMA reads 2,412 and the 16-bar WMA reads 2,404. The raw value is 2 x 2,412 minus 2,404, or 2,420. Price itself may be at 2,418, so the raw line sits ahead of price. That is exactly how the lag gets cancelled. The final 4-bar smoothing tidies the jittery raw values into a curve you can read.
Benefits of the Hull line over ordinary averages
The attractiveness of the HMA is that it answers two complaints at once. Short averages are punctual but wobbly, and long averages are smooth but stale. The HMA is both quick and tidy at the same period. On a clean reversal an HMA(16) usually turns a few bars before an EMA(16), and the slope change is plainly visible. Because the line hugs price action so closely, it also makes a generous trailing reference in a strong trend. A pullback that closes beyond it is a meaningful signal rather than ordinary noise.
How the HMA compares with SMA, EMA and WMA
Four averages of the same length behave in four distinct ways. The table assumes a 20-period setting on the same chart.
| Average | Weighting | Lag at period 20 | Smoothness | Behaviour in a range |
|---|---|---|---|---|
| SMA | Equal across all bars | Longest | High | Flattens, crossed rarely |
| EMA | Decays toward older bars | Shorter | Medium | Crossed often, modest overshoot |
| WMA | Straight-line, newest heaviest | Shorter than EMA | Medium | Similar to EMA |
| HMA | Two WMAs stretched, then smoothed | Shortest | High on trends, coarse on spikes | Flips colour frequently |
The EMA and WMA sit close together, and our EMA vs SMA comparison explains why the weighting scheme matters so little in practice. The HMA is the lopsided one. It wins on lag by a wide margin and pays with overshoot whenever a bar closes violently away from the recent range.
Choosing the period: a decision framework
Period choice is more consequential for the HMA than for an EMA, because the square-root smoothing step changes with it. Work through these criteria in order.
- Prefer a period whose square root is a whole number (9, 16, 25, 49, 100). Platforms round otherwise, and rounding adds ambiguity to the smoothing.
- Match the period to the swing you want to hold. Roughly 9 to 16 suits an intraday trigger, 20 to 34 a swing filter, 49 or more a regime line.
- Lengthen it on fickle instruments. Major forex pairs spend long stretches in ranges, where a short HMA flips in a disorderly way.
- Shorten it on markets that trend hard, such as gold or BTC on a strong day, where lag is the expensive error.
- Never pick the period that looked best on last quarter's chart alone. A 13 that beat 16 last month is curve fitting.
One rule mostly holds. If the line changes colour more than twice in a session while price goes nowhere, the period is too short for the current market conditions.
Trading strategies built on the HMA
Three uses cover most of what the indicator does well.
Slope turn. Enter when the HMA changes direction after a pullback inside a larger trend, confirmed by a closed bar. This is the humble default. It works best with a higher-timeframe filter from a slower line, the logic behind our guide to trend following.
Pullback to the line. In a trend, wait for price to touch or briefly close through the HMA, then re-enter when the next bar closes back on the trend side.
Dual HMA. Pair a short HMA with a long one, for instance 9 and 49, and trade the cross. It fires earlier than an EMA pair of the same lengths, though the usual moving average crossover whipsaw rules still apply.
Picture an account of 20,000 risking 0.5 percent, which is 100, on a 15-minute index CFD. The HMA(25) has been rising all morning, price dips to the line at 18,000 and the next bar closes above it. Entry at 18,000, stop under the pullback low at 17,960, a 40-point distance. That sets the size at 2.5 per point. The prior swing high at 18,100 is the first target, a 2.5 to 1 reward on the risk if it prints.
Common mistakes and limitations
- Reading overshoot as strength. After a single spike bar the raw line stretches beyond price, so the HMA can turn sharply on one print and reverse on the next.
- Trusting the colour flip in a range. Each false flip is a small loss plus the spread, and that cost accumulates quietly until the day is a thankless wash.
- Placing stops on the line. It is a smoothed average, not support, and price pokes through it before resuming.
- Stacking it with an EMA of the same period and calling the agreement confirmation. That is duplication.
- Treating a turn as a forecast. The HMA has scarcely any lag, but it still describes the recent past.
- Watching the live bar. The current value flickers until the close, and acting on the flicker is clumsy.
The honest summary: the HMA is a sharper tool than the SMA or EMA, and sharper tools cut both ways.
The HMA under prop-firm evaluation limits
A short HMA on a 5-minute chart during a sluggish range can flip five times before lunch. At 0.75 percent risked per flip, that is 3.75 percent gone, a hair under a 4 percent daily cap. Under evaluation rules, a period that is too short is not merely annoying, it is precarious.
Two Pivex rule details change how you size this. The total drawdown is a fixed figure that does not climb behind your equity peak, so a lumpy week does not shrink your remaining room. The daily loss allowance is restored at the start of every trading day. There is also no deadline on the attempt, so a longer HMA that trades seldom is a perfectly acceptable plan. Let the built-in assistant keep count of the daily allowance, and a whipsaw cluster never surprises you.
Pick one instrument and one timeframe, plot an HMA(16) and an EMA(16) together, and mark every point over the past three months where the two lines disagreed. That short audit tells you whether the earlier turn paid for the extra flips in your market. If you run the test under Pivex rules, size each entry so that three consecutive flips in one session still leave the daily limit comfortably intact.
Related guides: EMA vs SMA and Trend following.
Before participating, review Pivex's current program rules and decide whether a simulated proprietary trading evaluation fits your experience and risk limits. Participation does not guarantee a reward, income, or any particular trading result.