Hull Moving Average (HMA): Complete Trading Guide

Most moving averages face a cruel tradeoff: smooth the line to reduce noise, and you introduce lag; reduce lag, and the line becomes jagged and unreliable. Alan Hull cracked this problem in 2005 by combining two Weighted Moving Averages (WMAs) with an unusual twist — taking the square root of the period.

Here's the calculation broken into three steps. First, calculate a WMA using half the chosen period (for the default period of 14, that's WMA(7)). Second, calculate a standard WMA using the full period — WMA(14). Third, compute the difference: multiply WMA(7) by 2, then subtract WMA(14). This gives you a raw value. Finally, smooth that raw value with another WMA, but this time using the square root of the original period — for period 14, that's WMA(√14), or roughly WMA(4).

The formula looks like this: HMA = WMA(2 × WMA(n/2) − WMA(n)), √n

Why does this work? The doubling of the shorter WMA amplifies recent price action, while subtracting the longer WMA cancels out much of the historical weight dragging the average backward. The final smoothing step with √n cleans up the noise without reintroducing significant lag. The result is a line that hugs price more closely than a standard Exponential Moving Average (EMA) at equivalent periods — often by several candles — which matters enormously when you're trading M15 or H1 charts where a 3-bar lag can be the difference between a profitable entry and chasing a move.

The HMA generates signals in three distinct ways, each with different reliability profiles.

Direction Change Signals — The most direct signal comes from the HMA's slope reversing. When the HMA turns from falling to rising, that's a potential buy signal. When it turns from rising to falling, that's a potential sell signal. Because the HMA leads rather than lags, these turns often appear 2–4 candles before a traditional EMA would confirm the same move. On H1 charts, this can mean entering a trend near its beginning rather than its middle.

Price-Crossover Signals — When price crosses above a rising HMA, that confirms bullish momentum. When price crosses below a falling HMA, that confirms bearish momentum. The key qualifier here is direction: a price cross against the HMA's slope (price crosses above a falling HMA) is a warning sign, not a buy signal. Treat those crossovers as potential reversals requiring additional confirmation.

Divergence Setups — This is where the HMA becomes genuinely powerful. Because the HMA responds quickly to price changes, divergence between the HMA and an oscillator like RSI becomes meaningful. If price makes a new high but the HMA's rate of ascent is flattening while RSI is declining, momentum is deteriorating. That three-way divergence — price, HMA slope, RSI — has historically preceded some of the sharpest reversals on H4 charts.

One caution: the HMA's sensitivity is a double-edged quality. During sideways, choppy markets, the HMA will whipsaw — changing direction frequently without follow-through. Flat markets are the HMA's worst environment. Pairing it with an ADX reading above 25 filters out the majority of those false signals.

Counterintuitive fact: using the same period setting across all timeframes is one of the most common HMA mistakes, and it produces wildly inconsistent behavior.

The default period of 14 was designed as a general-purpose baseline. In practice, different timeframes respond better to different period lengths.

M15 Charts — At 15-minute intervals, price noise is high. A period of 14 responds quickly but generates frequent direction changes that don't always lead anywhere. Periods between 20 and 28 provide better noise filtering while preserving the HMA's low-lag advantage. On M15, the HMA works best as an entry timing tool after a higher-timeframe trend is already established.

H1 Charts — This is arguably the HMA's sweet spot. The default period of 14 performs well here, capturing intraday trends without becoming too reactive to individual candle spikes. Some traders prefer period 21 on H1 to align with the 21-period EMA that many institutional desks monitor. The H1 HMA slope direction alone can serve as a reliable trend filter for M15 entries.

H4 Charts — At four-hour intervals, the HMA with period 14 can still feel reactive. Periods between 9 and 14 work well for swing trading setups, where you want early signals on multi-day moves. Using two HMAs simultaneously — a faster HMA(9) and a slower HMA(21) — and watching for their crossovers gives you a dynamic signal system on H4 that adapts to changing trend strength.

Period selection ultimately comes down to your holding time. Scalpers targeting 10–20 pip moves on M15 need faster settings (period 9–14). Swing traders holding positions for 2–5 days on H4 benefit from slower settings (period 14–21) that don't trigger premature exits on normal retracements.

Theory becomes profit only when translated into a repeatable process. Here's how to build a structured approach around the HMA.

Step 1: Establish the Trend Context — Open your H4 chart and note the HMA(14) direction. This is your macro filter. Only take long trades on lower timeframes when the H4 HMA is pointing upward, and only short trades when it's pointing downward. This single filter eliminates a substantial portion of counter-trend trades that look attractive in isolation but fail in context.

Step 2: Identify the Entry Trigger — Drop to H1. Wait for the HMA to turn in the direction of your H4 bias. The turn itself — not just the slope, but the actual inflection point — is your entry zone. Set a limit order near the current HMA value rather than chasing the candle. The HMA's smoothness means it often acts as dynamic support or resistance during the first pullback after a trend begins.

Step 3: Define Risk Parameters — Place your stop-loss below the most recent swing low (for longs) or above the most recent swing high (for shorts), not below the HMA line itself. The HMA moves with price, so using it directly as a stop reference means your stop is always shifting — a structural problem for position sizing. Fixed structural levels give you clean risk-to-reward calculations.

Step 4: Manage the Trade — As the trend develops, the HMA slope flattening is an early exit signal. A slope reversal is your definitive exit. Partial profit-taking at 1:1 risk-to-reward preserves capital while leaving a runner for the full trend move.

Pulsar Terminal's built-in multi-level SL/TP tools let you set these exit levels directly on the chart based on HMA signals, making the execution of this process fast and precise without manual order management between timeframes.

The Simple Moving Average (SMA), Exponential Moving Average (EMA), and Hull Moving Average each occupy different positions on the lag-versus-smoothness spectrum — and each has legitimate uses.

The SMA gives equal weight to every bar in its period. A 14-period SMA weights yesterday's price the same as the price from 14 days ago. This creates maximum smoothness but also maximum lag. The SMA is useful for identifying long-term structural levels — the 200 SMA on daily charts, for example — where lag is irrelevant because you're defining support and resistance, not timing entries.

The EMA reduces lag by applying exponentially greater weight to recent bars. A 14-period EMA responds faster than a 14-period SMA. Most professional traders use EMAs for trend identification on intermediate timeframes. The EMA's main limitation is that it still carries significant lag compared to what price is actually doing right now.

The HMA eliminates most of that remaining lag. On an H1 chart during a strong uptrend, the HMA(14) will often be within 2–3 pips of current price, while the EMA(14) might trail by 8–12 pips and the SMA(14) by 15–20 pips. Those differences compound across multiple trades.

The tradeoff: the HMA generates more false signals in ranging markets than either the EMA or SMA. The SMA's lag, frustrating as it is in trending markets, actually protects it from whipsawing during consolidation. The practical solution is using all three in different roles — HMA for entry timing, EMA for trend direction, SMA for structural levels — rather than treating any single average as a complete system.