The Hull Moving Average (HMA) is one of those rare technical indicators that genuinely solves a real problem: regular moving averages either lag too much, or, when you shorten them to reduce lag, they become too noisy to be useful. Alan Hull’s insight in 2005 was that you could take a weighted moving average, run it twice with different lookbacks, subtract the slower one from twice the faster one, and then smooth the result over a square-root window. The result is a line that flips faster than an EMA at trend changes while staying smoother than a similarly fast simple moving average.
For NIFTY 50 and BANK NIFTY traders, the HMA is most useful on the daily chart as a trend filter, and on the 15-minute chart as a directional bias for intraday entries. This guide walks through what HMA is, how to compute it, what its strengths and weaknesses are, and how it fits alongside the other moving-average tools published on emaindicator.com.
What problem does HMA solve?
Every moving average is a compromise between two things: how quickly it reacts to a real trend change, and how often it flips on random short-term noise. A short Simple Moving Average (SMA) reacts fast but whipsaws constantly in a sideways market. A long SMA stays smooth but tells you about a trend change ten or twenty bars after it actually happened.
Alan Hull’s HMA reduces this trade-off in a clever way. By computing a weighted moving average over the full period, subtracting it from twice the weighted moving average over half the period, you produce a series that “leads” price. Then you smooth that lead with a final weighted moving average over the square root of the period, removing the noise that the lead introduced. The math is a little dense but the intuition is straightforward: capture the leading edge first, then re-smooth.
The output is an indicator that, on most equities and indices, flips noticeably earlier than EMA at trend changes while showing fewer false flips inside ranges than a fast SMA does.
The full formula has three steps. Let WMA(P, n) be the weighted moving average of price series P over n bars.
HMA(n) = WMA( 2 × WMA(P, n/2) − WMA(P, n), √n )
Walking through it:
- Compute the slow leg:
WMA(P, n). This is the standard weighted moving average of close prices over the full period.
- Compute the fast leg:
WMA(P, n/2) — the weighted moving average over half the period. This series leads the slow leg.
- Compute the leading series:
2 × WMA(P, n/2) − WMA(P, n). Subtracting the slow leg from twice the fast leg removes the lag that even WMA carries, but introduces choppiness.
- Smooth the leading series: take a weighted moving average of the result over
√n bars. The square-root window is short enough to preserve the lead but long enough to suppress noise.
For a 49-period HMA on NIFTY daily, that means the slow WMA is over 49 days, the fast WMA is over 24 days, and the final smoothing WMA is over 7 days (√49 = 7).
A worked example — HMA on NIFTY 50 daily
Take the most recent 50 daily closes of NIFTY 50. The 49-period HMA is computed in three sweeps:
- The 49-day weighted moving average gives more weight to the most recent days, less to older days.
- The 24-day weighted moving average gives a faster, more responsive line.
- Twice the fast minus the slow produces the lead — a value that anticipates where the average is heading.
- A 7-day weighted moving average of that lead is the final HMA value.
On a chart, the HMA line tends to sit very close to price during trending phases and curves smoothly through ranging phases. When the slope of the HMA flips from negative to positive, that is your bullish signal; the inverse is your bearish signal.
The number that matters in practice is the slope, not the cross with price. A flat HMA tells you the trend is gone; a steeply rising HMA tells you a clean uptrend is underway.
When HMA works on Indian indices
HMA shines in two contexts on NIFTY and BANK NIFTY:
Daily trend confirmation. A 50-period HMA on the daily chart is roughly equivalent in lookback weight to a 100-period EMA, but it flips earlier and stays smoother through gap days. Use it as a “are we in a long-side or short-side regime?” filter for swing trades.
15-minute intraday bias. A 21-period HMA on 15-minute NIFTY catches intraday swings cleanly when the broader market is trending. On expiry day or news days, when intraday moves are large and clean, HMA is one of the fastest ways to get an entry-side bias.
When HMA fails
HMA is not magic. It still fails in two places:
Tight intraday ranges. On a 5-minute NIFTY chart during a 60-point lunchtime range, the HMA will flip every 8–12 bars. The whipsaw rate jumps. This is what the whipsaw tracker is for — it counts these false flips so you know when HMA (or any moving average) is being whipsawed by chop.
Gap-prone instruments. HMA’s weighted formula gives a lot of weight to the most recent close. A large overnight gap can pull HMA sharply in one direction even though the post-gap price action is unchanged. On gap-heavy indices like SENSEX or single-stock charts with frequent earnings gaps, this is more of a problem than on NIFTY 50.
HMA vs EMA, SMA, and WMA
| Average |
Lag at trend change |
Smoothness in chop |
| Simple Moving Average (SMA) |
Highest |
Smoothest at long lookback, choppy at short |
| Exponential Moving Average (EMA) |
Medium |
Slightly noisier than SMA |
| Weighted Moving Average (WMA) |
Slightly less than EMA |
Similar to EMA |
| Hull Moving Average (HMA) |
Lowest of the four |
Comparable to medium-length EMA |
This is HMA’s selling point: it occupies a part of the lag-vs-smoothness frontier that the other three cannot reach.
For a head-to-head comparison of EMA vs SMA specifically — the most-asked of these comparisons — see our EMA vs SMA on NIFTY page, which publishes the lag in trading days, total cross counts, and forward-return profile of each, recomputed against 5 years of NIFTY 50 daily data.
Common HMA periods used by Indian traders
There is no single right answer, but these are the conventions:
- 9 or 14 HMA on 5-minute and 15-minute charts for scalping signals
- 21 HMA on 15-minute charts for intraday trend bias
- 49 or 50 HMA on daily charts for swing-trade direction filter
- 100 HMA on daily charts as a long-trend confirmation
- 200 HMA on daily/weekly charts as a primary trend backdrop
Shorter HMAs flip earlier, but the whipsaw rate climbs. Longer HMAs are stable but late. Combining a fast HMA (signal) with a slow HMA (filter) is a common setup — only take long signals from the fast HMA when the slow HMA is also rising.
How to use HMA alongside emaindicator.com tools
The HMA is a complement, not a replacement, for the live EMA tools on this site:
- Use the NIFTY 50 EMA regime page to see whether the broader regime is Trending Bull, Trending Bear, or Ranging right now. If it is Ranging, HMA signals are unreliable — sit on hands or fade extremes.
- Use the whipsaw tracker to gauge whether moving-average crossovers in general have been clean or noisy over the last 30 days.
- Use the multi-timeframe alignment table to confirm that 5-minute, 15-minute, 1-hour, and daily moving averages all point the same direction before taking an HMA-driven trade.
A practical rule we recommend: only trust an HMA flip on the 15-minute chart when the 1-hour regime call agrees with the new HMA direction. That single filter eliminates the majority of false intraday signals.
Settings to try first
If you have never used HMA before, start with these three configurations on NIFTY 50 daily:
- HMA 50 — primary trend filter
- HMA 21 on the 15-minute chart — intraday bias
- HMA 9 on the 5-minute chart — fastest signal (use only when the slow HMAs agree)
Plot all three on the same chart. If all three slope the same direction, you have a well-aligned trend. If they disagree, the trend is mixed and edge-cases dominate — exactly when an unfiltered moving-average signal is most dangerous.
Bottom line
The Hull Moving Average is genuinely useful on Indian-index daily charts as a trend filter and on 15-minute charts as an intraday bias. It is not a complete trading system. It is a single moving average with the best lag-vs-smoothness profile of any standard MA, used best as one component of a setup that also checks regime and multi-timeframe alignment.
Read HMA formula step by step for the full calculation walk-through, or HMA trading strategy for the rules-based system around the indicator.
Frequently asked questions
- What is the Hull Moving Average?
- The Hull Moving Average (HMA) is a moving-average indicator developed by Australian trader Alan Hull in 2005. It uses a weighted moving-average formula combined with a square-root lookback to remove the typical lag and noise of a regular moving average. The result is a line that hugs price closely while still smoothing out short-term volatility.
- Who created the Hull Moving Average?
- Alan Hull created the Hull Moving Average in 2005. He is an Australian author and trader who developed it as a faster alternative to traditional moving averages without introducing the choppiness that simply shortening the period would cause.
- What is the best Hull Moving Average setting?
- Most traders use a 14, 16, or 21-period HMA on intraday charts and a 50 or 100-period HMA on daily charts. There is no universally best setting — it depends on the timeframe and the asset. Shorter periods give earlier signals but more whipsaws; longer periods are smoother but slower to flip on real trend changes.
- Is HMA better than EMA for NIFTY?
- HMA flips faster than EMA at trend changes because of its square-root lookback term, but the trade-off is that on choppy intraday NIFTY sessions it can also flip more often inside ranges. For trend-following on daily NIFTY charts, HMA tends to give earlier turn signals than 50/200 EMA. For ranging markets, neither helps — that is what the regime classifier on emaindicator.com is for.
- Can HMA be used as a trend filter?
- Yes. A common use is: only take long entries when price is above the HMA and the HMA is sloping up; only take shorts when price is below the HMA and the HMA is sloping down. Combine the HMA-slope filter with a regime check (Trending Bull / Trending Bear) and the multi-timeframe alignment table for stronger setups.
Related on emaindicator.com