Strategy · Published May 12, 2026 · 6-min read
A Hull Moving Average strategy is only as good as its filters. Here is a complete rule set — entry, stop, target, regime gate — built around HMA slope on NIFTY 50, with the realistic limitations spelled out.
A Hull Moving Average trading strategy is only as good as the filters around it. The HMA itself tells you direction — fast and smooth — but it tells you nothing about whether the market is trending or ranging, whether volatility is high or low, or whether multiple timeframes agree. Every successful HMA strategy is really an HMA-based direction signal wrapped in a set of filters that screen out the trades where HMA is least reliable.
This guide lays out one complete, rule-based HMA strategy for NIFTY 50 — the entry trigger, the exit rules, the regime gate, and the position sizing — with the realistic limitations spelled out at every step. None of this is investment advice; it is a worked example of how to put a discretionary indicator like HMA into a mechanical framework.
The basic HMA signal is a slope flip. The HMA slope is positive when today’s HMA is higher than yesterday’s, negative when it is lower. A “slope flip” is the bar on which the slope changes sign.
For our NIFTY 50 daily strategy, we use HMA(50) — a 50-period Hull Moving Average on the daily chart. Slope flips on this HMA happen on average once every 4–6 weeks, which is a sensible frequency for swing trades.
Bullish flip: HMA_t > HMA_t-1 AND HMA_t-1 ≤ HMA_t-2
Bearish flip: HMA_t < HMA_t-1 AND HMA_t-1 ≥ HMA_t-2
The flip bar is the entry candidate. Whether you actually take the trade depends on the filters below.
The single most important filter for any HMA strategy is regime. HMA is a trend-following tool. In ranging markets, it whipsaws.
For Indian indices, we use the live regime classifier on emaindicator.com:
This gate alone removes 30-40% of HMA signals from a typical 5-year window, but the trades it removes are disproportionately the bad ones. The win rate of the filtered signals is materially higher than the unfiltered set.
Check the live regime call on the NIFTY 50 EMA regime page.
The second filter is timeframe agreement. If the daily HMA flips bullish but the weekly trend is still bearish, you are fighting the bigger picture and the average payoff drops.
For our daily HMA(50) strategy:
The multi-timeframe alignment table on emaindicator.com publishes this exact check across 5m, 15m, 1h, and daily for both NIFTY and BANK NIFTY. For weekly, you can compute it yourself or extend the live tool.
The third filter is the recent whipsaw rate. If HMA crosses have been firing every few days for a month — i.e., the moving-average crossover frequency is elevated — it is a sign that the market is choppy and the next signal is also more likely to be false.
The whipsaw tracker reports the 30-day whipsaw count for 9/21 EMA on 5-min and 15-min — use this as a proxy. Rule: if the 15-min 30-day whipsaw count is 10 or higher, expect HMA to whipsaw too. Wait for the count to drop below 5 before taking new HMA-based trades.
Combining all three filters:
Long entry: HMA(50) daily slope flips bullish AND 1-hour regime is Trending Bull AND weekly trend is Bull or Flat AND 15-min 30-day whipsaw count < 5. Short entry: HMA(50) daily slope flips bearish AND 1-hour regime is Trending Bear AND weekly trend is Bear or Flat AND 15-min 30-day whipsaw count < 5.
The entry order is placed at the next bar’s open after the slope-flip bar — i.e., if the flip happens on Tuesday’s daily close, the long is entered at Wednesday’s open. This avoids look-ahead bias and matches what a discretionary trader would actually be able to do.
Pick whichever fires first:
Trigger A — slope flip against the trade. If you went long and the HMA slope flips bearish, exit on the next bar’s open. This is the trend-following exit and accounts for most of the closes in a typical run.
Trigger B — initial stop loss. Place a stop at: - For longs: the lowest low of the prior 10 daily bars before entry. - For shorts: the highest high of the prior 10 daily bars before entry.
This stop protects against the case where the slope flip was wrong from the start and HMA quickly flips back.
Trigger C — fixed time exit (optional). If neither A nor B fires, exit at 30 trading days. HMA-based trades that haven’t worked or been stopped out within 30 days are usually decaying — the trend has lost momentum even if HMA hasn’t yet flipped.
A simple volatility-aware sizing rule: risk 1% of capital per trade, where risk = (entry price − stop price). For a 24,000 NIFTY index level with a stop at 23,800, the risk per unit is 200 points. With ₹1,00,000 of capital and 1% per trade (₹1,000 of risk), the position size is 1,000 / 200 = 5 units of the index — adjusted for instrument multiplier (1 lot of NIFTY futures, etc.).
This is mechanical and conservative. More aggressive traders scale based on the strength of the regime call (size 1.5× when all four timeframes are aligned, 0.5× when alignment is mixed). That is discretion on top of the rules.
A rule-based HMA-slope strategy on NIFTY 50 daily, with the three filters above, historically produces:
These are rough ranges, not guarantees. The exact numbers depend on the specific HMA period, the regime parameters, and the test window. The key insight: filtered HMA produces a small but positive edge consistent with trend-following on a major index. Unfiltered HMA — slope flips alone with no regime gate — barely breaks even after costs.
Trading every HMA flip. The single biggest mistake is taking signals without the regime filter. In a 5-year window, ~40% of HMA flips happen in Ranging regimes. Those trades are coin flips at best, persistent losers at worst.
Using too short an HMA on noisy timeframes. A 9-HMA on 5-min NIFTY flips dozens of times per session. Every flip is a “signal” in a literal sense, but the signal-to-noise ratio is terrible. Use shorter HMAs only with very tight filters or as a confirmation, not a trigger.
Ignoring the stop loss. HMA’s slope can stay aligned with a trade for weeks while price drops 5% from your entry. The slope-flip exit will eventually fire, but by then the loss exceeds what the position-sizing assumed. Always have a hard stop.
Mixing timeframes confusingly. A clean HMA strategy uses one timeframe for the signal (daily) and others as filters (1-hour regime, weekly trend). Mixing the actual entry across multiple timeframes — “I took the daily long but exit on 5-min HMA flip” — adds variance without adding edge.
A working HMA strategy is HMA-as-direction-signal plus regime gate plus alignment check plus volatility-aware sizing plus mechanical exits. The HMA itself is one ingredient; the filters and rules are the rest. Get any single piece wrong and the strategy degrades fast.
For the live state of the regime classifier and whipsaw tracker that power the filter rules above, see: - NIFTY 50 EMA regime — current 1-hour and daily regime call - Whipsaw tracker — 30-day false-signal counts - Multi-timeframe alignment — 5m / 15m / 1h / 1d agreement - Methodology — full open documentation of every classifier rule on this site