Backtest evidence

What happens when fills are priced at the Heikin Ashi close, and does Heikin Ashi reduce false signals?

Is the Heikin Ashi Color Flip Profitable? 960 Settings Tested

Finding

The same strategy becomes a different one. The Heikin Ashi close is (open + high + low + close) / 4 — an average of four prices and none of them. On USDJPY hourly bars it sits a median 4.55 pips from the real close, and when the bar is green it is below the real close 63.7% to 67.3% of the time. The same 6,107 trades came to −523.1 pips at the real close and +34,120 pips at the Heikin Ashi close, with no change to the strategy. The colour runs do lengthen: on hourly bars one colour lasted a mean of 1.95 bars on ordinary candles against 4.02 on Heikin Ashi, runs ending after a single bar fall from 50.4% to 23.8%, and colour changes halve from 3,185 a year to 1,548. What decided the result was the timeframe: filling at the real candles, 63 of 80 settings finished 2025 ahead on hourly bars against 29 of 80 on 15-minute, whose median was −509.3 pips.

Key results

The same 6,107 trades priced two ways
−523.1 pips at the real close, +34,120 pips at the Heikin Ashi close
Gap between the two closes on hourly bars
median 4.55 pips; on green bars it is below the real close 63.7% to 67.3% of the time
Mean colour run on hourly bars in 2025
1.95 bars on ordinary candles, 4.02 on Heikin Ashi

Scope

  • USDJPY, 2025-01-01 to 2025-12-31, with 2024 run identically for comparison
  • 15-minute, hourly and four-hour bars
  • 4 readings x 8 calculation variants (standard, better, smoothed 3/5/10, double-smoothed 3/5/10) = 80 per cell; 3 timeframes x 4 windows = 960 runs
  • 0.3 pip spread fixed, 0.1 lots; the fill-price experiment ran at zero spread

Method

Measured with Formiq's backtester. Fills use the real candle's close; Heikin Ashi only decides when to trade. Four readings — colour flip, a run of N same-coloured bars, a bar with no wick, and the doji forms. The only exit is the opposite signal, and since all four readings are symmetric, entry and exit share one rule set.

Limitations

  • One pair, two years. The gap between the Heikin Ashi close and the real one scales with how far price moves, so a different instrument changes the size of the invented profit too.
  • The fill experiment ran at zero spread. A real account pays spread on either calculation, but the gap between them is unchanged.
  • The spread is fixed at 0.3 pips. The hourly colour flip breaks even at 1.05, so that assumption is doing real work.
  • Exits are on the opposite signal. No stop-and-target combination improved anything by a margin worth reporting.
  • The MT4 export covers the standard variant only; the three smoothed variants need a rolling average of each component, so the export falls back to standard and says so in the emitted code.