Heikin Ashi vs Regular Candlesticks: Half as Many One-Bar Runs, a Different Colour on 1 Bar in 3, and Ahead in 7 of 12 Match-Ups but Never by More Than Noise

Most traders meet Heikin Ashi as a switch in the chart settings: same chart, calmer candles. The question that follows is simple. If I use the same rule I use on normal candles, does the Heikin Ashi version trade any better?

I tested exactly that. Same rules, same bars, same exit, one version reading Heikin Ashi colours and the other reading real candle colours.

What is actually different between the two?

A regular candle shows where price opened and closed in that bar. A Heikin Ashi candle shows an average: its close is (open + high + low + close) / 4, and its open is the midpoint of the previous Heikin Ashi candle. The full construction and what it does to a backtest are in Heikin Ashi strategy tested.

Here is what that averaging did to the colours on my data.

EURUSD M5 and M15 from June 3 to August 26, 2026, 16,926 and 5,642 bars; gold M5 and M15 from June 16 to August 26, 13,736 and 4,577 bars. Candles that opened and closed at the same price are left out of the run count.

That is the visual promise kept: on regular candles two runs in three were a single bar on EURUSD, on Heikin Ashi about one in four. The last row is the price of it. On one bar in three the Heikin Ashi candle was green while the real bar fell, or red while it rose.

How was the comparison set up?

• Three rules, applied to both candle types: colour change (first candle of the new colour), second candle (two in a row after a change), and flat candle (the first candle of a new run, within five bars, with no wick against the direction).
• Entry at the real close of the signal bar, for both. The Heikin Ashi version is not allowed to fill at its own averaged price.
• Exit: stop 1.5 ATR, target 3 ATR, 48 bars at most, one trade at a time.
• Cost: 1 pip on EURUSD, $0.30 an ounce on gold.
• Noise test: for each match-up I resampled both sets of trades 2,000 times and took the 5% to 95% range of the difference. A gap counts only if that whole range is on one side of zero. I compared candles against indicators the same way in candlestick patterns vs indicators.

Which won, rule by rule?

After cost, pips on EURUSD and dollars an ounce on gold. Positive difference means Heikin Ashi did better.

On EURUSD M5 the two versions were the same system. With more than 900 trades on each side, the difference was 0.01 to 0.13 pips a trade, and both lost about the spread. On EURUSD M15 and gold M15 Heikin Ashi was ahead in four of six, but with 146 to 410 trades per side the noise range was wider than the gap every time.

The only two clear results went against Heikin Ashi, both on gold M5. The regular colour change made +$0.06 a trade and the Heikin Ashi one lost $1.10. The regular flat candle, a real bar with no wick against it, made +$0.19, and the Heikin Ashi flat candle lost $2.42. A real marubozu-style bar is a rare event: 1,349 of them on gold M5 against 2,221 flat Heikin Ashi candles. The Heikin Ashi version fires on any calm stretch of a trend, including the end of it.

Neither version passed my full three-check test anywhere, the one used for every article on this blog. The regular-candle results on gold M5 were positive and beat the best 5% of random entries, but each lost money in one of the two halves of the data. Real-candle patterns on their own are covered in candlestick patterns tested.

Is Heikin Ashi later than regular candles?

I expected it to be. An average reacts after the price, so the colour change should come later. It did not.

“Already moved” is the distance from the extreme of the last 24 bars to the signal close. “Still to come” is the furthest price went in the signal’s direction in the next 24 bars.

Heikin Ashi was 0.02 to 0.11 ATR earlier on all four charts. The reason is in how often the colours change. A regular candle chart changes colour constantly, so the “first green candle” after a fall is often not the first green candle that mattered. Heikin Ashi only changes colour when the average turns, which happens close to where a regular trader would see a run of green bars start. Both still arrived after more of the move than was left, the same pattern I found in bar close vs live signal.

Does Heikin Ashi save money by trading less?

It does trade less. Always in the market and reversing on every colour change, Heikin Ashi flipped 81 times a day on EURUSD M5 against 92 for regular candles, and 22 times a day on gold M15 against 45.

Profit counted on real closes for both.

Heikin Ashi lost less on all four charts, by 0.03 to 0.21 a trade and by more than half in total on both gold charts, mostly because it paid the spread half as often on gold. That is the honest case for it: a smoother chart makes a flip-every-colour trader trade less. It does not make the trades good. All eight always-in runs lost money.

Which one should you trade from?

• For reading the trend, Heikin Ashi is easier. Runs were twice as long and one-bar colour flips less than half as common.
• For entries, it made no measurable difference. 7 of 12 match-ups went to Heikin Ashi, none by more than noise. The two clear gaps were both on gold M5, both in favour of regular candles.
• Do not use Heikin Ashi for flat-candle entries on gold M5. It was the biggest gap in the table, $2.60 an ounce per trade.
• Take entry prices and stops from the real candles. They differ in colour on one bar in three, and the Heikin Ashi close is an average, not a price.
• If your system flips on every colour change, Heikin Ashi will cost you less, by trading less. A slower chart does more of the same: see multi-timeframe analysis on MT4.

How to switch Heikin Ashi on in MetaTrader and what smoothing does to it is in Heikin Ashi settings for MT4.

Frequently asked questions

Is Heikin Ashi better than candlesticks?

Not for entries on this data. Heikin Ashi was ahead in 7 of 12 rule-by-rule match-ups, never by more than noise; regular candles were clearly ahead in 2, both on gold M5.

Is Heikin Ashi slower than regular candles?

No. Its colour change came 0.02 to 0.11 ATR earlier than the regular colour change on all four charts, because regular candles change colour so often that their first change is usually not the one that mattered.

Why does Heikin Ashi show green when price fell?

Because it is an average of the bar and the previous Heikin Ashi candle. On 33% to 34% of bars the Heikin Ashi colour differed from the real candle underneath.

Can I use candlestick patterns on Heikin Ashi?

The shapes are not the same events. A Heikin Ashi candle with no lower wick appeared 2,221 times on gold M5, a real bar with no wick against it 1,349 times, and the Heikin Ashi version lost $2.60 an ounce per trade more.

Does Heikin Ashi reduce trading costs?

If you flip on every colour change, yes. It traded half as often on gold and lost 0.03 to 0.21 less per trade, but every always-in run still lost money.