What Is Heikin Ashi?
Heikin ashi is a candlestick chart drawn from averages rather than traded prices. Each candle's close is the mean of that bar's open, high, low and close, and each open is the midpoint of the previous heikin ashi candle. The result smooths a trend but no longer shows the prices that actually traded.
Covered on this page: TradingView.
Heikin ashi looks like a candlestick chart and is not one. Understanding what it replaced is the difference between a useful second chart and a misleading first one.
How it works
Here is a real rise with real chop in it — 16 down candles inside a 29-bar advance.
The same data, redrawn. Two formulas do all of it:
Close = (Open + High + Low + Close) / 4
Open = (previous heikin ashi Open + previous heikin ashi Close) / 2
The high and low are then stretched to include those two values. Notice what the second formula means: today’s candle is built partly from yesterday’s drawn candle, so the smoothing is a chain, not a per-bar calculation.
What the smoothing is worth
Measured on this rise: 16 down candles became 8. Exactly half.
That is a real improvement and it is not the improvement the format is usually sold with. The pitch is an unbroken run of one color; what actually happened is that the chop halved.
The wick claim, measured
You will read that heikin ashi candles lose their lower wicks in an uptrend, and that a bar without one confirms strength.
On this rise, 9 bars out of 29 had no lower wick. Under a third — so the absence of a wick is not the routine condition of a trend, and a rule built on it is firing on about 31% of bars.
The problem with the prices
This is the part that matters most and gets said least. The close on a heikin ashi candle is an average of four numbers. Nobody traded at it.
On one candle of the worked example, the drawn close is 101.61 and the price the market actually finished at is 101.20 — 0.41 apart.
So any decision that needs a real price cannot be taken off this chart. A stop, a limit order, a level, a measured target. The chart is showing you a smoothed abstraction and it does not announce which numbers are real.
The gaps disappear
A consequence of the formula that almost nobody mentions, and it follows directly from the second line of arithmetic above.
On an ordinary chart a candle can open away from the previous close. A stock closes at 50 and opens at 47 the next morning; that three-point hole is a real event, and often the most informative thing on the chart.
Heikin ashi cannot draw it. Its open is defined as the midpoint of the previous drawn candle, so every candle starts inside the body of the one before it. Not usually — always. It is an identity in the formula, not a tendency.
So a gap becomes a long candle instead of a hole. The information is not lost exactly, but the thing that made it stand out is, and a chart where an overnight shock and an ordinary strong session look alike is a chart hiding something you wanted to see.
The name says as much: heikin ashi is Japanese for average bar. It was built to average, and this is one of the things averaging costs.
A worked example
You use it as a second chart, not the first one. Ordinary candles on the left, heikin ashi on the right, one question asked of the right-hand chart: is this move still going?
A run of same-color bars says the trend is intact. That is what the halving of chop buys you — fewer bars that make you doubt something that has not changed.
Every price you act on comes from the ordinary chart. Entry, stop, invalidation, target.
And you accept a bar of delay, for the structural reason in the next section, which means this is a tool for staying in a move rather than for catching a turn.
The original data
Across our study of 24,971 trading videos, 74 have heikin ashi in the title. The median one gets 20,370 views, only 57% fail to pass 50,000, and the median length is 10.3 minutes.
That is a high median for this glossary — well above chart patterns at 2,307 and moving average convergence divergence (MACD) at 3,645, though below chart reading at 77,589 and trend lines at 61,260 — and it comes from 74 videos, a small field.
The corpus carries description text for 63 of those 74, and across those 63, 1 mentions invalidation, failure, or what a bad read looks like.
When it fails
In a range it barely helps
Measured on the sideways stretch above: 17 color changes on the plain chart became 14 on heikin ashi. A reduction of about a sixth, against a half in the trend.
It smooths trends because trends are smooth. In a trading range there is no underlying direction for the averaging to bring out, so the flipping survives — and the range is exactly where you wanted the help.
It is late, and it has to be
Each candle’s open comes from the previous candle, so a turn cannot show up until at least the bar after it happens. In that example the color changed one bar after the actual high.
That is not a fault to be tuned out. It is what averaging means, the same trade-off as a moving average: smoother, and later.
You read the numbers as prices
Covered above and worth repeating because it is the expensive one. Never take a stop or a target off a heikin ashi chart.
You found the run afterwards
A completed run is obvious and an in-progress one is not. Cover the right-hand side and the smoothed chart is no more honest about the future than the raw one — it just looks calmer while being equally uncertain.
Related
Candlesticks is what a candle means before it is averaged, and the four prices heikin ashi rearranges.
Moving average is the other smoothing tool, and it makes the same smoother-but-later trade in a way that is easier to see.
And trading range is the condition where this tool measurably stops helping.
I do not run these as my main chart, and the reason is the price problem rather than the smoothing. If you are deciding where to put a stop, the numbers on the chart need to be numbers the market actually traded at, and on this chart they are not. It earns its place as a second chart you glance at to answer one question, which is whether a move is still going.
— Michael Whitman
This page is educational, not financial advice. Test every idea on your own charts before risking money. Some links on this page earn a commission if you buy through them. It costs you nothing and it does not decide what appears here or in what order — how these pages are made is set out in our methodology.