WhitmanTrading

How to Read a Heikin Ashi Chart

To read a Heikin Ashi chart, treat the colour and the wick pattern as a trend filter and ignore the prices entirely. Each candle is calculated from averages, so the open, high, low and close shown are not prices anything actually traded at.

A Heikin Ashi chart replaces each candle with one built from averages of the current and previous bars. The result is much smoother than a price chart, and the prices it displays are computed values rather than prices anything traded at.

Before you start

An understanding that the candles are averaged, so the prices shown are not tradeable prices. This is the single most important fact about the chart type.

A real price chart alongside it, because entries and stops need actual levels. Two charts, one for the trend read and one for everything you actually place an order at.

A decision about what you are using it for, since it answers exactly one question. Is there a trend, and is it still going. That is the whole scope.

The steps

1. Read the colour run, not the individual candle

A range-bound stretch of price with a smoothed sequence.
A long run of one colour is the signal. Illustrative chart - not real market data.

A long sequence of same-coloured candles is a trend. That is what the smoothing is for, and it is considerably clearer here than on a price chart.

2. Watch the wicks for the turn

A slice of price data with narrowing bodies.
Wicks on both sides mean the averaging is undecided. Illustrative chart - not real market data.

A run of candles with no lower wick is a strong advance. When lower wicks appear and bodies shrink, the averaged values are no longer moving one way.

3. Ignore every price on the chart

A long-horizon price series with computed values.
These numbers were never traded. Illustrative chart - not real market data.

The open, high, low and close are all derived. Placing an order at one of them is placing it at a price that does not correspond to anything in the order book.

4. Take levels from the real chart

A slow-moving stretch of price with an actual level marked.
Entries and stops come from the price chart. Illustrative chart - not real market data.

Swing highs and lows, ranges, prior closes — all from the price chart. On this site’s shared series the ninetieth percentile bar range is 1.101, and that measurement applies to real bars.

5. Accept the lag

The first half of a price series with a delayed transition.
Smoothing and lateness are the same property. Illustrative chart - not real market data.

Averaging means the chart turns after price does. That delay is what removes the noise, and trying to tune it away removes the reason for using the chart type.

6. Use it as a filter, not a trigger

A section of a price series with directional permission.
It says which direction, not when. Illustrative chart - not real market data.

Long setups only while the run is green. The setup itself comes from the price chart, and this decides which half of your setups you are permitted to take.

7. Never backtest an entry price from it

The first half of a price series tested against real data.
A test on computed prices is a test of the formula. Illustrative chart - not real market data.

A backtest that enters at a Heikin Ashi open is filling at a price nobody offered. The result describes the averaging formula rather than a strategy.

How to tell it worked

0 orders were placed at a price taken from this chart.

A real price chart was open alongside it, supplying every level.

The chart was used as a filter, so it declined at least 1 trade in the last 30 days.

And no backtest used its prices as fills, only its colour as a condition.

Why the prices are not real

A candlestick chart annotated with the round-trip cost of a switch.
You still transact at real prices, not averaged ones. Illustrative chart - not real market data.

The close is the average of the bar’s own open, high, low and close. The open is the average of the previous candle’s open and close. Neither is a price that traded.

A section of a price series drawn without volume context.
And smoothing makes a thin market look orderly. Illustrative chart - not real market data.

Which is also why it flatters illiquid instruments. Averaging turns erratic bars into a tidy sequence, and the tidiness is the calculation rather than the market.

What it is genuinely good at

Staying in a trend. The smoothing suppresses the individual bars that would shake you out, which is a real behavioural benefit and the honest case for the chart type.

Making a directional read fast. Colour and wick pattern communicate a trend state in a glance, more clearly than a price chart does.

And nothing else. It cannot show you a level, a gap, a rejection wick or a specific bar’s range, because all of those are properties of real prices that the averaging has removed.

The two-chart arrangement

One Heikin Ashi chart for the directional read, one price chart for everything else. Same instrument, same timeframe, side by side.

The Heikin Ashi chart has one job: say whether a trend is running and which way. It never supplies a number you act on.

The price chart supplies every level. Entry, stop, target, and the bar range you size against — all of them from bars that actually traded.

Which sounds like duplication and is the opposite. Each chart is doing the thing it is good at, and the failure mode being designed out is the specific one where a smoothed, averaged, deliberately lagging representation of the market gets used to choose a price.

The original data

Of the 24,971 unique videos in research/search-study-corpus.jsonl, 52 mention this chart type in the title, at a median of 33,639 views across 45 channels, and 65% of those titles are instruction-shaped. Renko charts appear in 72 at 5,347. The counts come from site/corpus_count.py.

A candlestick series with several gaps, the largest of them marked.
A gap disappears entirely into the averaging. Illustrative chart - not real market data.

52 videos at 33,639 — one of the largest audiences per video of any charting subject measured here. A chart that makes trends look obvious is very appealing to demonstrate, and almost none of that coverage mentions that the prices on it are computed.

A stretch of price bars cut short at a decision point.
14 green candles and no lower wicks. Buy the pullback? Illustrative chart - not real market data.

The answer to the question on that chart is that there may not have been a pullback. The averaging can show an unbroken run while the real chart shows several retracements — so the pullback you want to buy has to be located on the price chart, where it either exists or does not.

When it fails

The failure is treating the chart as prices, and it produces orders at levels that never existed. A stop placed at a Heikin Ashi low sits at a computed number. Sometimes it is close to a real level and behaves plausibly; sometimes it is far from one and gets hit by a move that never came near the level you thought you had chosen. Nothing on the display indicates which situation you are in, because every candle looks equally like a candle.

The second failure is backtesting fills from it. That tests the formula.

A third is using it alone. It supplies no levels at all.

A fourth is expecting it to lead. Averaging means it lags.

A fifth is reading a thin market’s tidiness as stability. That is the smoothing.

And a sixth is looking for gaps or rejection wicks. The calculation removed them.

Heikin Ashi covers the formula in detail. Candlesticks is the real-price chart this is derived from. And trend following is the approach it is built to support.

What I actually do

The mistake that cost me was taking a stop level off one of these charts. The low shown on a Heikin Ashi candle is a computed number — no order ever filled there. I placed a stop at a price that did not exist in the market and it behaved in a way I could not explain until I understood the calculation.

— Michael Whitman

This page is educational, not financial advice. Test every idea on your own charts before risking money.