Doji Candlestick: Types, Meaning and How to Trade It

A doji candlestick is a candle whose opening and closing prices are equal, or almost equal, so the body shrinks to a thin line. Price moved during the period, sometimes a lot, but it closed where it opened. Candlestick analysis reads it as a sign of indecision.

This guide covers what this candlestick pattern is, its anatomy, the five types of doji, how to identify one, the psychology behind it and how to use it in trading. It also shows what happened after it. Our test bench applied the same rule to full price history, which gave 1,196,277 occurrences on daily charts alone across stocks, crypto, forex, ETFs, indices and commodities, and compared the outcome with random days on the same assets.

What is a doji candlestick?

A doji candlestick is a single candle with a very small body: the open and close sit at the same level, or very close to it. The name comes from Japanese candlestick charting, introduced to Western traders by Steve Nison. On a chart it looks like a cross, a plus sign or an inverted cross, according to where the body sits.

This candlestick pattern is not a trading signal on its own. It is a frequent candle, about one daily bar in 7 in our data, and its meaning comes from where it appears: after a lengthy advance, after a sell-off, or in the middle of a flat stretch.

The anatomy of a doji candle

Every candlestick of this kind has three parts:

• The body: the distance between the opening and closing price. Here it is a line or a very thin box. Its color does not change the definition.
• The upper shadow (wick): the distance from the body to the high.
• The lower shadow: the distance from the body to the low.

How small must it be? No universal rule exists. A strict reading asks for exactly the same opening and closing price, which almost never happens on liquid markets. A relative test is more practical. Our bench uses the one from the TA-Lib library: the body is at most 10% of the average height of the previous 10 candles. On the median occurrence we found, the whole bar spanned 1.86% of price, or 0.67 times the average true range (ATR): the typical one is a quieter day than average, not a wide one.

What does a doji candle mean for traders?

It means that neither side won. Buyers drove it up, sellers dragged it down, and the close came back to the open. After a sustained advance or decline, classical candlestick analysis reads that balance as a sign that the side in control may be losing momentum; in a flat stretch, it reads very little into it.

So the candle is a question, not an answer. The next candlesticks give the answer: a close above its high suggests the bulls took over, a close under its low suggests sellers did.

The five main types of doji candlesticks

The types of doji candlestick patterns differ by the length and position of the shadows. Each one tells a slightly different story about the battle between the two sides.

Standard doji

The standard (neutral) version has short shadows of similar length on both sides. It shows a quiet day that stayed near the open throughout. On our data, doji candles of any kind appear once every 7 bars on day charts.

Long-legged doji

The long-legged doji has long shadows on both sides. Price traveled far in both directions, then closed where it opened. It shows more doubt than a standard one, because both sides made a serious attempt and failed. The rickshaw man is a long-legged doji centered in the middle of its span.

Dragonfly doji

The dragonfly doji has a long lower wick and little or no upper shadow: open, high and close sit at the top. Sellers drove it down, buyers brought it all the way back. After a decline, classical candlestick analysis reads the dragonfly doji as a potential bullish turn, close to a bodiless hammer.

Gravestone doji

The gravestone doji is the mirror image: a tall shadow on top and little or no shadow underneath, with open, low and close at the base. Bulls drove it up and failed to hold it. After a rally, classical candlestick analysis reads the gravestone doji as a potential bearish turn, the bodiless version of a shooting star.

Four-price doji

The four-price doji is a single horizontal line: open, high, low and close are the same. It means there was almost no activity, which happens on an illiquid company stock, on holidays or on very short timeframes. It says more about the absence of volume than about the balance between buyers and sellers. Our bench sets aside any bar with no range, so no four-price doji is counted in the statistics on this page.

How to identify a doji on a price chart

To identify this candlestick on a price chart, check three things in order:

1. The body is tiny compared with recent candles. Compare it with the average height of the last 10 candles, not with an absolute number: what counts as small on a company stock chart may be large on an hourly forex chart.
2. The candle is complete. Intraday, price can sit at the open and then move away before the end of the bar. The candle only exists once the bar has closed.
3. The shadows tell you which kind it is. Equal and short: standard. Both tall: long-legged. Only underneath: dragonfly. Only on top: gravestone. None: four-price.

Then look at the setting: the trend before it, the support and resistance levels nearby, and the volume of the bar.

Is a doji candlestick a bullish or a bearish sign?

Neither by itself. The pattern is neutral: it records a balance, not a direction. Classical candlestick analysis infers one from its location. After a sustained rise near resistance, it is read as a potential bearish reversal; after a long decline near support, as a potential bullish reversal. A red one and a green one mean the same thing, because its color carries no weight on such a thin bar.

We do not measure direction because the candle has no bias of its own: there is no expected way for price to go after it. We test the trading setup that follows the first breakout instead.

Market psychology behind the doji pattern

This candlestick is the picture of a bar where buying and selling pressure canceled out. In a steady move, one side dominates and the candlesticks have clear bodies pointing the same way. When this pattern appears after several decisive candlesticks, it shows that the dominant side stopped pushing. Holders may take profits, and others hesitate to add exposure.

That interpretation is reasonable, but it describes one bar. Such doubt can resolve either way, including a simple continuation of the move. Whether candlestick patterns like this one help you anticipate the next move is a question for data, which we answer in the reliability section.

How to trade the doji candlestick: a step-by-step guide

Because these patterns have no direction of their own, the classic trading strategy is to let buyers and sellers choose: place a buy order above its top and a sell order under its bottom, and follow the side that breaks first. This is the strategy our bench simulates on every occurrence.

Step 1: determine the market trend

Before acting, note the trend before the pattern. Classical analysis gives more weight to one that forms after a decisive move than to one inside a sideways range; our bench splits results by the prior rise or decline (below), not by range.

Step 2: locate support and resistance levels

Mark the levels nearby: a prior peak or trough, a round number, a moving average. Classical analysis reads a candle that forms right at one of these levels as the market hesitating where it has hesitated before; our bench does not test levels.

Step 3: look for a confirmation candle

Step 4: set your entry and stop-loss

Buy the break of the top, or sell short the break of the bottom. Put the stop loss at the other end of the pattern: if price breaks one side and then the other, the signal has failed. Size the position on that distance. A target at twice the risk gives a 2:1 reward-to-risk ratio. In our tests, trades that reached the target took a median of 3 days, and trades that were stopped took a median of 1 day.

Mistakes to avoid when trading doji patterns

• Treating all these patterns as reversal patterns. One appears between once every 5 and once every 9 bars, according to asset class and timeframe. Most carry no special meaning.
• Acting before the bar is complete. In the middle of the day, the setup is not yet confirmed.
• Placing the stop too tight. The candle range is the natural risk. In our data, of the daily trades that hit their stop, 23.6% had first been up 1 R.
• Ignoring the trend and the levels. Classical analysis reads its meaning from the setting; in our splits, the prior rise or decline barely changed the result.

Doji vs. spinning top: the differences

A spinning top also signals indecision, but it has a small, visible body with shadows longer than it. This pattern has almost none. It is a matter of degree: it is the extreme form of a spinning top. On our data, the two behaved almost alike: the doji setup reached its 2:1 target 33.7% of the time and the spinning top setup 33.4%, both against 33.5% for the same setup on random dates. Neither average result per trade is statistically different from random dates: the doji's (0.00 R against 0.00 R, No clear edge) and the spinning top's (−0.01 R against −0.01 R).

This candlestick also differs from a hammer. A hammer has a small body at the top of the range, a tall shadow underneath and must follow a decline. A dragonfly doji has the same shape, minus the body. The hammer is a reversal setup by definition, while this candlestick is neutral. In our tests, the hammer's 2:1 trade is rated No clear edge: 35.4% of those setups hit their target against 36.1% on random dates, and its average result per trade (+0.06 R against +0.09 R) is below that of random dates, a gap our test does not separate from chance. The inverted hammer, the cousin of the gravestone doji, is a different case: its daily 2:1 trade, all markets together, is rated No clear edge. Its expectancy is −0.08 R per trade before costs, against −0.08 R for the same trade on random dates: a gap our test does not separate from chance, and a trade that loses money either way.

Reliability of doji patterns across timeframes

This candlestick pattern on its own is not a proven trading signal. On daily charts across all markets, the first-breakout setup did about as well as the same setup on random dates: its average result per trade (0.00 R against 0.00 R) does not pass even a simple test, so its verdict below reads no clear edge. On hourly crypto charts its average result per trade was worse than chance, and that gap does survive the correction. Here are the numbers from our test bench.

Read the table in pairs of columns. The setup is the one described above: entry on the first break of the top or bottom, stop at the opposite extreme, target at twice the risk, results in R (multiples of the risk) before fees. "Same setup, random dates" replays the same entry, stop and target distances on random days of the same assets. Only the gap between the two columns measures the value of the pattern.

On crypto day charts, the pattern did better than chance: 34.0% of targets hit against 32.6%, and an average result of +0.02 R against −0.02 R per trade, but that gap clears only a simple test and falls short once we correct for multiple testing. An expectancy of +0.02 R per trade before costs is small enough for fees and slippage to erase. On hourly crypto charts, its average result per trade was slightly worse than on random dates, a gap that survives the correction; on hourly currency charts, it was below chance too, by a margin that passes a simple test but not the correction (weak, below chance). On 4-hour crypto charts, the setup also lagged random dates by a margin that passes only the simple test (weak, below chance). The all-markets daily row is rated No clear edge: it does not pass even the simple test. The other rows of the table read no clear edge.

Does the setting change the result?

We split those occurrences by setting. These breakdowns are descriptive and not corrected for the number of tests.

• After a rise: 33.3% of targets hit, −0.01 R. After a decline: 34.2%, +0.02 R.
• Price above its 200-day moving average: 33.4%, −0.01 R. Under it: 34.1%, +0.02 R.
• Volume above 1.5 times the 20-day average: 34.2%, +0.03 R. Normal volume: 33.6%, 0.00 R.

None of these filters turned the candle into a decisive edge. In raw numbers, the setting shows more when the candle is part of one of the larger candlestick patterns: the bullish doji star (a long black candle, then a doji that gaps beneath it, after a decline) reached its 2:1 target 37.7% of the time against 36.7% on random dates, with an average result of +0.18 R against +0.16 R per trade, rated no clear edge: our test does not separate that gap from chance. Its entry was triggered on only 28.6% of occurrences.

Real examples from our test bench

Here are real dojis, picked by our bench (best and worst results, textbook shapes, most recent, random draws), with their actual outcome. Trade results are in R, the initial risk of the trade.

Breakouts that reached their target

On November 2, 2025, DOT/USDT printed a doji with a top at 3.016 and a bottom at 2.866. The next day price broke the bottom, triggering a sell entry at 2.866, and reached the 2:1 target at 2.566 the same day. On January 23, 2026, AVAX/USDT printed one between 11.99 and 12.5; price broke the bottom the next day and reached the target at 10.97 on January 29, and AVAX was 16.4% lower 10 days after the candle.

Breakouts that failed

On August 7, 2024, DOGE/USDT printed a doji between 0.09428 and 0.10208. The next day the range covered both extremes, from 0.09354 to 0.108: the buy entry at 0.10208 triggered, and the bench counts the touch of the bottom that same day as a 1 R loss. DOGE closed that day at 0.10748, and was 7.5% higher 10 days after the candle: the call was right, the stop was hit first. XLM/USDT formed a doji on October 28, 2024 and broke its high the next day, then fell back through the bottom at 0.0927 on October 31 for a 1 R loss, before ending 8.2% higher 10 days after the candle. Both cases show how a stop at the opposite extreme of the candlestick can be hit by noise.

Combining doji candles with technical indicators

Doji candles say little on their own. Three tools can be paired with them:

• RSI or stochastic: in classical analysis, an overbought or oversold reading is a sign that momentum is fading. We have not tested this filter.
• Moving averages: in classical analysis, a candle that touches a moving average during a steady move can mark the end of a pullback. Our 200-day split above shows only a slight gap.
• Volume: high volume means many shares or contracts changed hands without a winner. In our data, setups on high volume reached the target 34.2% of the time against 33.6% on normal volume, a slight gap.

Limitations of using doji signals in isolation

• It is frequent: about one candle in 7, so most carry no special meaning.
• The definition varies between platforms, so two traders may not see the same candles on the same chart.
• On its own, its setup showed no edge that survives our correction on any market we tested, and did slightly worse than chance on hourly crypto.

Summary: mastering the doji candlestick

• A doji is a candle whose open and close are almost equal: a sign of indecision, not of direction.
• The five types of doji are the standard, long-legged, dragonfly, gravestone and four-price; our bench does not count the four-price bar.
• Across 1,194,453 daily trades, the first-breakout setup reached its 2:1 target 33.7% of the time, against 33.5% for the same setup on random dates.
• Crypto showed a small lead over random dates on one-day bars, with an expectancy of only +0.02 R per trade before costs, rated Promising, too weak to survive our correction; hourly crypto did slightly worse than chance.
• In raw numbers, the same shape inside a doji star after a decline did better than the lone candlestick, without passing our test against random dates.

Frequently asked questions about the doji candlestick