Marubozu Candlestick Pattern: A Complete Guide to Trading the Signal

A marubozu candlestick is a single candle with a long real body and no shadows, or almost none: the session opens at one extreme of its range and closes at the other. A white (bullish) marubozu opens at the low and closes at the high; a black (bearish) marubozu opens at the high and closes at the low. Classical candlestick analysis reads it as a sign that one side, buyers or sellers, controlled the entire session.

This guide covers the meaning of the marubozu, its types, how to identify it on a chart and how to trade it. It also shows what happened next. Our test bench applied the same rule to full price history, 145,183 white and 135,183 black ones on daily charts alone across stocks, crypto, forex, ETFs, indices and commodities, and compared the outcome with random bars on the same assets.

What is a marubozu candlestick?

The marubozu candlestick pattern is a one-candle formation defined only by its shape: a long body that fills the whole range of the session. It can be bullish or bearish, and it can appear anywhere in a trend.

Definition and origin of the term

Marubozu is a Japanese word translated as "bald head" or "shaved head": the candle has no hair (no wick) at either end. Steve Nison, who introduced Japanese candlestick charting to Western traders in Japanese Candlestick Charting Techniques (1991), lists it among the basic candle lines. In technical analysis terms, it is a long day candle with the shadows removed.

Our bench uses a rule a computer can apply without judgment, adapted from the open-source TA-Lib library (function CDLMARUBOZU): a body that is long compared with the bodies of the prior candles, and an upper and lower shadow that are both very short compared with their recent range. Our bar for a long body is stricter than TA-Lib's: more than 1.3 times the average body of the previous 10 candles, where TA-Lib asks for the average itself. No prior trend is required.

Unique physical characteristics

• A long real body, clearly larger than the average of the recent candles.
• No upper wick, or a very small one: the high is at, or next to, the top of the body.
• No lower wick, or a very small one: the low is at, or next to, the bottom of the body.
• Color decides the direction: white or green is bullish, black or red is bearish.

One detail surprised us. On the median occurrence our bench found on daily charts, a white marubozu spans 2.76% of price, or 1.06 times the average true range (ATR) of the asset. The typical marubozu is not a giant candle: its range is close to an ordinary session. What makes it special is that the whole range is body.

Understanding the psychology behind a marubozu

A bullish marubozu tells the story of a session where buyers never let go. Price opens, never trades below the opening level, climbs through the session and closes at the high. No seller managed to push price back during the session, and no buyer took profits before the end. A bearish marubozu is the mirror: sellers in control from the open to the close, with no rebound.

Classical candlestick analysis reads this as momentum: the side that controlled one session may control the next ones. That reading is plausible, but it describes one candle. Whether one session of strong buying or selling predicts the following bars is a question for data, and our data gives a clear answer in the reliability section below.

Types of marubozu candles

There are three types of marubozu, depending on which wick is missing.

Full marubozu

The full marubozu has no wick at all: open and close are the high and low of the session. It is the strict textbook version. On real charts a perfect one is rare, so detection rules, including ours, accept very short shadows.

Opening marubozu

The opening marubozu has no wick at the opening end: a bullish one opens at the low, a bearish one opens at the high, and a small shadow remains at the closing end. When it appears after a decline (bullish) or after a rally (bearish), it is also called a belt hold.

Closing marubozu

The closing marubozu has no wick at the closing end: a bullish one closes at the high, a bearish one closes at the low, with a small shadow at the opening end. One reading ranks it as the stronger of the two partial types, because the close is the price that the market accepted at the end of the session. Our bench does not test the opening marubozu on its own, so it cannot settle that ranking. It does show that the closing version did worse than chance: on 370,811 daily white closing marubozu candles, 52.8% were higher 10 bars later, against 54.1% for random bars, an average excess return of −0.12%, rated worse than chance, close to the result of the full white marubozu.

Bullish vs. bearish marubozu: how to identify them

The bullish marubozu: high buying pressure

A bullish marubozu opens at or near the low of the session and closes at or near the high. The candle is white or green, with a long body. In terms of price action, buyers controlled the market from the first trade to the last. Classical candlestick analysis reads it as bullish continuation in an uptrend, and as a possible reversal after a decline.

The bearish marubozu: intense selling pressure

A bearish marubozu opens at or near the high and closes at or near the low. The candle is black or red. Sellers controlled the session from start to finish, and buyers never managed a rebound. The classical reading sees bearish continuation in a downtrend, and a possible reversal after a rally.

The charts above are not drawn by hand. They show the median shape of the marubozu candles our bench detected, and the lines of the trade we test: for the bullish version, entry above the high, stop loss at the low, target at twice the risk; for the bearish version, the same trade upside down.

How to identify a marubozu on your charts

Three checks are enough, in this order.

Step 1: analyze the candle body length

Compare the body with the last 10 candles. On our bench, a marubozu body is more than 1.3 times their average body, a stricter bar than TA-Lib's. A small candle without wicks is not a marubozu: it carries little information because little happened during the session.

Step 2: verify the absence of wicks

Check that the upper and lower shadows are both very short compared with the range of recent candles. A long wick at either end means the session was contested, and the candle becomes a long day candle or, with a small body, a spinning top.

Step 3: confirm with trading volume

Classical candlestick analysis reads volume above average on the marubozu as a sign that the move had participation. We measured it. On daily charts, white marubozu candles printed on volume above 1.5 times the 20-day average reached the 2:1 target in 36.5% of cases, against 35.9% on normal volume, with an expectancy of +0.10 R against +0.07 R. For black marubozu candles, high volume changed nothing (28.4% against 29.3%). The breakdown is descriptive (not corrected for the number of tests): treat volume as a mild plus, not a filter that changes the odds.

The importance of market context

The same marubozu candle is read differently depending on where it appears.

Marubozu in a trend continuation

A white marubozu in an uptrend, or a black one in a downtrend, is read in classical analysis as the trend gaining strength. We split daily occurrences by the prior trend (price compared with 10 bars earlier). White marubozu candles after a rise were higher 10 bars later in 52.7% of cases; after a decline, 52.7%. Random bars on the same assets rose 54.2% of the time. Neither the continuation case nor the reversal case beat chance.

Marubozu at support and resistance levels

A bullish marubozu that closes above a resistance level, or a bearish one that closes below a support level, is read in classical analysis as a breakout with conviction. A marubozu that stops right at a level is weaker: the level may hold on the next session. We have not tested support and resistance levels, so treat this as a reading rule, not a measured one.

The danger of marubozu at market exhaustion

After a long run, a very large marubozu can mark the last burst of a move rather than its start: late buyers rush in, and no one is left to buy. Our data shows a related effect. On hourly charts, the next bar after a white marubozu rose only 42.6% of the time, against 47.9% for random bars; after a black marubozu, the next bar fell 42.5% of the time, against 48.5%. In the short term, price tended to give back part of the move rather than extend it. These next-bar figures are descriptive, outside the tests behind our verdicts.

How to trade the marubozu pattern

The classic marubozu trade follows the direction of the candle: buy a break above a bullish marubozu, sell a break below a bearish one. This is the trade our bench simulates on every occurrence.

Setting up your entry strategy

There are two ways to enter. The aggressive one enters at the close of the marubozu. The conservative one places a buy stop just above the high of a bullish marubozu (or a sell stop below the low of a bearish one) and waits for the next sessions to trigger it. Because the marubozu closes at its extreme, the stop entry is very close to the close. In our tests on daily charts, the entry was triggered in 88.3% of white marubozu setups; in the other 11.6%, price broke the low first and the setup was canceled.

Determining valid stop-loss placement

Managing take-profit targets

Our bench sets the target at twice the risk (a 2:1 reward-to-risk ratio); the next resistance level for a long trade, or support level for a short trade, is an alternative we did not test. On daily charts, white marubozu trades that reached the 2:1 target took a median of 7 days; trades that were stopped took a median of 3 days. A trap worth knowing: of the daily trades that hit their stop, 24.4% had first been up 1 R. A trailing stop or a partial exit at 1 R changes that, but we have not tested it.

Comparing marubozu to other candlestick patterns

Marubozu vs. engulfing patterns

An engulfing setup has two candles: the second body covers the first. The second candle can be a marubozu or close to one. The difference is that the engulfing requires a prior trend and a reversal of color, while the marubozu is judged alone. On daily charts, the bullish engulfing reached the 2:1 target in 36.9% of 48,554 trades against 35.7% on random dates, the white marubozu 36.0% on 128,026 trades against 36.4%. Neither average result per trade passes even a simple test against random dates.

Marubozu vs. long day candles

A long line (long day) candle has a long body but can keep normal wicks. The marubozu is a long day candle with the wicks removed. If the shaved ends carried information, the marubozu should do better than the long line. It does not: on daily charts, 53.2% of long white lines were higher 10 bars later, against 52.7% for white marubozu candles.

Marubozu vs. doji

A doji is the opposite shape: almost no body, with the opening and closing price nearly equal. The doji signals indecision, the marubozu signals control. Both are single candles. The marubozu never beat the rest of the market in a way that survives our correction; the doji has no expected direction, so our bench tests only its breakout trade, rated No clear edge on daily charts.

Common mistakes to avoid when trading marubozu

• Taking every marubozu as a signal. The candle shows who controlled one session, not what comes next (see the limitations below).
• Entering at the close and expecting the move to extend right away. On hourly charts, the next bar more often went against the candle than random bars did.
• Setting a stop too wide for the account. The stop covers the full candle, so a large marubozu means a small position.
• Calling a candle with a visible wick a marubozu. The definition requires very short shadows; otherwise it is a long day candle.
• Ignoring the market. Black marubozu trades lost money on stocks because the stocks in our data rose over time: the same short on random dates lost about as much (−0.12 R, against −0.14 R for the pattern).

Statistical reliability of the marubozu pattern

The marubozu is not reliable on its own. No line beat chance once corrected for multiple testing. The closest is the white marubozu on daily crypto charts (4,241 candles, an average excess return of +0.84% after 10 days, rated promising: it passes a simple test, not our correction), and even there the 2:1 trade did no better than on random dates. Many more lines did worse than chance after the correction: both colors on hourly charts and on daily US stocks and indices, and black candles also on 4-hour charts, on daily ETFs and on all daily markets combined. Here are the numbers from our test bench, which applies the same rule to full price history and compares each occurrence with random bars on the same assets and with the rest of the market.

White (bullish) marubozu

Black (bearish) marubozu

Read each table in pairs of columns. On daily charts, 52.7% of white marubozu candles were followed by a gain 10 days later. That sounds bullish until you see that random bars on the same assets rose 54.2% of the time, because these assets rose a little more often than they fell over our data. The excess return over the rest of the market after 10 days is −0.10%, with a 95% confidence interval from −0.15% to −0.04%: rated Weak, below chance: the gap below chance passes our tests, but it changes sign between the two halves of the history (see below), so we do not count it as confirmed.

The 2:1 trade column tells the same story. The white marubozu trade on daily charts has a positive expectancy, +0.07 R per trade before costs, but the same entry, stop and target on random dates averaged +0.09 R, the same result. The gain comes from the rise of these assets over our data, not from the candle. For the black marubozu, the expectancy is negative (−0.13 R) for the same reason: shorting these assets on random dates lost money over our data (−0.11 R), and the candle does not change it.

The clearest negative is on hourly forex: both the white and the black marubozu did worse than chance over 10 bars, and the gap survives our statistical correction. The black marubozu trade averaged −0.24 R against −0.02 R on random dates, rated worse than chance; the white trade, −0.07 R against 0.00 R, is rated worse than chance.

Does the larger trend help?

We split daily white marubozu candles by the 200-day moving average. Above it, 52.9% were higher 10 days later; below it, 52.3%. The 2:1 trade gave +0.07 R and +0.07 R. The long-term trend filter barely moved the result. The result was not stable over time: the excess return was −0.21% in the first half of the history and +0.02% in the second.

The method is published in our methodology.

Best timeframes for identifying marubozu signals

No timeframe showed a confirmed edge once compared with chance; the closest, white candles on daily crypto charts, passes only a simple test. Weekly charts show the best raw numbers for the white marubozu (39.5% of 2:1 targets hit on 24,588 trades, +0.19 R), but random weeks on the same assets did slightly better (40.9%, +0.23 R): the assets rose over our weekly data, and any long trade benefited. On hourly charts, marubozu candles are frequent, about one every 74 bars. On daily US stocks, a white marubozu appears about once every 58 bars on average and a black one every 62 bars, so a watchlist of 20 stocks shows about two of each per week.

Two of the subgroups that look better on raw numbers are on indices. On daily charts, white marubozu trades reached the 2:1 target in 41.4% of 5,323 trades against 40.5% on random dates (+0.24 R against +0.21 R). On weekly charts, the white marubozu was followed by an excess return of +0.42% after 10 weeks. The daily trade is rated No clear edge: its gap does not pass even a simple test. The weekly excess return is rated promising: it passes a simple test, not our correction for multiple testing. Neither counts as an edge. And on daily indices, the white marubozu's own 10-bar excess return was below chance (−0.14%, rated worse than chance).

Case studies of the marubozu

The Real examples section shows real marubozu candles picked by our bench (best and worst results, textbook shapes, most recent, random draws), with their actual outcome. The cases written up below are on daily charts, on crypto and on the Indian Sensex index.

Case study: a bullish marubozu that worked

On October 20, 2021, LINK/USDT printed a white marubozu from 25.9 to 27.49, with a high of 27.66 and a low of 25.85. The buy stop at the high was triggered the next day, with a stop at 25.85 and a 2:1 target at 31.28. Price climbed for two sessions, then reached the target on October 23. LINK was 8.4% higher 10 days after the candle.

Case study: a bearish marubozu, target hit, then a rebound

On April 16, 2002, the Sensex printed a black marubozu that opened at its high of 3,459.08 and closed at 3,413.72, just above its low of 3,411.73. The sell stop at 3,411.73 was triggered on April 18, and the 2:1 target at 3,317.03 was reached on April 29, seven sessions later. The index bottomed at 3,296.88 that day, then rebounded to 3,394.89 by May 6, and it was only 2.2% below the candle's close 10 sessions after it. The trade worked because it had a fixed target; holding for continuation would have given back most of the gain.

Case study: failed marubozu setups

On July 1, 2025, UNI/USDT printed a black marubozu with a high at 7.183 and a low at 6.59. The next day price traded below the low, triggering the short, then rallied above the high in the same session: entry and stop on the same day, a full 1 R loss, and UNI was 31.9% higher 10 days later. On September 22, 2021, UNI/USDT printed a white marubozu from 19.07 to 21.62, after a sharp two-day drop. The long trade at 21.66 had its stop at 18.85: a very wide risk. The entry was triggered the next day, and the day after, price fell to 18.17 and hit the stop. UNI then rallied to 27.23, just short of the 27.28 target, and was 21.0% higher 10 days after the candle, yet the trade lost 1 R.

Limitations and risks of the marubozu pattern

• On its own, it showed no edge over chance after our correction on any line (white candles on daily crypto come closest, without passing it), and on many, hourly charts above all, it did worse.
• The stop covers the full candle, so the risk per trade is large on the biggest candles.
• On daily charts, 11.6% of white marubozu setups were canceled before entry, and of the daily white marubozu trades that hit their stop, 24.4% had first been up 1 R.

The strengths are real but modest: the candle is easy to read, it defines a clear risk, and it shows where one side was in control. Use it as a piece of context, not as a trading system.

Frequently asked questions

Summary and final trading tips

• A marubozu is a candle with a long body and little or no wick: a white one opens at the low and closes at the high, a black one does the opposite.
• There are three types: full, opening and closing marubozu.
• On 145,183 daily white marubozu candles, 52.7% were higher 10 days later, against 54.2% for random bars on the same assets.
• The 2:1 trade averaged +0.07 R before costs on daily charts, like the same trade on random dates (+0.09 R).
• On hourly charts, the next bar more often went against the candle; on hourly forex, the black marubozu trade did worse than chance.

If you trade this candle, treat it as information about one session, not as a forecast. Define the stop at the other end of the candle before entry, size the position on that distance, and take profits at a fixed target rather than waiting for continuation.

About the author and the data. Written by Amaury, founder of PatternsWizard, who is passionate about finance, quantitative trading and data analysis. He built the PatternsWizard test bench, which detects each setup with fixed, published rules on full market history: crypto, forex, US stocks, ETFs, indices and commodities. Trades and win rates are compared with random bars on the same assets, excess returns with the rest of the market, with error margins and a correction for multiple testing, and the whole pipeline is checked on random-walk prices, where nothing can be predicted. Statistics on this page were generated on November 5, 2026. The full methodology explains every step. This page is educational and is not investment advice: trading involves a risk of loss, and past results do not predict future ones.