Takuri Candlestick Pattern: Definition, Trading Rules and Tested Results

The takuri candlestick pattern is a single candlestick with a small body at the top of its range, a very long lower shadow (at least three times the body) and little or no upper shadow, printed after a downtrend. It is a stricter version of the hammer, and technical analysis reads it the same way: a potential bullish reversal, because sellers pushed the price far down during the session and buyers took it all back before the close.

This guide covers the definition, how to identify a takuri line on candlestick charts, how traders use it in trading, and what actually happened to prices after it. Our test bench detected 10,013 takuri candlestick patterns 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 the takuri candlestick pattern?

The takuri (also called the takuri line) is a one-candlestick bullish reversal pattern: it indicates that a downtrend may be ending. The takuri candlestick appears at the bottom of a decline and looks like a hammer with an even longer handle: the opening price and the closing price sit near the high of the session, and the lower shadow hangs far below them.

Definition used on this page

Our bench uses a rule a computer can apply without judgment: after a downtrend (the previous close is at least one average true range lower than the close 10 bars earlier, and in the lower half of the range of the 40 previous closes), a short body at the top of the candlestick, a lower shadow at least 3 times the body and at least two thirds of the whole range, a very short upper shadow, and a range at least the average of the 10 previous candlesticks. Sizes are measured against the 10 previous candlesticks, as in the open-source TA-Lib library, but the definitions differ: our rule finds only 8% of the cases TA-Lib's takuri function flags. The color of the body does not matter: a white (bullish) or black (bearish) candlestick both count.

Visual representation

The chart below is not drawn by hand. It is the median shape of the takuri candlestick patterns our bench detected, with the three previous candlesticks, and the lines of the position we test: entry above the takuri high, stop loss below its low, target at twice the risk.

Anatomy of the takuri line

A takuri candlestick has three parts, and the proportions are what separate it from its neighbors on candlestick charts.

The long lower shadow

The lower shadow is the defining feature. It must be at least three times the size of the real body and cover at least two thirds of the high-low range. On the median daily takuri our bench found, the whole candlestick spans 3.72% of price, or 1.14 times the average true range (ATR) of the asset: the typical takuri is not an unusually large candle, it is a normal-sized candle with most of its length below the body.

Small or non-existent body

The real body, the distance between the opening price and the closing price, is small and sits at the top. When the body shrinks to almost nothing, the takuri candlestick becomes very close to a dragonfly doji (see the doji family below).

Little or no upper shadow

The open or the close is at, or very near, the high of the session. A long upper shadow on top of it turns the candle into a high-wave candlestick or a spinning top, which signal indecision rather than a bullish reversal. Indecision candlestick patterns indicate a pause in the trend, not a change of trend.

How to read a candlestick before you look for a takuri

Candlestick charts are traditionally attributed to Japanese rice traders, and every candlestick shows four prices for one period: the open, the high, the low and the close. The real body is the box between the opening price and the closing price. A white (or green) candlestick closes above its open and is read as bullish; a black (or red) candlestick closes below its open and is read as bearish. The thin lines above and below the body are the shadows, also called wicks: the upper shadow reaches the high of the period, the lower shadow reaches the low.

Single candlestick patterns are defined by these proportions. A long body shows strong control by one side; a small body with long shadows on both sides signals indecision; a long shadow on one side only means that one side pushed the price far and was then pushed back. The takuri belongs to this last group: almost all of it is a lower shadow, and its body is small and white or black. Reading the open, high, low and close of each candlestick, and comparing it with the previous candlesticks, is the basis of all candlestick patterns, from the single doji to three-candlestick reversal patterns such as the morning star and the evening star.

Historical context and origin of the takuri

The takuri candlestick comes from the Japanese candlestick tradition that Steve Nison introduced to Western traders in Japanese Candlestick Charting Techniques (1991). It belongs to the hammer family and is defined by the length of its wick. The TA-Lib library lists it as a separate function, CDLTAKURI, described as a dragonfly doji with a very long lower shadow. Definitions therefore vary between sources and tools: some require a real body, others a doji. Always check which one your scanner uses.

Takuri vs. dragonfly doji vs. hammer

Three candlestick patterns share the same idea of a long wick below the opening price, and they are easy to confuse.

• Hammer: small body at the top, lower shadow at least twice the body, after a decline.
• Takuri: the same, with a lower shadow at least three times the body. In practice, a takuri is a hammer with a longer shadow.
• Dragonfly doji: no real body at all (opening and closing price equal or almost equal) near the high, a long lower shadow, and no requirement of a prior decline.

The difference matters for statistics. Our bench treats the dragonfly doji as a doji with no expected side and trades the first one that breaks, while the takuri and the hammer are traded as bullish signals. On daily charts the three ended up with very similar results (see the comparison table below). A standard doji, with a cross shape and wicks on both sides, is not part of this family; a doji only becomes a dragonfly doji when the open and close sit at the high.

Takuri and the doji family

A doji is a candlestick whose open and close are equal or almost equal, so the real body is a thin line. There are four common doji types: the standard doji (short shadows on both sides), the long-legged doji (long shadows on both sides), the gravestone doji (a long upper shadow, open and close at the low) and the dragonfly doji (a long lower shadow, open and close at the high).

The takuri sits between the hammer and the dragonfly doji. With a real body, it is a long-tailed hammer; when the body becomes a doji, it is a dragonfly doji with a very long wick, which is how the TA-Lib library defines it. The doji reading changes the meaning: a doji indicates indecision, so our bench gives the dragonfly doji no expected side. On daily charts, trading the first breakout of a dragonfly doji reached its 2:1 target 35.5% of the time, against 35.0% on random dates, and its average result per trade (+0.06 R against +0.04 R) does not pass our statistical test. The takuri, traded only on the bullish side, reached 35.7% against 35.9% (see the reliability section).

Single candlestick patterns that look like a takuri

Several candlestick patterns have a small body and long shadows. Telling them apart on a chart is the most important part of reading a takuri correctly.

• Hammer: the same shape with a shorter lower shadow (at least twice the body), after a decline. Bullish.
• Hanging man: the hammer shape after an uptrend. Bearish.
• Inverted hammer: small body at the bottom and a long upper shadow, after a decline. Bullish.
• Shooting star: the inverted hammer shape after an uptrend, with a long upper shadow that shows rejected buying. Bearish.
• Dragonfly doji: open and close at the high, long lower shadow, no real body.
• Gravestone doji: open and close at the low, long upper shadow, no real body.
• Long-legged doji and high wave: long shadows on both sides, a sign of indecision.
• Spinning top: small body in the middle of the range, shadows of similar size on both sides.

The quick test for a takuri candlestick: a small body at the very top, no visible upper shadow, a lower shadow at least three times the body, and a decline before it. If any of these is missing, it is one of the shapes above.

How to identify the takuri pattern on a chart

To identify a takuri candlestick on a price chart, check four points in order. Each one removes a look-alike candlestick.

Step 1: confirm the prior decline

The takuri is a reversal pattern, so there must be a decline to reverse. A simple test: the previous close is at least one average true range below the close 10 bars earlier, in the lower half of its recent range. Without it, the shape is only a candle with a long wick.

Step 2: measure the lower shadow

Compare the lower shadow with the body itself. Two times is a hammer; three times or more is a takuri. Also check that the shadow covers at least two thirds of the high-low range.

Step 3: check the upper shadow

The high should be at, or very close to, the top of the body. If the upper shadow is long, it is not a takuri.

Step 4: wait for the close

The pattern only exists once the candle has closed. During the session, a candle can show a long wick and then fall back into the close.

Market psychology behind the takuri pattern

The takuri tells the story of one session. Price opens, sellers take control and push it far below the opening price, sometimes to a new low of the move. Then buyers absorb the selling and lift the price all the way back near the open and the high by the close. The very long lower shadow is the trace of that rejected selling: the market tested much lower prices and refused them, which classic analysis reads as a sign that control may change hands.

The same price action can be read from the other side. Sellers who pushed prices to the low of the session were trapped when the price closed near the high, and some of them may buy back their position if the price keeps rising. A large wick at the end of a long decline can also indicate exhaustion. Each of these readings indicates a potential reversal; none of them indicates how likely it is.

That reading is plausible, but it describes one candle. Whether one rejected session changes the trend is a question for data, and the answer is below.

Is the takuri pattern bullish or bearish?

By definition, the takuri is a bullish reversal pattern: it appears after a downtrend and indicates a potential turn up. The same shape after an uptrend is not called a takuri; with a lower shadow of at least twice the body it is a hanging man, a bearish reversal signal. In our data, the takuri's average excess return over the rest of the market never beat chance: on daily crypto, hourly crypto, daily commodities and weekly charts it was below chance by gaps that pass a simple test but not our correction, so "bullish" describes the textbook reading, not a measured edge. Commodity results come from continuous futures that are not adjusted at contract rolls, so a few trades span a price jump between two contracts.

How to trade the takuri candlestick pattern

The classic takuri setup buys a break above the high of the candlestick, with a stop loss below its low and a profit target set as a multiple of the risk. This is the position our bench simulates on every occurrence.

Determining the entry point

There are two classic entries. The aggressive one buys at the close of the takuri. The conservative one places a buy stop just above the high and waits for the price to move through it within the next sessions. In our tests the buy stop was triggered 79.3% of the time on daily charts; in 20.5% of cases, the price broke the low of the takuri first and the setup was canceled before entry.

Setting a stop loss for risk management

Identifying profit targets

The target we test is twice the risk (a 2:1 reward-to-risk ratio). On daily charts, positions that reached it took a median of 8 days; positions that were stopped took a median of 3 days. A previous resistance level or a falling moving average above the price is another possible target, which we have not tested.

The importance of market context

Trading takuri in a downtrend

Every takuri in our statistics follows a decline by construction. What changes is the larger trend. On daily charts, takuri lines above the 200-day moving average (a pullback in a long-term uptrend) were higher 10 days later 54.3% of the time, against 48.6% below it. The excess return over the rest of the market was +0.15% above the average, within the margin of error, and −0.29% below it, so neither group beat chance.

Confirmation signals to look for

A stricter approach waits for confirmation: a white candlestick that closes above the high of the takuri, ideally on higher volume, or a bullish engulfing pattern or morning star right after it. Classic analysis gives more weight to a takuri at a support level, such as a prior low; we have not measured support levels. The buy stop above the high is a form of confirmation, and it is the version we measured.

Confirmation can also fail. A bearish engulfing candlestick or a black candlestick that closes below the low of the takuri the next day is a bearish reversal signal in its own right: it shows that the rejection of lower prices did not hold. A trader who waits for a second white candlestick, or for a pattern such as three white soldiers, gets a later entry and a wider stop, so the reward-to-risk ratio of the trade changes.

Reliability: what the data shows

The takuri candlestick pattern is not reliable on its own: across every market and timeframe we tested, its average excess return over the rest of the market never beat chance, and on daily crypto it was below chance by a gap that passes a simple test but not our correction. Here are the numbers from our bench.

Table quoted in the text (verdicts and every market: see All measurements)

Market and timeframe	Takuri tested	Higher 10 bars later	Random bars	Closed trades	2:1 target hit	Same setup, random dates	Expectancy (random)
All markets, daily	10,013	50.6%	53.3%	7,933	35.7%	35.9%	+0.07 R (+0.08 R)
US stocks, daily	8,294	51.5%	54.1%	6,504	36.0%	36.1%	+0.08 R (+0.09 R)
Crypto, daily	972	39.7%	43.9%	851	30.3%	34.0%	−0.08 R (+0.03 R)
Crypto, 4-hour	5,145	50.9%	47.4%	4,401	33.5%	33.4%	+0.02 R (+0.02 R)
Crypto, hourly	25,737	48.7%	48.2%	22,504	29.8%	32.6%	−0.09 R (−0.01 R)
Forex, hourly	2,715	48.3%	50.7%	2,362	30.7%	33.3%	−0.08 R (0.00 R)
Indices, daily	296	59.7%	57.6%	248	45.2%	36.7%	+0.34 R (+0.10 R)
All markets, weekly	1,170	57.2%	57.9%	1,010	35.3%	40.3%	+0.06 R (+0.22 R)

Read the table in pairs of columns. On daily charts, 50.6% of takuri patterns were followed by a gain 10 days later, against 53.3% of random days on the same assets. The excess return over the rest of the market after 10 days is −0.16%, with a 95% confidence interval from −0.32% to +0.01%: zero is inside, so there is no clear edge on direction.

The 2:1 setup tells the same story. On daily charts it reached its target 35.7% of the time against 35.9% for the same entry, stop and target on random dates, for an expectancy of +0.07 R per trade before costs against +0.08 R on random dates. That result is rated no clear edge: the positive expectancy comes from assets that rose over their history, not from the takuri, and costs would cut it to +0.04 R. On hourly forex the result is slightly worse than chance: 30.7% against 33.3%, with an expectancy of −0.08 R against 0.00 R, a gap that passes a simple test but not our correction for the thousands of tests we run. Daily indices look better (45.2% against 36.7%, or +0.34 R against +0.10 R per trade) on only 248 trades, but that gap passes only a simple test, not our correction for multiple testing: rated promising, not proven.

What changed the results

These breakdowns are descriptive, not corrected for the number of tests. On daily charts, takuri patterns printed with volume above 1.5 times the 20-day average reached the 2:1 target 35.8% of the time, against 35.5% on normal volume. The 200-day moving average changed the hit rate little: 37.7% above it, 34.6% below, for +0.13 R and +0.04 R per trade, a descriptive split we did not test. The excess return over the rest of the market after 10 days was negative in both halves of the history (−0.07% and −0.25%).

Timing of the setup

Takuri setups are decided fast. On daily charts the median winner reached its target in 8 bars and the median loser was stopped in 3 bars. And of the positions that hit their stop, 23.7% had first been up one full R.

The full table, with every market, timeframe and verdict, is in the statistics section of this page. The method is published in our methodology.

Takuri vs. other bullish and bearish candlestick patterns

Here is how the takuri compares with related candlestick patterns on the same bench, all markets, daily charts. Each pattern is traded the same way: a stop order at the high (bullish) or low (bearish) of the pattern, a stop at the other end, a target at twice the risk.

Two lessons. First, the bullish signals have higher hit rates and expectancies than the bearish ones mainly because most of the assets we test rose over their history, as the random columns show too: compare each row with its random column, not with the other rows. Second, the takuri does not improve on the plain hammer: requiring a longer wick did not make the signal better. None of these candlestick patterns beat its random baseline on the 2:1 trade after our correction, and the bearish ones still have a negative expectancy.

The table also explains why candlestick patterns are grouped as bullish and bearish reversal patterns rather than judged one by one. The bullish ones (takuri, hammer, inverted hammer, piercing line, morning star, bullish engulfing pattern, three white soldiers) all appear after a decline in price and expect the price to rise; the bearish ones appear after an uptrend and expect the price to fall. In both groups, the result after the signal stayed close to the result on random dates on the same chart. To learn how each one is detected and traded, follow the links in the table.

Limitations of the takuri pattern

• On its own, it showed no confirmed edge on any market or timeframe we tested: the best-looking row, daily indices, passes only a simple test, not our correction, and on daily crypto it did worse than chance, by a gap that does not survive our correction either.
• The stop is wide relative to the body, so position sizing matters.
• On daily charts, 20.5% of setups are canceled before entry, and when the position does enter, the median time to the stop is only 3 bars.
• Definitions differ between books and scanners, so two traders can disagree on whether a candlestick is a takuri.

Combining takuri with technical indicators

Using the relative strength index

One rule is to take a takuri only when the RSI is oversold (below 30), on the idea that the decline is stretched. We have not tested an RSI filter: if you use one, test it on your own market first.

Using moving averages for trend validation

Moving averages are the filter we did test. On daily charts, the takuri setup above the 200-day moving average reached the 2:1 target 37.7% of the time with an expectancy of +0.13 R, against 34.6% and +0.04 R below it. This split is descriptive and not corrected for multiple testing.

Volume analysis and the takuri

Classic analysis reads high volume on the takuri as buyers stepping in with size. On daily charts it helped the setup a little (35.8% of targets hit against 35.5%), but the move 10 days later was not better: 49.8% of high-volume takuri lines were higher, against 50.9% on normal volume. Treat volume as a mild plus, not a filter that changes the odds.

A trader's checklist for the takuri

Before you trade a takuri, go through this short list, in the order a trader meets the points on a live chart.

1. Is there a downtrend before the candlestick? If not, it is not a takuri.
2. Is the lower shadow at least three times the body, with almost no upper shadow?
3. Is the candlestick at a support level or after a significant decline? Support is a classic part of the reading; we have not measured whether it changes the odds.
4. Where is the entry (above the high), the stop (below the low) and the target? Write them down before the trade is started.
5. Does the position size limit the loss to an amount you accept if the stop is hit?

A scanner with predefined rules applies its shape test without judgment, but each scanner has its own: the TA-Lib function wants a doji, not a body with a shadow 3 times as long. The descriptions in candlestick books are looser, so a trader who learns the pattern from a chart may see more takuri lines than a scanner. Our data gives no reason to trade the takuri as a system on its own; whether it helps as one piece of evidence among several is something we have not tested.

Common mistakes when trading the takuri

• Entering before the candle closes: a long wick in the middle of the session is not a takuri yet.
• Ignoring the prior trend: without a decline, the shape is not a reversal signal.
• Placing the stop inside the shadow: the low of the candlestick is the level that proves the pattern wrong.
• Trading it in isolation: on its own, its direction 10 days later was no better than a random day on the same asset.
• Picking only the examples that worked: looking back at candlestick charts, the takuri lines that failed are easy to forget.

Takuri patterns across different timeframes

Day trading with the takuri

The takuri is frequent on intraday charts: about once every 592 bars on hourly charts. On hourly crypto (25,737 occurrences) the average excess return was slightly below zero, and on hourly forex the 2:1 trade did slightly worse than chance, two gaps that pass a simple test but not our correction. A day trader also faces spreads and fees that are large compared with a one-hour candle.

Swing trading and long-term analysis

On daily charts it appears about once every 847 bars, and on weekly charts once every 1,328. Weekly takuri candles reached the 2:1 target 35.3% of the time, less often than random weeks on the same assets (40.3%), and their move 10 weeks later was rated weak, below chance.

Takuri patterns in forex, crypto, and stocks

On US stocks (8,294 daily occurrences on 1,494 companies since 1970, 6,504 closed trades), the takuri setup reached its target 36.0% of the time against 36.1% on random dates; its average result per trade (+0.08 R against +0.09 R) is no better than chance. On crypto (972 daily occurrences on 395 coins since 2018), 39.7% were higher 10 days later against 43.9% for random days, and the average excess return (−1.07%) is rated weak, below chance, a gap that passes a simple test but not our correction. On forex (2,715 hourly occurrences on 25 currency pairs since 2012), the setup did slightly worse than random entries, a gap that passes a simple test but not our correction. No market gave it a confirmed edge.

Takuri case studies

The gallery in the Real examples section shows real takuri candlestick patterns picked by our bench, with their actual outcome. The case studies here come from our bench's detections on daily crypto charts.

Case study: higher 10 days later, still stopped

On September 26, 2019, ETH/USDT printed a takuri with a high at 171.01 and a low at 152.11, after a decline. The buy stop above the high was triggered the next day, with a stop at 152.11 and a 2:1 target at 208.81. Ten days after the takuri, ETH was 2.5% above its closing price, but the target was never reached and the stop was hit on November 22, 2019: a full 1 R loss.

Case study: canceled before entry

On January 19, 2026, DOGE/USDT printed a takuri with a high at 0.13163 and a low at 0.12. The price never traded above the high: on January 25 it broke below the low, and the setup was canceled before entry. Ten days after the takuri, DOGE was 9.2% below its closing price: the buy stop kept the trader out of the decline.

Case study: a failed takuri

On April 25, 2022, UNI/USDT printed a takuri with a high at 8.78 and a low at 8.16, small body at the top of a long lower shadow. The next day traded both above the high, to 8.82, and down to 7.91, through the low; our bench counts such a bar as an entry followed by the stop: a full 1 R loss. Ten days after the takuri, UNI was 15.9% lower. More real cases are shown in the gallery of the Real examples section.

Frequently asked questions about the takuri pattern

Is the takuri pattern reliable?

Not on its own. On 10,013 daily occurrences, 50.6% were higher 10 days later, against 53.3% for random days on the same assets, and the average excess return (−0.16%) is within the margin of error.

What does the long lower shadow represent?

It shows that sellers pushed the price far below the open during the session and that buyers brought it back near the high by the close: lower prices were tested and rejected.

How does the takuri differ from a hammer?

Only by the length of the lower shadow: at least three times the body for the takuri, at least twice for the hammer. In our data the two had almost identical results.

Can the takuri appear in an uptrend?

The same shape can appear after a rally, but it is then read as a bearish warning, not as a takuri.

What is the best timeframe for the takuri?

No timeframe gave an edge. On hourly forex, the 2:1 trade did slightly worse than chance, rated weak, below chance, a gap that passes a simple test but not our correction; on weekly charts the move 10 weeks later was below chance with the same caveat, and on daily crypto it was rated weak, below chance.

Summary of the takuri candlestick pattern

• A takuri has a small body at the top, a lower shadow at least three times the body and little or no upper shadow, after a downtrend.
• It is a stricter hammer; the same shape after a rally is read as bearish.
• On 10,013 daily occurrences, 50.6% were higher 10 days later, against 53.3% for random days.
• The 2:1 setup reached its target 35.7% of the time, against 35.9% for the same setup on random dates, for +0.07 R against +0.08 R per trade before costs: no edge over chance.
• A longer wick did not make it a better signal than the hammer.

Further reading on candlestick charts

To learn more about the hammer family, read our guides to the hammer, the inverted hammer and the dragonfly doji. For the origins of candlestick charts in the West, Steve Nison's Japanese Candlestick Charting Techniques (1991) remains the standard reference.

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, not financial advice: trading can lose money, and past results do not guarantee future ones.