A long-legged doji is a candlestick with almost no real body and long upper and lower shadows: the opening and closing prices are nearly equal, while the market traveled above and below them by more than the recent shadows. It reflects indecision in the market, not a bullish or bearish signal on its own, and it appears in every financial market.

This article explains how to identify the long-legged doji candlestick pattern, what it says about buyers and sellers in technical analysis, how it differs from other doji patterns, and how to trade it. It also answers the question that matters for a trader: what happened next. Our research bench applied the same rule to full market history, which gave 100,516 occurrences in daily data alone across stock, crypto, forex, ETF, index and commodity markets, and compared the outcome with random dates on the same assets.

What is a long-legged doji?

The long-legged doji is a single-bar pattern from Japanese candlestick analysis, sometimes written long legged doji without the hyphen. Steve Nison, who introduced Japanese candlestick charts to Western traders in Japanese Candlestick Charting Techniques (1991), describes it as a doji with very long upper and lower shadows, and treats it as a sign that the market has lost its sense of direction. Like every doji candlestick pattern, its open and close are at, or very close to, the same price.

Our bench writes the rule so a computer can apply it without judgment: a doji (a body of at most 10% of the average range of the 10 previous candles) whose upper and lower shadows are both longer than the average half-shadow of those 10 candles, whose whole range is at least the average range of those candles, and whose shorter shadow is at least a third of the longer one. No prior trend is required. Sizes are relative to recent candles, as in the open-source TA-Lib library, but our shadow test is stricter: on the bars compared, we find 30.4 percent of the long-legged dojis that library flags.

How to identify a long-legged doji candlestick

To identify a long-legged doji on a chart, check three points once the session is over: the gap between the opening and closing prices, the two shadows, and the size of the whole range compared with recent sessions.

Long upper shadow> avg half-shadow, last 10
Doji bodyopen near close
Long lower shadow> avg half-shadow, last 10

Median shape of 1,349 daily occurrences drawn at random from all markets, those with the candle colors drawn here (45% of the 3,000 drawn), found by our test bench.

Visual structure and characteristics

The doji looks like a cross with extended arms up and down. The figures below are not drawn by hand: they are the median shape of the clean occurrences our bench detected, with the faded candles showing what came before, and the lines of the two trades we test, one on each side.

Bullish setup on the median shape of 1,349 daily occurrences drawn at random from all markets, with entry, stop loss (SL) and 2:1 target (TP)

Bearish setup on the median shape of 1,349 daily occurrences drawn at random from all markets, with entry, stop loss (SL) and 2:1 target (TP)

Defining the shadows and the body

Real body: a thin line or a tiny rectangle. The opening price and the closing price are almost the same, so the color carries little information.
Upper shadow: long, showing that buyers pushed prices well above the open at some point.
Lower shadow: long as well, showing that sellers also pushed prices well below it.
Position of the real body: anywhere in the middle half of the range under our rule, since the shorter shadow must be at least a third of the longer one. When it sits exactly in the middle, it is called a rickshaw man.

One detail is easy to miss. In daily data, the median long-legged doji our bench found spans 2.86% of price, or 1.09 times the average true range (ATR) of the asset. The legs are extended compared with recent shadows, but the whole candle is about the size of an average session.

The psychology behind the long-legged doji

Market indecision explained

The candle tells the story of one session. Price opens, moves one way, then the other way, and the closing price comes back to the opening level. Neither side kept control. The two shadows are the trace of two rejected moves: buyers were rejected at the high and sellers at the low.

The battle between buyers and sellers

Buyers and sellers both had a turn and both failed to hold it. Classical candlestick analysis (Nison) reads that balance as a pause: after a strong trend, it may indicate that momentum is fading and that a potential reversal is near; in a range, it simply reflects a market that has not chosen a side. That reading is plausible, but it describes one candle. Whether the indecision leads anywhere is a question for data, which we answer in the reliability section.

Long-legged doji versus other doji patterns

All doji patterns share the same feature: an opening price and a closing price nearly equal. What changes is the length of the upper and lower shadows. Here is how the long-legged doji compares with related patterns, with each one's 2:1 result in daily data, taken from its own page.

Comparison with dragonfly doji

The dragonfly doji has a long lower shadow and almost no upper shadow: open, top and close at the same price. Its 2:1 setup hit the target in 35.5% of 58,832 trades against 35.0% on random dates, and its average result per trade (+0.06 R against +0.04 R) does not beat random dates by a margin that passes our statistical test.

Comparison with gravestone doji

The gravestone doji is the mirror image: a long upper shadow and almost no lower one. Its 2:1 setup hit the target in 32.0% of 49,732 trades against 32.2% on random dates: about the same hit rate, and the average result per trade does not beat random dates by a margin that passes our statistical test.

Comparison with a standard doji

A standard doji candlestick only requires a close almost equal to the open; its shadows can be of any length. The long-legged doji is the subset with extended shadows on both sides. Over 1,194,453 daily doji trades, the 2:1 setup hit the target in 33.7% of cases against 33.5% on random dates, rated No clear edge: its average result per trade is not statistically different from random dates.

Comparison with a spinning top

A spinning top has a small but visible real body, with shadows longer than it. The long-legged doji has almost none. Both are indecision patterns, and both showed the same result: the spinning top hit its target in 33.4% of 787,847 trades against 33.4% on random dates, with no measurable difference in average result per trade (−0.01 R against −0.01 R). The high wave candle, with both shadows at least three times the body, sits between the two.

Significance of the long-legged doji in market trends

In classical candlestick analysis, context gives the pattern its meaning: the same doji candlestick reads differently at the top of a rally, at the bottom of a decline and in the middle of a range.

Long-legged doji after an uptrend

When a long-legged doji forms after a strong uptrend, near a resistance level, the classical reading sees a warning that buyers are losing control: they pushed prices up, could not hold them, and the close came back to the open. It can be the middle candle of a bearish reversal candlestick pattern such as the evening star. In daily data, occurrences after a rise hit the 2:1 target in 32.9% of cases, for an expectancy of −0.02 R.

Long-legged doji after a downtrend

When it forms after a downtrend, near a support level, the same doji is read as a sign that sellers are losing momentum, sometimes the middle candle of a bullish reversal pattern, the morning star. In daily data, occurrences after a decline hit the target in 33.5% of cases, for an expectancy of 0.00 R. The gap between the two groups is small, and these breakdowns are descriptive, not corrected for the number of tests.

Trading strategies for the long-legged doji

Because the pattern points neither up nor down, one approach in doji trading is to let the next candles decide. These trading strategies use the doji range as the trigger.

Buy stop: At the pattern high, stop loss at the low
Sell stop: At the pattern low, stop loss at the high
Target (TP): Twice the risk (2:1), on the side hit first

Whichever order is hit first is traded, the other is canceled. The trade our bench tests, drawn on the typical shape. Median shape of 1,349 daily occurrences drawn at random from all markets, those with the candle colors drawn here (45% of the 3,000 drawn).

A clean case from our bench: US stocksChosen on its shape alone, never on the outcome. Stock prices are adjusted for splits and dividends, so they can differ from the prices quoted at the time.

TextbookPANW August 12, 2015Daily chartPrice fell 1.8% after 10 bars2:1 trade: stopped out, −1.00 R

Trading the breakout of the doji range

This is the setup our bench simulates on every occurrence: a buy stop just above the high of the doji and a sell stop just below its low, valid 20 bars. The first side to break is traded, and the break itself serves as confirmation, with a stop loss at the opposite end of the candle and a target at twice the risk. In daily data one side almost always broke: the entry was triggered on 99.9% of occurrences.

Confirmation using price action

A stricter entry waits for confirmation through price action: a candle that closes beyond the range of the doji, not just a touch. A close gives a later entry, further from the stop, and filters out some false breaks. Our statistics measure the stop order entry, not the close, so consider a confirmed close as a different, untested rule.

Using volume as a confirmation tool

Classical candlestick analysis reads higher volume on the doji as a stronger clue. We measured it: in daily data, occurrences with volume above 1.5 times the 20-day average hit the 2:1 target in 33.3% of cases (0.00 R), against 33.2% on normal volume (−0.02 R). On hourly crypto, 31.5% against 31.1%. Volume barely changed the result.

Integrating technical indicators with the long-legged doji

Combining with support and resistance zones

Classical candlestick analysis gives more weight to a long-legged doji that forms at support or resistance already marked on the chart: a prior swing, a round number, a gap. The logic is that the indecision happens where a decision on the trend is expected. We have not measured the effect of support or resistance zones.

Using relative strength index (RSI) for confluence

One rule takes the pattern only when the RSI, a momentum indicator, is overbought (above 70) or oversold (below 30). We have not tested this indicator filter: if you use it, test it on your own market first.

Applying moving averages for trend context

Moving averages are the filter we did test. In daily data, occurrences above the 200-day moving average hit the 2:1 target in 32.9% of cases (−0.02 R), against 33.7% below it (0.00 R). The larger trend changed the target-hit rate by about one percentage point.

Bollinger band extremes and volatility

Another filter looks for the doji outside the Bollinger Bands, where prices are stretched and volatility is high. This is a matter of confluence between technical indicators, and we have no data on it.

Risk management when trading doji patterns

Setting stop-loss orders correctly

Determining profit targets

The target we test is twice the risk; another option is the next support or resistance level, which we have not tested. In daily data, trades that hit the 2:1 target took a median of 5 bars; stopped trades took a median of 3 bars.

Calculating risk-to-reward ratios

With a 2:1 target, a position must win more than one time in three to break even before costs. The long-legged doji hit that target in 33.2% of 100,256 daily trades: break-even, with fees and slippage still to pay. Also note that, of the trades that hit their stop, 24.2% had first reached +1 R, which is the argument for a partial exit or a trailing stop.

Advantages and limitations of the pattern

Why the pattern appeals

Easy to spot: a cross with extended shadows on both sides.
Clear levels: the two ends of the candle define both the entry and the stop.
Frequent: roughly one occurrence every 84 bars in daily data and every 78 on hourly bars.

Common pitfalls and false signals

No bias: the pattern does not say which way prices will break.
No edge on its own: in daily data the breakout setup did almost exactly as well as random dates.
On hourly crypto, the breakout did slightly worse than chance in average result per trade, a difference confirmed by our correction.
Calling a reversal before price confirms it, or reading every cross as a signal, including those in quiet sideways market conditions.
Trading every occurrence the same way, whatever the trend and market conditions around it.

Is the long-legged doji reliable?

Not on its own. On most markets and intervals we tested, the breakout of a long-legged doji hit its 2:1 target about as often as the same trade placed on random dates; on hourly crypto charts, and on all markets combined on hourly charts, it did worse, a gap that holds after our correction for multiple testing; a few other rows pass a simple test but not our correction, such as weekly ETFs (weak, below chance); and no row showed an edge over chance that holds after our correction. Here are the numbers from our test bench, which applies the same rule to full history.

33.2%vs 33.0% at random

Hit the 2:1 target, compared with the same trade on random dates.

Measured on 100,256 closed trades on 100,516 daily long-legged doji occurrences across 2,338 assets. The same trade on random dates on the same assets is the honest baseline.

No clear edge

Long-legged doji against random bars on the same assets, by market, on daily candles. Bars of almost the same length mean no edge.

2:1 trade hit its targetLong-legged doji tradeSame trade, random dates

All markets
All markets, Long-legged doji trade: 33.2%
All markets, Same trade, random dates: 33.0%
Crypto
Crypto, Long-legged doji trade: 35.1%
Crypto, Same trade, random dates: 33.4%
US stocks
US stocks, Long-legged doji trade: 33.0%
US stocks, Same trade, random dates: 32.8%
Forex
Forex, Long-legged doji trade: 34.8%
Forex, Same trade, random dates: 32.3%
ETFs
ETFs, Long-legged doji trade: 33.1%
ETFs, Same trade, random dates: 34.4%
Indices
Indices, Long-legged doji trade: 35.0%
Indices, Same trade, random dates: 36.9%
Commodities
Commodities, Long-legged doji trade: 31.3%
Commodities, Same trade, random dates: 36.8%

All statistics

Full results for the long-legged doji, market by market

Every occurrence our rule finds on full price history, compared with what the same assets did on random bars. A pattern only earns an edge if it beats random entries after correcting for the thousands of tests we run.

Our verdict

As a 2:1 trade: Not a reliable signal on its own here. Use it as context, not as a trigger.

2:1 trade No clear edge

Long-legged doji: 100,516 occurrences tested

Neutral pattern · All markets · Daily · 2,338 assets · 1970-03 to 2026-10

2:1 trade hit the target

33.2%

vs 33.0% for the same trade on random dates

Average result per trade

−0.01 R

vs −0.02 R on random dates

What this means. This pattern has no expected direction, so only the 2:1 trade is compared with the same trade on random dates. For the 2:1 trade: no measurable difference from the same trade on random dates on the same assets.

Timeline

Entry triggered, after the pattern: 1 day · middle half 1 to 1 · on 100,449 triggered entries; Stop order beyond the pattern's extreme. 193 of these entries are not among the 100,256 trades: still open at the end of our data, or left out by the trade rules.
Time to the 2:1 target, when hit: 5 days · middle half 3 to 11 · on 33,279 trades that hit it
Time to the stop, when hit: 3 days · middle half 1 to 6 · on 66,735 trades that hit it
Time in the trade, all exits: 4 days · middle half 2 to 8 · on 100,256 trades

Size and trade path

Pattern range, high to low: 2.9% · middle half 2.0% to 4.3% · on 100,516 occurrences
Range vs average bar (ATR): 1.09× · middle half 1.00× to 1.24× · on 100,516 occurrences
Best open profit during the trade: 0.98 R · middle half 0.30 R to 2.10 R · on 100,256 trades
Stopped trades that were first up 1 R: 24.2% · on 66,735 stopped trades; Trades a stop moved to breakeven at +1 R would have closed flat instead of at a loss
Entry order triggered: 99.9% · on 100,516 occurrences

Medians, each on the cases named on its line; "middle half" spans the 25th to 75th percentile; no value under 30 cases.

How we detect it. Doji whose upper and lower shadows are both longer than the average half-shadow of the previous 10 candles. The candle's range is at least the average range, and the shorter shadow is at least a third of the longer one. Candlestick counts are not checked against planted examples, so they may not be exhaustive.

All measurements (20 rows)

2:1 trade: target hit and average result

R = initial risk of the trade. Entry by an order at the trigger level, filled at the open when a bar opens beyond it. Exit at the stop or the target, or at the open when a bar opens beyond either one; otherwise at the close after 100 bars. When one bar touches both the stop and the target, the loss is counted. No fees or slippage. n/a = not measured. Per trade: average result, against the same trade on random dates. Figures in gray: fewer than 30 closed trades, not interpreted.
Market	Trades	Target hit	Same trade, random dates	Per trade	2:1 trade verdict
Crypto · 1-hour	183,675	31.2%	33.2%	−0.06 R vs +0.01 R	Opposite of the textbook
Forex · 1-hour	32,923	32.5%	32.5%	−0.02 R vs −0.02 R	No clear edge
All markets · 1-hour	216,598	31.4%	33.1%	−0.05 R vs 0.00 R	Opposite of the textbook
Crypto · 4-hour	44,777	32.4%	33.2%	−0.02 R vs +0.01 R	No clear edge
Forex · 4-hour	7,365	33.6%	33.0%	+0.01 R vs −0.01 R	No clear edge
All markets · 4-hour	52,142	32.6%	33.1%	−0.01 R vs 0.00 R	No clear edge
Crypto · Daily	6,530	35.1%	33.4%	+0.07 R vs +0.02 R	No clear edge
US stocks · Daily	87,899	33.0%	32.8%	−0.02 R vs −0.02 R	No clear edge
ETFs · Daily	2,780	33.1%	34.4%	−0.03 R vs +0.01 R	No clear edge
Indices · Daily	1,256	35.0%	36.9%	+0.04 R vs +0.10 R	No clear edge
Forex · Daily	954	34.8%	32.3%	+0.05 R vs −0.02 R	No clear edge
Commodities · Daily	837	31.3%	36.8%	−0.08 R vs +0.10 R	Weak, below chance
All markets · Daily	100,256	33.2%	33.0%	−0.01 R vs −0.02 R	No clear edge
Crypto · Weekly	716	35.9%	32.2%	+0.09 R vs −0.02 R	Not enough data
US stocks · Weekly	14,801	33.9%	33.8%	+0.02 R vs +0.02 R	No clear edge
ETFs · Weekly	496	31.5%	36.4%	−0.06 R vs +0.09 R	Weak, below chance
Indices · Weekly	239	40.6%	36.9%	+0.21 R vs +0.11 R	No clear edge
Forex · Weekly	188	31.4%	30.1%	−0.06 R vs −0.09 R	No clear edge
Commodities · Weekly	171	38.6%	38.3%	+0.15 R vs +0.15 R	No clear edge
All markets · Weekly	16,611	34.0%	33.8%	+0.02 R vs +0.02 R	No clear edge

Data updated November 5, 2026 · code version c2a1683f · Read the methodology. We measure every pattern on crypto and forex (1-hour, 4-hour, daily and weekly charts) and US stocks, ETFs, indices and commodities (daily and weekly charts). A market or timeframe from that list missing from the tables above had no occurrence.

What you often read

“The long-legged doji signals indecision before a big move.”

What our data shows

On 100,256 daily closed trades, the 2:1 trade taken on the side of the first breakout hit its target 33.2% of the time, against 33.0% for the same trade on random dates: no clear edge, since the average result per trade is not statistically significant.

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

Market and interval	Closed trades	2:1 target hit	Same trade, random dates	Expectancy	Verdict
All markets, daily	100,256	33.2%	33.0%	−0.01 R	No clear edge
US stock market, daily	87,899	33.0%	32.8%	−0.02 R	No clear edge
Crypto, daily	6,530	35.1%	33.4%	+0.07 R	No clear edge
Crypto, hourly	183,675	31.2%	33.2%	−0.06 R	Worse than chance
Crypto, 4-hour	44,777	32.4%	33.2%	−0.02 R	No clear edge
Forex, hourly	32,923	32.5%	32.5%	−0.02 R	No clear edge
All markets, weekly	16,611	34.0%	33.8%	+0.02 R	No clear edge

Read the table in pairs of columns. In daily data, the breakout trade hit its 2:1 target in 33.2% of 100,256 trades, against 33.0% for the same entry, stop and target placed on random dates of the same assets. The expectancy, the average result per trade in units of risk (R) before costs, is −0.01 R: no clear edge. Because the pattern has no expected direction, we do not report a "moved as expected" rate.

Daily crypto looks better at first sight (35.1% against 33.4%, +0.07 R), but on 6,595 occurrences (6,530 closed trades) the gap in average result per trade does not pass our test, so we do not count it as an edge. On hourly crypto, the breakout did worse than random dates (31.2% against 33.2%), a difference that holds after our correction for the thousands of tests we run. The full table, with every market and interval, is in the statistics section of this page, and the method is in our methodology.

Real examples of the long-legged doji

The gallery in this section shows real long-legged dojis picked by our bench (best and worst results, textbook shapes, most recent, random draws), with their actual outcome. The cases written up in this section are on daily crypto charts.

Example: a breakout that worked

On July 19, 2018, ADA/USDT printed a long-legged doji after a sharp rally: open at 0.18061, close at 0.18191, high at 0.19217 and low at 0.17149, with shadows of about the same length on each side. The next day price broke the low, triggering a sell entry at 0.17149 with a stop at 0.19217 and a target at 0.13013. The target was hit on August 2, and ADA was 10.6% lower 10 days after the doji.

Example: two breakouts that failed

On November 2, 2022, UNI/USDT printed one between 6.83 and 7.38. The following session, price broke the top, triggering a buy entry at 7.38, and on November 4 it dipped to 6.81 and hit the stop at 6.83: a full 1 R loss. UNI then collapsed and was 21.7% lower 10 days after the doji, so the stop did its job. On October 26, 2021, LINK/USDT printed one between 31.31 and 33.82 after a rally. The next session traded from 27.91 to 33.91, through both extremes: the buy entry at 33.82 triggered, and the bench counts the touch of the bottom that same day as a 1 R loss. LINK was virtually unchanged 10 days after the doji (+1.0%).

Best practices for backtesting your strategy

Before you put money from your trading account on this pattern, run the same test we run, at a smaller scale:

write the rule down: what counts as an extended shadow, which side you trade, where the stop and target go;
mark the last 50 occurrences on your own chart, winners and losers alike;
record each result in R, then place the same trade on random dates and compare;
apply the same rules on every chart, without keeping only the setups that look good afterwards.

TradingView can flag candlestick patterns, and the open-source TA-Lib library has a CDLLONGLEGGEDDOJI function. Check the definition your tool uses: thresholds for an extended shadow differ, and so do the results.

Frequently asked questions

Is a long-legged doji a bullish or bearish signal?

Neither. It is an indecision candlestick pattern. It becomes a bullish or bearish signal only through context and the next candles, and in our data, on daily charts, the first breakout did as well as a random trade; on hourly crypto it did worse.

Does the color of the doji candle matter?

Very little. The open and close are so close that the difference between a green and a red one is a few ticks. Our statistics count both colors together.

How often does the long-legged doji occur?

Often: about one candle in 84 on daily bars, one in 78 on hourly bars and one in 92 on weekly bars.

Can this pattern be used for day trading?

It appears on intraday charts too. On hourly crypto, its breakout hit the target in 31.2% of 183,675 trades against 33.2% for random dates, and on hourly forex 32.5% on 32,923 trades against 32.5%. Neither result is a positive edge, and on hourly crypto the setup did worse than chance, a gap in average result per trade confirmed by our correction.

How does volatility impact this pattern?

A long-legged doji is a session with long shadows that ends where it started, and the median candle is about the size of an average session. Since the stop sits at the end of a wick, a wider candle means a wider stop and a smaller position for the same loss at the stop.

Is the long-legged doji reliable in sideways markets?

We have not measured this split. In classical candlestick analysis, a doji in the middle of a range adds little, while one at the edge of the range marks a level to watch.

What does a long shadow imply for future price movement?

It shows that prices were tested and rejected during the session. It does not imply an upward or downward move: in daily data, trades after a prior rise and after a prior decline both stayed close to break-even (−0.02 R after a rise, 0.00 R after a decline).

Can I use this pattern on all timeframes?

Yes, the definition is the same on every timeframe. Hourly charts stood out, in the wrong direction: on hourly crypto, the average result per trade was −0.06 R against +0.01 R, a gap that survives our correction; 4-hour charts showed a smaller gap of the same sign, which our tests do not separate from chance. On weekly bars, all markets together, the average result (+0.02 R against +0.02 R) did not differ measurably from chance.

How to distinguish between a reversal and a continuation?

Wait for the break and read it against the trend. A close above the doji in an uptrend, or below it in a downtrend, is read in classical analysis as continuation; a close the other way as a potential reversal. We have not tested this reading, and there is no guarantee either way.

Summary

A long-legged doji has almost no body and long upper and lower shadows: open and close are nearly equal, and both shadows are longer than usual.
This candle signals indecision, not a bullish or bearish move.
On 100,516 daily occurrences (100,256 closed trades), the breakout setup hit its 2:1 target in 33.2% of cases, almost exactly as often as the same trade on random dates.
On hourly crypto it did slightly worse than chance (31.2% against 33.2%).
Volume, the prior trend and the 200-day moving average changed the target-hit rate by about one percentage point at most.

If you trade this pattern, treat it as a location for a trade, not a signal on its own. Define the stop before entry, size the position on that distance and keep the risk per trade small. Nothing in this content is investment advice, and past results are no guarantee for future investment decisions: trading involves risk of loss, and no candlestick pattern removes it.

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 applies the same published rules to 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.

How we tested this pattern

Every pattern's results on one sheet