Gravestone Doji: Meaning, Formation, and What the Data Shows

The gravestone doji is a single candlestick whose opening, closing and low prices are at or near the same level, with a long upper shadow above them. It looks like an upside-down "T", or a gravestone. In technical analysis it is read as a potential bearish reversal pattern when it appears after an uptrend: buyers pushed the price higher during the session, then sellers drove it all the way back to the opening price by the close.

This guide explains how the gravestone doji pattern forms in technical analysis, how to identify it on a chart, how traders use it, and how it compares with the dragonfly doji and the shooting star. It also answers one more question: what happened after the pattern. We applied the same rule to full market history, which gave 49,807 occurrences on daily charts alone across stocks, crypto, forex, ETFs, indices and commodities, and compared the outcome with random dates on the same assets.

What is a gravestone doji?

A gravestone doji is a one-candle candlestick pattern from the doji family. A doji is a candlestick with almost no body: the opening and closing prices are nearly the same. The gravestone is the doji formation where that open and close sit at the bottom of the range, so the only visible part is a long upper shadow.

Steve Nison, who introduced Japanese candlestick charting to Western traders in Japanese Candlestick Charting Techniques (1991), presents the gravestone doji as a bearish signal at market tops: the market opened, rallied, and could not hold any of the gain. Like every doji pattern, it is first a sign of indecision. Its bearish meaning as a potential trend reversal comes from its location, after a rise, and from the rejection of higher prices that the upper shadow records. That is why it belongs to the reversal patterns of technical analysis rather than to the continuation patterns.

Anatomical structure of the gravestone doji

Identifying the shape and characteristics

A gravestone doji candlestick pattern has three features:

• No real body, or a very thin one: the opening and closing prices are the same or almost the same.
• A long upper shadow: the high of the session is well above the open and the close.
• No lower shadow, or a very short one: the low of the session is at or near the opening price.

The color does not matter. With no body to speak of, a green or red gravestone doji carries the same information.

Our test bench writes those features as a rule a computer can apply without judgment. The body must be at most 10% of the average range of the 10 previous candles (the standard doji definition of the open-source TA-Lib library), the lower shadow must be very short, and the upper shadow must reach at least half of that average range. That is stricter than TA-Lib's own gravestone function: our rule finds 44% of the cases it flags. On the median occurrence we found on daily charts, the whole pattern spans 2.01% of price, less than the average true range (ATR) of the asset (the statistics card gives the ratio): a typical gravestone doji is a smaller than average candlestick, not a spectacular one.

How the gravestone doji is formed

The chart below is not drawn by hand. It is the median shape of the gravestone doji patterns our bench detected on daily charts across all markets (the caption gives the count), with the faded candles showing the bars before the formation.

The formation of the pattern takes one session. Price opens, buyers take control and push it upward to a new high for the session, then the upward move fails and selling pressure brings the price back down, the first step of a potential reversal. The session ends at or near the opening level, which is also the low. Everything gained during the session is lost before the close.

The market psychology behind the gravestone doji

The gravestone doji pattern tells the story of a failed rally, and it indicates where the market refused higher prices. During the session, buyers were in control long enough to lift the price well above the open. By the close, sellers had taken back every point, and the selling pressure left no gain on the chart. The long upper shadow is the trace of that rejected buying pressure.

After an uptrend, Nison reads this as a shift in market sentiment: the market tried to go higher and was refused. That is why the gravestone doji is classed as a bearish reversal pattern and why it is watched near resistance. After a decline, the same formation is harder to read: it can be a rejected bounce (bearish) or an early sign of buying (bullish, the logic of the inverted hammer).

It is important to keep that in proportion. The reading is plausible, but it describes one session. Whether one rejected rally can indicate a lasting trend reversal in the market is a question for data, not for the story. We come back to it in the reliability section.

How to identify a gravestone doji on a chart

To identify a gravestone doji pattern on a stock, forex or crypto trading chart, check four points in order. Each step helps you understand whether the signal is worth acting on.

Step 1: Confirming the preceding trend

The bearish reversal reading needs an upward trend to reverse: no trend, no reversal. A simple test: the closing price before the pattern is higher than 10 sessions earlier. Without a prior rise, the shape is a gravestone doji, but not a bearish reversal pattern setup. Our bench does not require a trend, which lets us check whether the trend matters: on daily charts, about half of the occurrences we found came after a rise (25,680) and half after a decline (23,953).

Step 2: Checking the shape

Look at the candle once it has formed, at the open and close: they must be nearly equal and both sit at the low of the session, with an upper shadow that is long compared with recent candles. If the body is visible, the pattern is closer to a shooting star. If there is a lower shadow as well, it is a long-legged doji or a plain doji.

Step 3: Analyzing volume and market context

The classical reading gives more weight to a gravestone doji printed on high volume, near a resistance level or after a long rally, because those market conditions make the rejection more significant. We measured volume: on daily charts, the gravestone doji pattern printed with volume above 1.5 times the 20-day average reached the 2:1 target of our test trade 32.0% of the time, against 31.9% for the others. The difference is small, and this breakdown is descriptive, not tested, so treat high volume as a mild plus at best, not a signal that changes the market odds.

Step 4: Checking for confirmation candles

The doji pattern only exists once the session has closed: during the session, price can still end far from the open. It is important to wait for that close. One confirmation is a bearish candlestick on the next session, another a close below the low of the gravestone doji. A break below the low is also the conservative entry for a short position, and it is the clearest way to confirm the signal.

Where does the gravestone doji appear?

The gravestone doji candlestick pattern appears in all financial markets and on every timeframe. On our data, the pattern shows up roughly once every 156 bars on hourly charts, once every 170 on daily charts and once every 189 on weekly charts. On a list of 20 shares, that is roughly one gravestone doji every week or two.

The pattern is less common in the crypto market (once every 220 daily bars) than on stock charts (once every 167), and it is most frequent on daily and weekly forex charts (once every 98 and 89 bars). It is a common pattern, not a rare event, which is one important reason to be careful with each signal.

The gravestone doji in the stock market, forex and crypto

The formation is defined the same way in every market, but market structure changes what it shows.

Stock market

In the stock market, each daily session has a real opening and closing auction, so both prices carry weight. A gravestone doji candlestick pattern on a stock chart means the whole rally of the session was given back by the closing bell. It is the market where we have the most daily data: 43,770 daily occurrences on 1,664 US stocks since 1970. Those stock figures carry a survivorship bias we have not sized: most delisted companies are missing from our data.

Forex market

The forex market trades around the clock during the week, and it can be read on intraday charts. A daily candle there depends on the closing time chosen by the broker. On hourly charts, it appears once every 212 bars, much less often than on daily forex charts (once every 98) but still a steady flow of potential signals, and, in our data, it was followed by worse trading results than random hours (a gap that clears a simple test but not our correction), which is a reason to be careful with it in forex.

Crypto market

The crypto market never closes, so the daily open of our bars is simply the price at midnight UTC. The gravestone doji is less frequent there than on stocks (once every 220 daily bars), but on hourly crypto charts it appears once every 150 bars, and there it did worse than random hours by a margin that survives our correction. On daily crypto charts, the trade hit its target more often than on random days (32.7% against 32.2%), but its average result per trade (0.00 R against −0.02 R) is not statistically significant.

Gravestone doji vs. dragonfly doji: key differences between two doji patterns

The gravestone doji and the dragonfly doji are mirror images, and each is a reversal signal in its own direction. Both are doji patterns with the open and close at the same level, but on opposite sides of the range.

On our data, neither doji stands out on daily charts. The dragonfly doji's two-sided trade is labeled no clear edge (an expectancy of +0.06 R against +0.04 R for random dates). The gravestone doji showed no measurable difference from chance (−0.05 R against −0.04 R).

The shooting star: the same idea with a body

The shooting star has the same long upper shadow, but with a small real body at the bottom of the range, and it requires an uptrend. Both say the same thing about rejected buying and a potential reversal. On daily charts, the shooting star was followed by a decline 10 days later 44.3% of the time, against 46.0% for random days on the same assets, and its average excess return (−0.10%) is within the margin of error: no clear edge either.

The inverted hammer: the same shape after a decline

The inverted hammer has the same shape with a small body, after a downtrend, and is read as a potential bullish reversal. The comparison shows how much the name of a candlestick pattern depends on the trend before it rather than on the shape itself.

Trading the gravestone doji pattern

The classic gravestone doji trading strategy treats the pattern as a potential reversal signal: a short position triggered by a break below its low, with a stop above its high and a profit objective set as a multiple of the risk.

Setting entry points for short positions

There are two entries. The aggressive entry sells at the close of the gravestone doji. The conservative entry places a sell stop just below its low and waits for price to trade through it. Because this doji pattern is classed as indecision in our bench, we test a version that takes no side in advance: it enters on whichever side breaks first, short below the low or long above the high. On daily charts, one side or the other was triggered in practically every case.

Determining optimal stop-loss placement

Calculating profit targets and risk-reward ratios

The objective we test is twice the risk (a 2:1 reward-to-risk ratio). On daily charts, trades that reached it took a median of 3 sessions; trades that were stopped took a median of 2 sessions. A prior support level or a moving average below the price is another reasonable objective. Note that of the trades that hit their stop, 23.8% had first been up 1 R: a rule for moving the stop or taking partial profits matters with this pattern.

Common mistakes when trading the gravestone doji

Trading in isolation without confirmation

On its own, the gravestone doji is not a trading system, and no trading rule should rest on it alone. Our data shows why: on daily charts, its average result per trade (−0.05 R) did not beat the same trade on random dates (−0.04 R) by a margin large enough to separate from chance. Treat it as one input among others (trend, resistance, risk management) when you make trading decisions.

Misinterpreting the signal in a ranging market

Ignoring the importance of volume indicators

In the classical reading, low volume makes any single candlestick pattern less informative, especially in thin, illiquid sessions. Volume is a reasonable filter to test, but our data shows it changes the odds only slightly.

Confusing color with direction

The color does not matter, so do not reject a gravestone doji because it is green. The trend does matter for the bearish reading: the same pattern after a decline is not a bearish reversal pattern in the textbook sense.

Strengthening your trading with technical indicators

Classical analysis does not use the gravestone doji alone: it pairs candlestick signals with technical indicators that describe the trend or the momentum around it, to make more informed trading decisions.

Using relative strength index (RSI) for overbought signals

One filter is to act on gravestone doji signals only when the RSI, a momentum indicator, is overbought (above 70), on the idea that the uptrend is stretched. We have not tested an RSI filter: if you use one, test it on your own market before relying on it.

Applying moving averages for trend validation

Moving averages are the trend indicators we did test. On daily charts, the doji pattern above the 200-day moving average reached the 2:1 target 32.1% of the time (expectancy −0.05 R), against 31.8% below it (−0.06 R). The long-term trend barely moved the result.

Resistance levels and price action

The classical reading gives the gravestone doji the most weight at a level that already mattered, where the price action of the past sets the context: a prior high, a round number, a falling trend line. The formation then shows a rejection where sellers already had a reason to act. We have not measured this confluence, so it remains a matter of price action judgment, one of the technical tools that combine candlestick patterns with chart patterns.

Reliability of the gravestone doji: what our data shows

The gravestone doji pattern is not reliable on its own as a reversal signal: across every market and timeframe we tested, a trade on its break never beat the same trade on random dates in average result per trade by a margin that holds up to our tests, and it did worse on hourly charts, clearly so on crypto. Here are the numbers from our test bench, which applies the same rule to full market history and compares each occurrence with random dates on the same assets.

How to read the table

Read it in pairs of columns. Our test trade is the one described above: an entry on the first break of the high or the low, a stop at the other end of the pattern and a target at twice the risk, without fees. On daily charts it reached its target 32.0% of the time. That number means little alone, so we replay the same entry, stop and target distances on random dates of the same assets: 32.2%. The hit rate is about the same, the average result per trade does not beat random dates by a margin that passes our statistical test, and it is negative in both cases (−0.05 R against −0.04 R per trade, before costs). Verdict: No clear edge.

Because the gravestone doji is a two-sided pattern in our bench, we do not publish a directional figure (how often the price was lower 10 days later) for it: that measure is absent for this pattern. The two-sided trade is the measure we have.

Is the gravestone doji reliable in forex and crypto trading?

No. In forex trading, on hourly charts (10,240 occurrences on 25 currency pairs since 2012), the trade did worse than the same trade on random dates, a gap that clears a simple test but not our correction; on 4-hour forex charts the gap was within the margin of error. In crypto trading, on hourly charts (98,094 occurrences on 587 coins since 2017), the gap was also negative, and it is statistically significant after our correction for multiple testing: worse than chance. On daily crypto and stock market data, the average result per trade showed no measurable difference from chance.

Does the prior trend change the result?

Only slightly. On daily charts, gravestone doji patterns after a rise (the textbook bearish setup) reached the 2:1 target 31.4% of the time, with an expectancy of −0.07 R; after a decline, 32.6% and −0.04 R. These breakdowns are descriptive and not corrected for the number of tests.

The higher raw numbers on indices and commodities come with higher numbers on random dates too. One possible reason, which we have not tested, is that those markets trended in a way that helped any trade, not the pattern. On large US stocks, the result agrees with academic work: Marshall, Young and Rose, in "Candlestick technical trading strategies: Can they create value for investors?" (Journal of Banking & Finance, 2006), found that candlestick strategies did not add value on large US stocks. The All measurements table, with every market, timeframe and verdict, is in the statistics section of this page, and the method is in our methodology.

Limitations and risks of the gravestone doji

• On its own, the pattern did not clearly beat chance on any market or timeframe we tested, and it did worse than chance on hourly charts, clearly so in crypto trading.
• It is frequent (one bar in 89 to 220 depending on market and timeframe), so it produces many signals.
• The stop is wide relative to the body, and the median losing trade is stopped within 2 bars on daily charts.
• Its meaning depends on the trend before it, and there is no single rule for that trend, so two readers can judge the same candle differently.

What the pattern does offer is practical: it is easy to identify, it gives a clear invalidation level (the high), and it is available on every market and timeframe. None of that is an edge in our data.

Real-world examples of the gravestone doji

The gallery of this section shows real gravestone doji patterns, picked by our bench across markets (best and worst results, textbook shapes, most recent cases and typical draws), with their actual outcome. The cases written up below are on daily crypto charts.

Example 1: a successful bearish break

On December 17, 2020, FIL/USDT printed a gravestone doji: open 28.41, high 29.46, low 28.28, close 28.44, the day after a drop to 27.82. Two days later the price traded below the low, triggering a short entry at 28.28 with a stop at 29.46 and a 2:1 target at 25.93. Price reached the target on December 21, and FIL was 21.8% lower 10 days after the pattern. It is a clean example of the shape, not of the full setup: price had not risen before it, so step 1 fails.

Example 2: in profit, then stopped

On April 22, 2026, ETC/USDT printed a gravestone doji (high 8.75, low 8.41) after several quiet sessions. The short entry at 8.41 was triggered the next day, and on April 29 the price fell to 8.13, about 0.8 R in favor of the trade. It never reached the target at 7.73: price rallied and hit the stop at 8.75 on May 4, a full 1 R loss. It fell just short of the 1 R gain that 23.8% of stopped trades had reached before failing.

Example 3: both sides in one session

On December 19, 2021, XLM/USDT printed a gravestone doji with a high at 0.2672 and a low at 0.2513. The next session traded above the high (triggering the long side of our two-sided trade) and below the low on the same day. When one bar touches both the entry and the stop, our bench counts the loss, so this trade is a 1 R loss. XLM was 5.4% higher 10 days later.

Trading tools for spotting the gravestone doji

Charting platforms and technical analysis tools with a candlestick scanner can look for the gravestone doji. The open-source TA-Lib library provides a CDLGRAVESTONEDOJI function. Check which definition your tool uses: the thresholds for "no body" and "no lower shadow" vary, and so do the results.

Before trading the gravestone doji with real money, mark the last 50 occurrences on your market in a demo account, write down the entry, stop and target you would have used, and record the outcome. Compare it with the same trade on random days. That exercise is the test we run at scale, and it shows quickly whether the pattern improves your trading decisions.

Frequently asked questions

Summary

• A gravestone doji has its opening, closing and low prices at about the same level and a long upper shadow.
• After an uptrend it is read as a potential bearish reversal; the pattern is the mirror of the dragonfly doji.
• On 49,732 closed trades on daily occurrences, a 2:1 trade on its break reached the target 32.0% of the time, against 32.2% for the same trade on random dates: about the same hit rate, and a gap in average result per trade too small to separate from chance.
• On hourly charts, especially in crypto trading, it did worse than chance.
• Volume, the prior trend and the 200-day moving average changed the results only slightly.

If you trade the gravestone doji pattern, treat it as a location for a trade, not a reason for one, and write it into a trading plan. Define the stop above the high before entry, size the position on that distance, and keep the risk per trade small. Any edge has to come from what you combine it with.

Risk notice. This page is educational and is not investment advice. Positions in stocks, forex and crypto carry a risk of loss, all figures are before fees and slippage, and past results do not predict future ones.

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 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 (the Benjamini-Hochberg false discovery rate procedure), 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.