Tasuki Gap Candlestick Pattern: Rules, Examples and Tested Results

The tasuki gap candlestick pattern is a three-candle continuation pattern. Two candlestick bars in the direction of the trend leave a price gap between them, then a third one of the opposite color moves back into the gap but fails to close the gap. Technical analysis textbooks read it as a pause in the trend: the trend is likely to continue because the counter-move could not fill the gap.

This guide covers the definition, the two versions, how to identify the pattern on a candle chart and use it in a trading plan, and what actually happened after it. Our test bench applied the same rule to full price history, found 560 upside and 364 downside occurrences on daily charts alone, and compared the outcome with random bars on the same assets, so you can see whether the signals provide anything a random entry does not.

What is the tasuki gap pattern?

The tasuki gap is a continuation pattern: it does not announce a reversal, it suggests that the current trend will resume after a short pullback. It comes in two versions. The upside version forms in an uptrend and is bullish. The downside version forms in a downtrend and is bearish.

The name is Japanese. A tasuki is a cord used to hold back the sleeves of a kimono, an image of something that holds the move in place. Steve Nison, who brought Japanese candlestick analysis to Western traders in Japanese Candlestick Charting Techniques (1991), describes these candlestick patterns in his chapter on windows, the Japanese term for a price gap. In that tradition, an open window acts as support in an uptrend and as resistance in a downtrend, and this pattern is a test of that window that holds.

The structure of the tasuki gap

The pattern consists of three candles, and each one has a precise role. Understanding these roles is the start of any analysis of the setup. Our bench uses the same conditions, written so a computer can apply them without judgment.

First candle: confirmation of the existing trend

The first candle moves with the trend: a green (white) body in an uptrend, a red (black) body in a downtrend. It shows that buyers or sellers are in control. Our rule requires a trend before the pattern: for the bullish case, the last close before the pattern is at least one average true range above the close 10 bars earlier, and in the upper half of the last 40 closes.

Second candle: creating the gap

The second candle consists of a body of the same color as the first and opens with a gap. In an uptrend, its low is above the high of the first bar; in the downside version, its high is below the low of the first bar. This is a true price gap: an empty zone on the chart where no trade took place.

Third candle: the partial gap fill

The third candle has the opposite color: black (red on color charts) in an uptrend, white (green) in a downtrend. It starts inside the real body of the middle bar, moves back toward the gap and ends inside it, without reaching the first bar. In an uptrend, it closes below the opening price of the middle bar but above the high of the first one. Its real body is about the same size as the middle body. Because the final price stays inside the zone, the window remains open: the counter-move did not close the gap.

Types of tasuki gap patterns

Upside tasuki gap

The upside tasuki gap pattern is the bullish one. In an uptrend, a rising bar is followed by another rising bar that gaps up, then by a black candle that starts inside the middle body and ends inside the upward gap. The chart above is not drawn by hand: it is the median shape of the clean bullish occurrences our bench found, with the levels of the trading setup we test.

Downside tasuki gap

The downside formation follows a downward trend. A black bar is followed by another black bar that gaps down, then by a rising candle that starts inside the middle body and ends inside the gap, below the low of the first one. The chart below is the median downside shape from our data.

The psychology behind the tasuki gap

The story is one of strength, a pause and a failed counter-attack. In a rising market, buyers drive the first candle, then demand is strong enough to open the next one above the previous high. On the third day, in the classic reading, profit taking or bets on a correction pull price back into the gap. But it holds inside the zone: sellers could not close it below the high of the first bar, even when the session low trades deeper. Traders read this as a sign that the market still respects the gap as support and that the move is likely to continue.

The downside story is the mirror: sellers gap the price down, a relief rally pushes it back up, and the rally stalls inside the gap, under the old support that has turned into resistance. This reading is plausible, but it is a story. Whether the failed counter-move predicts anything is a question for data, which we answer in the reliability section below.

How to identify the tasuki gap on a price chart

To identify the tasuki gap, check these conditions in order:

1. A clear trend: up for the bullish version, down for the bearish one.
2. Two bars in the trend direction, with a real gap between the shadows of the first and the second candle. This is how the setup starts.
3. A third bar of the other color that starts inside the middle real body.
4. That third bar ends beyond the middle real body but does not close the window: the space stays partly open.
5. The middle and last bodies are roughly the same size.

Wait for the candle to close. During the session it can look like a tasuki gap and then fill the window entirely, which turns it into a different pattern. The pattern is rare, at least with our strict definition: we found one upside occurrence every 13,907 eligible bars and one downside occurrence every 21,395. On a single stock, with about 252 sessions a year, that means more than a decade between two occurrences of the same version, on average.

Why the pattern is so rare in crypto and forex

The pattern needs a gap, and gaps need a market that pauses. Crypto trading runs around the clock, so each bar opens almost exactly at the previous price. Forex trading runs without a break from Monday to Friday. Our data shows the effect: across all the crypto coins we test, we found no upside occurrence on any timeframe and 1 downside occurrence on hourly charts, one every 14,681,353 hourly bars, and in forex none at all, on any of the timeframes we test. Gaps rarely occur in these markets. In practice, it is a pattern of stocks, ETFs and indices, the financial markets with a session that ends overnight. A gap can form there every morning.

How to trade the tasuki gap

The classic trading strategy is to enter in the direction of the trend once price confirms that the pullback is over, with a stop loss on the other side of the formation and a target set as a multiple of the risk. This is the trading setup our bench simulates on every occurrence.

Setting entry points

Placing stop-loss orders

A tighter stop goes just beyond the window: if the counter-move continues and closes the gap, the reading is wrong. Our bench uses the other extreme of the whole formation, which is wider. That width matters: the median upside pattern spans 2.13 times the average true range (ATR) of the asset, so the exit level is far from the entry. Size the position on that distance, not on the size of one bar, to limit the potential loss.

Determining profit targets

The target we test is twice the risk (a 2:1 reward-to-risk ratio); other choices are the next resistance level for a long position and the next support level for a short. In our data, winning upside trades took a median of 26 bars to reach the 2:1 target; losing ones were stopped out after a median of 11 bars.

Advanced trading strategies for the tasuki gap

Because the pattern alone gave weak results in our test, it can be combined with technical indicators, chart analysis and price levels. Consider the strategies below as a start for your own analysis. Each filter below should be tested on your own market before you risk capital on it.

Using moving averages for trend confirmation

A continuation pattern makes more sense inside a strong trend, and one simple filter among technical indicators is the 200-day moving average. We measured it. Upside occurrences above the 200-day average reached the 2:1 target in 39.4% of cases, against 25.0% below it, with expectancies of +0.21 R and −0.19 R. Downside setups below the average reached it in 14.6% of cases (expectancy −0.48 R), and 15.1% above it (−0.54 R). These breakdowns are descriptive and rest on uneven groups: 471 upside setups above the average against 83 below it, and 114 downside setups above it against 247 below. The filter did not turn the downside version into a winner.

Combining the pattern with RSI divergences

One filter skips a bullish setup when the relative strength index (RSI) shows a bearish divergence, on the idea that momentum is fading even though price gapped. We have not tested an RSI filter, so we have no figure for it.

Integrating Fibonacci levels with the tasuki gap

Another approach looks for the third bar to stall near a Fibonacci retracement of the middle bar, or near another technical level, as extra evidence that the gap is acting as support or resistance. This is a matter of confluence, and we have not measured it.

Important distinctions: tasuki gap vs. gap three methods

The upside gap three methods starts the same way: two rising bars with a window between them. The difference is the last bar. In the tasuki gap, it ends inside the window and leaves it open. In the gap three methods, it ends inside the first body and fills the space completely. Both are traditionally read as bullish continuation. On our data, that pattern (56 occurrences, 19 closed trades) reached its 2:1 target in 21.1% of cases against 35.8% for the same setup on random dates, too few trades to read a result. The bearish twin is the downside gap three methods.

Other continuation patterns include the side-by-side white lines, where two bars of the same color sit next to each other after the window, and the rising and falling three methods, which hold inside the range of a long candle without a gap.

Advantages of using the tasuki gap in your strategy

• A precise definition: three candles, a window, a partial fill. It leaves little room for judgment.
• It trades with the trend, not against it, so it fits a trend-following trading strategy.
• Clear invalidation: if the space is filled, the setup is gone.

Common mistakes to avoid when trading the tasuki gap

• Counting a third bar that fills the window. That is a different pattern, with a different reading.
• Trying to enter before the third candle closes.
• Ignoring the trend: this pattern in a sideways range has no trend to continue.
• Taking the downside version as a strong short trade, with a tighter exit than the formation allows. In our data it did worse than chance, a gap that passes a simple test but does not survive our correction for multiple testing.
• Expecting it on crypto or forex charts, where it almost never forms.

Limitations and risk management

The main limitation is that the pattern did not clearly beat chance on any market we tested. It is also rare, which makes it hard to build a trading system around it, and the stop is wide. Risk management decides the outcome more than the trading signals do, and it protects your money: define the exit before entry, risk a small fixed share of the account on each position, and keep a record of every one so you can compare your results with chance. Past performance, ours included, does not guarantee future results or future trends, and nothing here is financial advice.

Performance of the tasuki gap in different markets

The tasuki gap candlestick pattern does not provide reliable signals on its own. Across every market and timeframe with enough data, neither version beat random bars on the same assets by a margin our tests separate from chance. Here are the numbers from our test bench, which applied the same rule to full price history.

Read the table in pairs of columns. "Moved as expected" means higher 10 bars later for the upside version and lower for the downside version. On daily charts, 52.1% of upside occurrences were followed by a rise, but 55.0% of random bars on the same assets rose too: the pattern did slightly worse than an average session. The excess return over the rest of the market after 10 bars was −0.14%, with a 95% confidence interval from −0.59% to +0.32%. Zero is inside the interval: no clear edge.

The 2:1 trading setup tells the same story. The bullish version reached its target in 37.8% of cases against 37.9% on random dates, for an expectancy of +0.17 R per position before costs, against +0.21 R for the same setup at random. The positive number comes from rising stock markets, not from the pattern. The downside version lost: 14.5% of targets hit, an expectancy of −0.50 R against −0.30 R at random. Shorting after this pattern, all markets combined, did worse than shorting at random dates: the 2:1 trade reads weak, below chance, a gap in average result per trade that passes a simple test but not our correction for multiple testing. On the 10-bar move, the downside version also went the wrong way, but its excess return of −0.28% reads no clear edge: zero is inside its margin of error.

US stocks make up about nine in ten occurrences and show the same picture: 35.5% of upside targets hit against 36.4% at random, and 14.8% of downside targets against 22.6%. The bullish row on the longer timeframe looks better (50.0% of targets hit against 45.8%), but it rests on only 8 positions, too few to interpret. Index data are thin: 24 upside and 3 downside occurrences. Both directions and their trades (16 upside, none downside) are too few to rate: not enough data.

Does volume or the long-term trend change the result?

Not in a useful way. Upside setups with high volume on the last bar (more than 1.5 times the average) reached the 2:1 target in 31.8% of cases, against 38.3% with normal volume. Downside setups with high volume reached it in 13.5% of cases, against 14.9% with normal volume. Across the decades, the downside excess return was negative in both the first and the second half of the history (−0.18% and −0.39%), while the upside one changed sign (+0.06% then −0.33%): no stable edge for either version.

The full table, with every market, timeframe and verdict, is in the statistics section of this page. To understand how we form the random benchmark and learn how each occurrence is counted, read the method in our methodology.

Real examples of the tasuki gap candlestick pattern

Here are real tasuki gaps, picked by our bench (best and worst results, textbook shapes, most recent, random draws), with their actual outcome. The case studies below are on stock charts.

Upside tasuki gap that reached its target

On August 31, 2017, Micron (MU) shares rose with a white candle. The next session opened at 31.33, above the previous high of 31.22, and rose again. On September 5, a black candle opened inside the middle body and ended at 31.27, below the middle candle's open of 31.33 and above the first candle's high of 31.22, so the gap stayed open at the close, even though the session low of 30.86 had traded below that high. The buy order at the top of the pattern (32.06) was triggered on September 7, with a protective level at 30.69 and a 2:1 target at 34.79. Price reached the target on September 18, 2017.

Downside tasuki gap that hit its stop-loss

On July 30, 2010, Micron (MU) gapped below the low of the previous black candle, and the following session a rising candle ended inside the gap. The sell order at 7.08 was triggered on August 5, with a protective level at 8.10. Price fell as low as 6.20 on August 31, short of the 5.03 target, and was 6.1% lower 10 bars after the pattern; it then climbed back and gapped above the stop 64 sessions after entry: a loss of 1.15 R.

A setup canceled before entry

On January 21, 2026, Meta Platforms (META) completed a downside tasuki gap, closing at 611.39, inside the window between 609.83 and 618.49. The next session, on January 22, price traded above the top of the pattern (627.47) without having reached the entry at its low (598.46): the setup was canceled without a position, and the stock was 9.1% higher 10 bars later. Cancellations like this one happened on almost half of the downside setups.

Trading tools for finding the tasuki gap

Technical analysis software such as TradingView can flag candlestick patterns on a chart, and the open-source TA-Lib library provides a CDLTASUKIGAP function. The same rules apply on any financial chart, but definitions differ between tools, for example on whether the gap is measured between shadows or between bodies, so check which one your tool uses. Ours is stricter than most: on the bars we compared, it finds about 18% of the upside setups and 17% of the downside setups that the TA-Lib function flags, so a scanner will usually flag more cases than our counts. Before trading the pattern with real money, learn its rules and review its past occurrences on your market in a demo account, and compare the outcome with the same setup on random dates.

Frequently asked questions

Summary of key points

• The tasuki gap is a three-candle continuation pattern: two candles with a gap, then a candle of the other color that ends inside the window without filling it.
• On 560 upside occurrences on daily charts (325 closed trades), the 2:1 target was hit in 37.8% of cases, against 37.9% for the same setup on random dates.
• On 364 downside occurrences (186 closed trades), the expectancy was −0.50 R, worse than the −0.30 R of the same setup at random.
• The pattern needs gaps, so it is almost absent from crypto and forex charts.
• Treat it as context in a trend, with a defined exit, not as a trading signal on its own.

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. This page is educational content, not financial advice.