Downside Gap Three Methods Candlestick Pattern: Rules, Trading and Real Statistics

The downside gap three methods candlestick pattern is a three-candle bearish continuation pattern: in a downtrend, two long bearish candles separated by a gap down, then a bullish candle that opens inside the second body and closes inside the first one, closing the gap. Candlestick analysis reads the third candle as short covering inside a falling market, not as a reversal.

This guide covers the definition, how to identify the pattern, how it differs from the downside tasuki gap, how traders use it, and how to code it. It also shows what happened after it. Our test bench detected 67 occurrences on daily charts across stocks, ETFs and indices since 1980, and compared each one with random days on the same assets.

What is the downside gap three methods pattern?

The downside gap three methods is a bearish continuation candlestick pattern. It appears during a downtrend and says that a rebound which fills the gap is a pause, not the end of the decline. It is the mirror image of the upside gap three methods, and a close relative of the downside tasuki gap, which has the same first two candles but a shallower third candle.

The chart below is not drawn by hand. It shows the median shape of the daily occurrences our bench found across all markets: the faded bars show the trend before the pattern, and the lines show the position we test, entry below the pattern low, stop above its high, target at twice the risk.

How to identify the downside gap three methods on a chart

Check each part in order. Our bench applies these rules without judgment.

Defining the bearish trend

The pattern is a continuation signal, so there must be a downtrend to continue. Our test: the last close before the pattern is at least one average true range (ATR) below the close 10 bars earlier, and in the lower half of the 40 previous closes. The same candles inside a range carry no continuation meaning.

The first candle: long bearish candle

A long bearish candle, with a close well below the open, extends the decline. Its size is measured against the 10 prior bars on the same asset.

The gap: separation between candles

The next session opens below the low of the first candle and its high stays below that low: a true gap, with no overlap between the shadows. A gap between the bodies only does not count in our definition.

The second candle: long bearish candle

The second candle is also long and bearish, and closes near a new low of the move. Some summaries describe it as small or bullish: that is a different pattern. In the standard definition, both candles before the gap fill are long and bearish.

The third candle: closing the gap

The third candle is bullish. It opens inside the body of the second candle and closes inside the body of the first, which fills the gap. It must not close above the open of the first candle: the rebound stays inside the decline.

The psychology behind the pattern

The first candle shows strong selling pressure. The gap down shows that sellers were still in control before the next open, and the second candle confirms it with a new low. On the third trading session, short sellers take profits and buyers step in, and price climbs back through the gap. Read in the bearish way, the gap was filled without a close above the first candle's body: the selling that started the move is still in control, and the rebound is temporary.

The opposite reading is just as easy to defend: a gap that fills in one session can be the first sign that sellers are running out. Which reading wins is a question for data, which we answer in the reliability section.

Downside gap three methods vs downside tasuki gap

The two patterns share the same first two candles: two bearish candles with a gap down between them. The difference is the third candle.

• Downside tasuki gap: the bullish third candle closes inside the gap, which stays partly open.
• Downside gap three methods: the bullish third candle closes inside the first body, so the gap is completely filled.

If an open gap acts as resistance, as gap theory holds, the downside tasuki gap should be the stronger bearish signal. Our data does not support either one. On daily charts, the downside tasuki gap (364 occurrences) was lower 10 days later in 41.4% of cases, against 45.3% for random days, and its 2:1 short trade (186 trades) reached the target 14.5% of the time against 22.1% for the same trade on random dates. Its average excess return after 10 days was even negative (−0.28%), within the margin of error, a result rated no clear edge. The downside tasuki gap is also far more common: one occurrence every 21,395 eligible daily bars, against one every 115,091 for the downside gap three methods.

Other related patterns

The falling three methods is a five-candle bearish continuation pattern with no gap: a long bearish candle, three small candles inside its range, then a new bearish candle. On daily data, its 2:1 short trade reached the target 29.2% of the time against 30.0% on random dates. The upside gap three methods is the exact mirror in an uptrend: on 56 daily occurrences, it did not beat chance either.

The Real examples section shows downside gap three methods patterns from our bench with their actual outcome. Trade results are in R, the initial risk of the trade.

How to trade the downside gap three methods

The classic trading strategy sells short on a break below the pattern low, with a stop above the pattern high and a target set as a multiple of the risk. This is the position our bench simulates on every occurrence.

Entry strategy and signal confirmation

The aggressive entry sells at the close of the third candle. The conservative entry places a sell stop just below the low of the pattern, valid for 20 bars, and waits for price to trade through it. In our tests on daily data the sell stop was triggered in only 32.8% of cases. In 65.7% of cases, price rose above the highest high of the pattern before breaking the low, and the setup was canceled before any trade: in those cases, the rebound that filled the gap kept going. The few remaining setups touched neither level within the 20 bars the order stays valid.

Setting stop-loss orders

Determining profit targets

The target we test is twice the risk. On daily data, only 7 trades reached that target, too few to read a median; trades that were stopped took a median of 18 bars. Of the trades that hit their stop, 30.8% had first been up 1 R, so a partial exit at 1 R or a trailing stop changes the result. A prior support level below the entry is another reasonable target.

Essential indicators for pattern confirmation

The pattern can be combined with a filter. Here are three, and our data covers the first two, as descriptive splits that are not corrected for multiple testing.

Using volume to validate the trend

Classical candlestick analysis reads heavier volume on the bearish candles than on the rebound as a sign that sellers are still in control. On daily data, occurrences with high volume on the pattern (more than 1.5 times the average, 32 cases) were lower 10 days later in 48.4% of cases, against 56.3% for the others. The excess return over the rest of the market, −2.18% (positive favors the pattern), has a 95% confidence interval from −4.45% to +0.50%: it includes zero, so this is not a proven filter.

Moving averages for trend context

A trader who wants to sell with the main trend takes the pattern only when price is below its 200-day moving average. Our data cannot judge that logic: below the 200-day average (49 occurrences), price was lower 10 days later in 50.0% of cases; above it, only 18 occurrences (61.1%), fewer than 30, too few to compare. On the 2:1 short trade, the 15 closed trades below the average (fewer than 30) hit the target 33.3% of the time (expectancy +0.20 R) and the 7 above it are both too few to read, so this split says nothing about the trend filter.

Relative strength index for overbought conditions

An RSI filter sells the pattern only when the RSI rebounds toward its middle zone during the gap fill, on the idea that the bounce has used up its momentum. We have not tested an RSI filter: if you use one, test it on your own market first.

Common mistakes to avoid when trading this pattern

• Calling a downside tasuki gap a downside gap three methods: if the gap is not filled, it is a different pattern.
• Ignoring the trend: the same candles after a rise carry no continuation meaning.
• Selling short at the close of the third candle without a stop.
• Treating the pattern as a complete strategy: on its own, it did not beat chance.
• Expecting the pattern often: on daily stock charts it appears about once every 118,054 bars.

Advantages and limitations

Advantages of the pattern

• Clear rules: three candles, one gap, one fill.
• Clear risk: the stop goes above the high of the pattern.
• Easy to code and to scan for, with TA-Lib or a few lines of your own, but TA-Lib flags far more cases: our rule finds 4% of them.

Limitations and risks of the pattern

• Very rare on stocks, almost absent on crypto and absent on forex.
• On its own, it did not beat chance on any market or timeframe with enough data.
• Two setups in three are canceled before entry, and the stop is wide.

Reliability and success rate

The downside gap three methods is not reliable on its own: on daily charts, its average excess return moved against the expected direction relative to its market (−1.98%, within the margin of error), even though it was followed by a decline more often than a random day on the same assets, and its 2:1 short trade, though ahead of the same trade on random dates on paper, rests on only 22 closed trades, fewer than 30: too few to interpret. Here are the numbers from our test bench, which scans full price history with a definition much stricter than the TA-Lib function: on the bars compared, we find 4% of the occurrences it flags.

Read the table in pairs of columns. On daily data, 53.0% of occurrences were followed by a decline 10 days later, against 45.0% of random days on the same assets: on these assets, mostly US stocks, random days fell less often than they rose, pattern or not, so a bearish pattern starts with a handicap. The excess return over the rest of the market after 10 days is −1.98% (positive favors the pattern, so this went against it), with a 95% confidence interval from −4.31% to +0.82%. Zero is inside the interval: no clear edge. This is in line with academic work on candlesticks: Marshall, Young and Rose (Journal of Banking & Finance, 2006) found that candlestick trading strategies did not create value on US stocks.

The 2:1 short trade reached its target 31.8% of the time, against 4.8% for the same entry, stop and target distances on random dates, for an expectancy of +0.09 R per trade against −0.51 R, before costs, but on only 22 closed trades, fewer than 30: not interpreted. On weekly data, 6 occurrences read Not enough data, and only 3 closed trades, fewer than 30: not interpreted.

Can this pattern be used in forex and crypto markets?

In practice, almost never. The pattern needs a true gap between two bars. Crypto trades 24 hours a day and forex around the clock five days a week, so a bar almost always opens where the previous one closed. Our bench found no occurrence at all on forex (data since 2012, hourly to weekly). On crypto it found 1 on hourly charts (one every 14,681,353 eligible bars) and none on 4-hour or daily charts. Stocks gap overnight, which is why 62 of the 67 daily occurrences are on US stocks.

Other markets and stability

Indices (4 daily occurrences) and ETFs (1) have too few cases to measure anything, and the bench finds none on commodities. The excess return after 10 days was −1.79% in the first half of the history and −2.21% in the second, with wide error margins: not a proven edge. The full table, with every market and verdict, is in the statistics section of this page, and the method is published in our methodology.

Real-world application and case study

A US stock case from our bench, with prices adjusted for splits and dividends, shows both the shape of the pattern and why a 10-day reading and a trade result can disagree.

Caterpillar, December 1980

On December 10, 1980, Caterpillar (CAT) printed a bearish candle from 2.615 to 2.571, with a low at 2.565. On December 11 it opened at 2.554, which was also its high, under that low, and closed at 2.510. On December 12 it opened at 2.521 and closed at 2.593, inside the first body: the gap was filled.

The sell stop below the pattern low (2.493) was triggered on December 22, with a stop above the high at 2.620 and a 2:1 target at 2.240. Price fell as low as 2.471 on December 24, then turned. On January 5, 1981, the stock reached the stop, a loss of 1 R. Ten days after the pattern, price was 2.8% below the close of the third candle, so a reading based only on the direction after 10 days would have counted it as a success. The trade tells a different story.

The upside version has three stock examples, two losers and a canceled setup, on the upside gap three methods page.

Code implementation and TA-Lib

TA-Lib, the open-source technical analysis library, detects both versions with the CDLXSIDEGAP3METHODS function. It takes the open, high, low and close series and returns a negative value on a downside gap three methods, a positive value on an upside one, and zero otherwise. The logic our bench uses, for the three candles c1, c2 and c3:

is_pattern = (
    c1.close < c1.open and c1.is_long and     # long bearish candle
    c2.close < c2.open and c2.is_long and     # second long bearish candle
    c2.high < c1.low and                      # true gap down
    c3.close > c3.open and                    # bullish third candle
    c2.close < c3.open < c2.open and          # opens inside the second body
    c1.close < c3.close < c1.open             # closes inside the first body
) and downtrend_before(c1)

Thresholds vary between tools, so two scanners can disagree on the same chart.

Frequently asked questions

Summary of key points

• The downside gap three methods is a three-candle bearish continuation pattern: two long bearish candles with a gap down, then a bullish candle that fills the gap.
• If the gap is not filled, it is a downside tasuki gap.
• On 67 daily occurrences, 53.0% were lower 10 days later, against 45.0% for random days on the same assets, a difference not distinguishable from chance.
• The 2:1 short setup reached its target 31.8% of the time on 22 trades, against 4.8% for the same trade on random dates: too few trades, fewer than 30, to interpret.
• Two setups in three were canceled before entry, and the pattern almost never appears on crypto and never on forex.

If you trade this pattern, treat it as a place to look for a short trade inside a downtrend, not as a reason for one. Put the stop above the pattern high before entry, size the position on that distance, and keep the risk per trade small. Short selling adds its own risks, such as borrowing costs and sharp rebounds. Our data shows no edge from the pattern itself: any edge has to come from what you combine it with. Results measured on historical data, ours included, do not guarantee future performance, and this page is not investment advice.

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.