Advance Block Candlestick Pattern: Meaning, Trading and Tested Results

The advance block candlestick pattern is a three-candle bearish reversal pattern that forms during an uptrend: three consecutive green candles close higher, but each body is smaller than the one before, or the last two show tall upper wicks. Price is still rising, yet the buying is losing strength. In the classic reading, it warns that the move up is running out of fuel.

This guide explains how to identify the pattern, what it shows about buyers and sellers, how a trader can use it, and what actually happened after it. Our test bench applied one rule to full price history and found 2,212 on the daily timeframe alone across US stocks, crypto, forex, ETFs, indices and commodities, then compared the outcome with random bars on the same assets.

What is the advance block candlestick pattern?

Steve Nison, who introduced Japanese candlestick charting to Western traders in Japanese Candlestick Charting Techniques (1991), presents it as a variation of the three white soldiers: the same three rising sessions, but with signs of weakness. The name says it all: the rally runs into a block of sellers. Where strong buying sessions show buyers in control, this pattern shows buyers who still push the close higher but meet more and more selling near the highs.

It is classified as a bearish reversal pattern, but it is a warning more than a sell signal. Nison reads it as a reason for protecting long positions rather than for opening new short positions. This guide waits for a confirmation, described below, before acting on it.

Rules of the pattern

Our bench uses the following rules, built on the idea of the open-source TA-Lib library (function CDLADVANCEBLOCK) but clearly different from it: on candle shape alone, at TA-Lib's own long-body threshold, they find only 16% of the cases that function flags.

• The market is rising beforehand: the close before the first candle is at least one average true range above the close 10 bars earlier, and in the upper half of the last 40 closes.
• Three bullish candles close higher, one after the other.
• The second and third candles open inside the body of the candle before them.
• The first candle has a large body.
• The momentum fades: each body is smaller than the one before, or the second and third have tall upper wicks.

The chart above is not drawn by hand. It is the median shape of a random sample of the occurrences our bench detected on daily charts across all markets, with the faded candles showing the trend before the pattern and the lines of the setup we test: entry below the low of the pattern, stop above its top, target at twice the risk.

How to identify the advance block pattern

Read the three consecutive candles one at a time. Each candlestick tells part of the story, and you need to know what each one shows before you name it.

The first candle: strong bullish momentum

The first candlestick is a large white candle that closes near its peak, in the direction of the existing trend. Nothing looks wrong yet: buyers are in control and the market makes a new peak.

The second candle: signs of slowing momentum

The second candle opens inside the body of the first and closes higher, but its body is shorter, or it leaves a tall upper wick. Buyers reached a higher level during the session, then sellers pushed the close back down. The second candle is where the loss of momentum first shows on the chart.

The third candle: exhaustion and reversal potential

The third candle again opens inside the previous body and closes a little higher. Its body is the smallest of the three, or its upper wick is the tallest. Three higher closes, but each gain is smaller and harder to keep: the rally is stalling.

Reading bodies and shadows

Two parts of each candle carry the information. The body, from the open to the close, shows how much ground buyers kept by the close: a shrinking body shows a weaker push from one candle to the next. The upper shadow, from the top of the body to the session peak, shows how far buyers got before supply pushed price back: a longer shadow on each candle shows growing resistance. A candle can show one sign or both. Color adds nothing here: all three sessions are bullish, with no red candlesticks inside the formation. In the textbook reading, the warning is clearest when both appear together on the second and third candle: a smaller body and a longer shadow.

Size of the pattern

On the daily timeframe, the median occurrence spans 4.68% of price from its lowest to its highest point, or 2.23 times the average true range (ATR) of the asset. It is a wide pattern, which matters for the stop.

Where it fits among candlestick patterns

Classical candlestick analysis sorts patterns into two groups: reversal formations, where the shape suggests a change of direction, and continuation formations, where the trend is expected to resume. On paper, this one belongs to the first type. In practice, it is important to know that a candlestick pattern only describes what buyers and sellers did over a few sessions. It provides a context, not a forecast, and our data shows that the trend went on more often than it turned.

A quick way to see the difference: a bearish reversal needs sellers to take over, while a continuation only needs buyers to pause. The second candle and the third candle of this candlestick pattern show a pause, not a takeover.

The psychology behind the pattern

Each candle shows the same fight. Buyers open the session inside the previous body and push price higher, but sellers appear near the highs, take profits, and leave upper wicks behind. The bodies still step up, so the trend looks intact on a line chart. On a candlestick chart, the shrinking bodies and the growing wicks show that each new peak costs buyers more effort.

The pattern shows buyers losing pressure as the advance slows: the upward push is still there, but it costs more and more. That reading is plausible. Whether a fading three-session rally turns into a decline is a question for data, not for the story. We come back to it in the reliability section.

How to trade the advance block pattern

There are two ways to use it: as an exit warning for a buy position, or as a candlestick setup to sell. Both start from the same three sessions, but they are different decisions. Opening a short position on the third close needs a significant drop to pay off, and the potential loss is large because the stop sits above a wide formation. Tightening the stop on a long position only needs the upward pressure to fade.

Entry strategy and signal confirmation

Two confirmations are possible: a red candle that closes below the body of the third candle, or a break below the low of the pattern. Our bench uses the second rule: a sell stop just below the lowest point of the three candles. The order only triggers if price falls through the whole pattern.

This confirmation filters out most setups. On the daily timeframe, the sell stop was triggered in 13.7% of cases; in 85.8% of cases, price broke above the top of the pattern first, and the setup was canceled before any entry. The few left expired: price touched neither level within the 20 bars the order stays active, or the data ended first. In other words, after a typical occurrence, the trend went on.

Setting stop loss and take profit levels

The stop goes above the highest wick of the pattern. If price trades there, buyers have pushed through the blocking area and the idea is wrong. Because it spans about two ATR in the median case, size the position on the distance between entry and stop.

The target we test is twice the risk (a 2:1 reward-to-risk ratio); another choice is the next support level below. On the daily timeframe, trades that reached the 2:1 target took a median of 17 days; trades that were stopped took a median of 9 days.

Using it to manage an open position

Differences between advance block and three white soldiers

The three white soldiers and this pattern both have three white candles with higher closes, each opening inside the previous body. The differences:

On daily charts, the three white soldiers setup reached its 2:1 target in 34.4% of cases against 33.3% for the same setup on random dates. Neither pattern showed a clear edge in our tests.

Related patterns

The stalled pattern, also called deliberation, is the close cousin: two large white candles, then a small third candle at the top. The shooting star shows the same rejected buying in a single candle with a tall upper wick, the hanging man is another one-candle warning after a rally, and the evening star is a three-candle top with a clearer turn. On daily charts, none of them made money on the 2:1 trade in our data: the shooting star setup reached its target in 28.8% of cases against 29.7% on random dates, the bearish stalled pattern 27.7% against 26.0%, and the evening star 26.6% against 25.8%, and none of these gaps passes our statistical test. The hanging man is no exception: 25.0% against 24.1%, and its average result per trade, −0.25 R against −0.28 R on random dates, does not survive our correction for multiple testing either: it lost money, like random dates.

Critical limitations and risks of the pattern

• It is a warning, not a reversal: on daily charts, 45.5% of occurrences were lower 10 days later, against 45.4% for random days, and the average excess return (+0.07%) is not statistically significant.
• The definition is loose: "smaller bodies" and "tall upper shadows" can be measured in many ways, so two scanners will not find the same occurrences.
• The stop is wide, about two ATR in the median case.
• Selling into an uptrend fights the main direction of the market, which has a cost on assets that rise over time, such as stocks and indices.
• No market and timeframe showed a proven edge. None even passes a simple test. The closest call is the 10-bar move on hourly crypto (53.4% lower against 50.6% for random bars), rated no clear edge because its excess return over the rest of the market is close to zero. On 4-hour crypto, the 2:1 trade hit its target less often than on random dates (14.8% against 16.7%, on 54 trades) but lost less per trade (−0.23 R against −0.36 R), both gaps within the margin of error.

Technical confluence

Since it shows no proven edge on its own, a trader can combine it with other signals and technical indicators that help filter the candlestick setups. The important point is to know which filters we actually measured. Here is what our data says about two of them, and which ones we have not tested.

Using volume to validate the reversal

Is the pattern more meaningful on heavy volume? We measured it on hourly crypto data: occurrences printed with volume above 1.5 times the 20-bar average were followed by a decline in 52.6% of cases after 10 bars, against 54.1% on normal volume. On hourly charts of all markets, their 2:1 setup reached its target about as often (25.0% against 24.4%), with a better result per trade (+0.13 R against −0.14 R), a descriptive gap. Heavy volume did not make a move down more likely. This breakdown is descriptive analysis, not corrected for the number of tests, so treat both signals with care.

Combining with support and resistance levels

A setup that forms right under a known resistance level, such as a prior peak, a round number or a falling trendline, has a reason behind it in the classic reading: sellers who waited at that level are the ones writing the upper wicks. That is the classic use of support and resistance here. We have not measured it, so treat these levels as context, not as a proven filter.

Integrating oscillators and technical indicators

A trader can also look for a divergence on the RSI or MACD: price makes a higher high on the third candle while the indicator does not. It is the same idea as the shrinking bodies, measured another way. We have not tested this filter: if you use it, backtest it on your own market. Indicators can help you read the price movement, but none of these signals provides certainty.

The 200-period moving average on hourly charts

We split hourly occurrences by position against the 200-period moving average. Above it, 51.5% declined after 10 bars; below it, 55.2%. Random bars of the same assets were lower in 50.0% of cases. The trend filter changed little. This split is descriptive, not in the tables and not corrected for the number of tests.

Reliability and success rates

The advance block pattern is not a reliable bearish signal on its own: on every market and timeframe we tested, its average excess return over the rest of the market did not pass our statistical test. No line passes even a simple test, on either measure. Two results come closest, described below the table: the 10-bar move on hourly crypto, rated no clear edge, and the 2:1 trade on 4-hour crypto, which hit its target less often than random dates but lost less per trade. Here are the numbers.

How to read the table

Read it in pairs of columns. On daily US stocks, 45.5% of occurrences had a lower price 10 days later. That sounds like a failure, until you see that only 45.2% of random days on the same stocks declined: stocks rise more often than they fall. The fair comparison is always with random bars, and here the gap is under one point, which is not statistically significant.

The second pair of columns tests the setup described above, sell stop below the pattern, stop above it, target at twice the risk, against the same setup placed on random dates. Expectancy is the average result per trade in R, where 1 R is the amount risked, before fees and slippage. On daily stocks it was +0.01 R, against −0.09 R for random dates: both close to zero, with no meaningful difference.

Hourly crypto: the closest call, still not an edge

On hourly crypto charts, 53.4% of 2,606 occurrences were lower 10 bars later, against 50.6% for random bars, with an average excess return of +0.01% (95% confidence interval from −0.22% to +0.23%). The excess return sits close to zero, well inside its margin of error: rated No clear edge, like every other line, and the move is small. Its 2:1 setup reached the target in 24.7% of cases against 20.7% on random dates, with an expectancy of −0.01 R against −0.22 R: roughly break-even before costs, and the gap with random dates is not statistically significant.

Forex: within the margin of error

On hourly forex charts, 50.1% of 1,663 occurrences were lower 10 bars later, against 49.1% for random bars, a gap rated no clear edge. On daily forex charts, the setup reached its target in 28.6% of cases against 0.0% on random dates, with an expectancy of +0.50 R against −0.80 R: on only 7 trades, too few to judge. Neither is a result to act on.

Weekly stocks: below random dates

On weekly stock data, the setup reached its target in 9.3% of cases against 23.9% on random dates, for an expectancy of −0.70 R against −0.23 R. The gap in average result per trade points the wrong way for a seller, but it does not pass our statistical test, so the verdict stays No clear edge. Shorting this weekly setup on stocks has been a losing idea in our data.

How often does it occur?

It is common for a three-candle formation: one occurrence every 3,833 daily bars across all markets, every 3,947 hourly bars, and every 1,305 hourly bars on forex. The full method is published in our methodology.

Real examples of the advance block pattern

The gallery in this section shows occurrences, picked by our bench, with their actual outcome. The three cases described next, on daily charts of two stock indices and a US small cap ETF, share the most frequent outcome: the sell stop was never triggered, because price broke above the top of the pattern first. On daily charts, all markets together, 85.8% of setups were canceled that way, a rate the tables on this page do not show.

Nasdaq 100, January 1988: the warning was right, the entry never came

From January 4 to 6, 1988, the Nasdaq 100 printed three green candles closing at 161.81, 164.28 and 166.175. Each body was smaller than the one before (5.56, then 2.47, then 1.90). The sell stop sat below the low of the pattern at 156.25, with the stop at 166.745. The next day, price traded up to 168.705, above the top of the pattern: the setup was canceled. Then the rally faded: the index closed 8.3% lower 10 days later. The decline it warned about did come, but only after a final push higher.

IWM, January 2018: never sell a strong rally on one warning

From January 19 to 23, 2018, the IWM small cap ETF printed three green candles closing at 142.414, 143.178 and 143.636 on adjusted prices, with bodies shrinking from 1.68 to 0.94, then 0.65. The next session, price traded up to 144.237, above the pattern top of 143.941, and the setup was canceled. IWM then fell and closed 6.4% lower 10 days later. A trader who sold at the close of the third candle, without waiting, with a stop above the pattern, would have been stopped out on that push, before the drop.

Hang Seng, February 2019: the trend went on

From February 8 to 12, 2019, the Hang Seng index printed three white candles closing at 27,946.3, 28,143.8 and 28,171.3, the last with a small body (78 points, against 238 and 216 for the first two). The next day, price traded up to 28,533.4, above the pattern top of 28,219.6, and the setup was canceled. The index closed 2.1% higher 10 days later.

Quick checklist before you act

1. Is the market rising before the first candle, with the close before it at least one average true range above the close 10 bars earlier, and in the upper half of the last 40 closes?
2. Is the first candle a large bullish candlestick?
3. Are the three candlesticks consecutive, with no other candle between them, and do they show a shift from strength to hesitation?
4. Does each following candle open inside the body of the candle before it?
5. Do the bodies shrink, or do the second and third candles show a tall wick on top?

If one of these five answers is no, you are looking at something else: three white soldiers, a stalled pattern, or just three ordinary sessions. If all five are yes, you have an advance block, but not yet a reason to act: wait for a red candle that closes below the third candlestick body, or for a break below the whole pattern, the rule our bench uses.

Tools and resources to learn the pattern

If you want to learn the shape, start by scrolling back through a daily chart of a strong market and marking every place where three rising sessions close higher with shrinking bodies: it shows how often the trend kept going, so you know what to expect. Charting tools can help you find it: TradingView has community scripts for candlestick patterns, and the free open-source TA-Lib library provides the CDLADVANCEBLOCK function. Check how each tool measures shrinking bodies and tall shadows on each candlestick, because the thresholds differ and so do the occurrences it flags. To learn the background, Nison's book cited above covers this pattern and its cousins.

Common mistakes to avoid when trading this pattern

• Selling too early. In 85.8% of daily cases, price broke above the pattern before breaking below it. Entering at the close of the third candle means trading against that majority.
• Confusing it with its bullish twin. Three large candles that close near their highs after a downtrend are a different pattern, with the opposite meaning.
• Ignoring the trend. The pattern only exists after a rise. Three small candles in a trading range are noise, even if the third one prints a new high.
• Expecting a large decline. Even when it works, the typical movement is small: on hourly data, 54.8% of occurrences were lower one bar later, against 48.9% for random bars (descriptive figures, not in the tables), and the advantage did not grow with time.
• Using a tight stop. A stop placed inside the pattern can be hit by ordinary swings. Above the top of the pattern, or no entry.

Advance block pattern frequently asked questions

Summary of key points

• The pattern is three consecutive white candles with higher closes in an uptrend, with shrinking bodies or tall upper wicks.
• It is a bearish warning, which Nison reads as a reason to protect long positions rather than to sell.
• On 2,212 daily occurrences, 45.5% were lower 10 days later, against 45.4% for random bars on the same assets.
• With a sell stop below the pattern, 85.8% of daily setups were canceled before entry because price broke above the pattern first.
• In technical analysis terms, the shape suggests a pause in a bullish trend, not a top: the most important step is to wait for confirmation.
• The 2:1 setup showed no proven edge on any market or timeframe, and reached its target less often than random dates on weekly stock data.

Final thoughts on risk management

Use this candlestick pattern as a reason to look closer, not as a reason to sell. If you own the asset, it is a good moment to move your stop under the low of the pattern. If you want to sell, wait for price to break below the pattern, keep the stop above its top, and size the position on that distance. Keep the risk per trade small and fixed, whatever the pattern.

This page is for educational purposes only and is not investment advice. Past results do not guarantee future results, and trading involves a risk of loss.

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.