Stalled pattern (deliberation): rules, examples and real results

The stalled pattern, which Japanese traders call deliberation, is two confident white candles followed by a hesitant one. The first two sessions push prices up with long bodies that close at or near their highs. The third session opens near the top of that move, goes almost nowhere, and leaves a small body sitting on the "shoulder" of the second candle.

The name says what the market is doing: thinking. After two days of uncontested buying, the third session is a pause in which neither side commits. Nison classes it as a topping warning for traders who are long, a moment to take some money off the table rather than a call that prices must fall.

Below you will find what separates the stalled pattern from the Advance block candlestick pattern: a rally running out of breath, the precise test TA-Lib runs, a case where the pause came one day before a 9.7% drop and a case where it preceded a 29% gain, and the record of 67 detections on large US stocks.

Anatomy of the stalled pattern

Three candles, all white, each closing above the last:

1. First candle: long white. A decisive up day.
2. Second candle: another long white that opens inside or just above the first body and closes at or very near its high. The missing upper shadow is important: buyers were still in full control at the bell.
3. Third candle: a small white body that opens at or above the upper part of the second body (sometimes with a small gap) and closes only slightly higher.

The contrast between the two strong sessions and the small third one is the whole signal. Where the advance block shows fatigue building over two days, the stalled pattern shows strength that stops abruptly.

The exact rules Finabase uses

Finabase detects the pattern with TA-Lib's CDLSTALLEDPATTERN on daily bars. All sizes are relative to the stock's own recent candles.

Two things are worth knowing.

"Rides on the shoulder" is measured from the open. TA-Lib does not require a gap. It requires the small third body to start near the top of the second body, so the small candle visibly sits on the prior close rather than inside the prior range.

No trend check. Like most TA-Lib candle functions, this one never looks at the prices before the first candle. Two long white candles and a higher close make an uptrend of at least a few days by construction, and 76.1% of our detections came after a ten-session rise. The rest formed in rebounds within a falling or sideways market.

Why traders read the pause as a warning

Two long white candles that close at their highs usually draw in late buyers: momentum traders, short sellers covering, and people who watched the first day and committed on the second. The third session is the test of whether that demand continues at the new, higher price.

When the third body comes out small, the answer is "not much". The session may still trade higher intraday, and it may even close up a few cents, but it has stopped expanding. The market is deliberating whether the new level is fair. If the next session brings sellers, the late buyers from day two are the first to be under water, which can speed up a decline.

That logic is why the classic response is defensive. Holders bank part of a gain or move a stop up; few traditional sources treat the stalled pattern by itself as an entry for a short position.

What would confirm it

• A lower close below the third candle's low. This is the confirmation rule in our test; 19 of the 67 patterns met it.
• A gap down or a long black candle that cuts back into the second white body.
• A failure at a level that mattered: a previous high, the top of a trading range, or a widely watched moving average.
• Overbought momentum rolling over, for example the Relative Strength Index (RSI): how to read and use it turning down from above 70.

Two real stalled patterns

Netflix, April 16, 2026. After a 12.8% rise in ten sessions, NFLX printed a long white candle on April 14 (body $3.16), another on April 15 (body $1.73) that closed at $107.71 just 14 cents below its high, and then on April 16 a $0.32 body that opened at $107.47 and closed at $107.79. The next session the stock closed at $97.31, down 9.7% in one day, and it was 13.2% lower ten sessions later. The same three candles also met TA-Lib's advance-block rules, which shows how much the two siblings overlap. A one-day drop of that size is the kind that follows news, and no candle pattern forecasts news; what the pattern did capture was that buying had already stopped expanding.

Cisco, May 1, 2026. CSCO rose 6.5% over ten sessions, then printed white bodies of $2.25 and $1.68 on April 29 and 30 and a near-doji on May 1 ($91.75 open, $91.85 close). On the same bar TA-Lib also flagged a doji star and an advance block, so the chart was crowded with bearish labels. The next close was higher, at $92.63, which never confirmed the warning, and ten sessions later the stock was up 28.7%. The pause was a pause, nothing more.

What the data says

Over two years of daily bars (October 30, 2024 to October 29, 2026) for 60 large US stocks and ETFs, CDLSTALLEDPATTERN fired 67 times on 38 symbols.

• The direction count was a coin flip. 46.3% lower after five sessions is the same as any random day.
• The average return leaned negative. −0.26% over five sessions and −0.75% over ten, against positive baselines. With 67 cases, a few large drops like Netflix's carry much of that average.
• Confirmation lowered the average further, not the hit rate. The 19 confirmed cases averaged −1.09% over the following five sessions but were lower only 47.4% of the time.

The stalled pattern and its relatives

The rule of thumb: soldiers keep the bodies long, the advance block lets them shrink gradually, and the stalled pattern shrinks only the last one.

An educational outline

An outline of how the pattern is commonly used. It isn't a recommendation to trade.

1. Ask what you hold. The traditional use is for someone already long: it is a prompt to protect profits from the two strong days.
2. Note the levels. The third candle's low marks where the pause would turn into a decline; the third candle's high marks where buyers would be back in charge.
3. Let the next session vote. A close under the third low confirms; a close above the high, as at Cisco, cancels the warning.

How Finabase shows the stalled pattern

• Chart markers. The Stalled Pattern study marks the third candle of every occurrence TA-Lib finds on the daily chart. Because the advance block often fires on the same bar, adding both studies shows where they overlap.
• Scanner. Detections are sorted by the sign of TA-Lib's output. The stalled pattern is always −100 and appears only on the bearish list.
• Signals tab. The Candlestick gauge on a symbol page such as AAPL's signals takes one vote per pattern: a stalled pattern within the last eight bars is a Sell vote. Votes combine as (buy − sell) / (buy + sell + min(neutral, max(buy, sell))), with Strong Sell below −0.6, Sell below −0.2, Neutral below 0.2, Buy below 0.6 and Strong Buy above.

Frequently asked questions

Why is it also called deliberation?

It is the English rendering of the Japanese name. The third small candle shows a market deliberating over whether the two-day advance went too far, rather than acting.

Does the third candle have to gap up?

No. A small gap is common in textbook drawings, but TA-Lib only requires the third body to open near or above the top of the second body.

Is the stalled pattern the same as an advance block?

They overlap but are not the same. The stalled pattern needs two long white bodies and a nearly shadowless second candle; the advance block allows shrinking from the second candle onward. On some days, such as Netflix on April 16, 2026, both are flagged.

What timeframe works best?

The rules are the same on any timeframe. Finabase's markers and scanner use daily bars, which is what the statistics on this page measure.

Summary

Deliberation is two days of strong buying followed by one of hesitation. It shows that demand stopped expanding at a higher price, not that supply took over. In our sample of 67 cases the pattern was no better than a random day at predicting direction, though the average outcome leaned slightly negative because of a few sharp drops. Its practical value is as a checkpoint for an existing position, with the third candle's low and high marking where the market will show which way the deliberation went.