A single candle records one session. A two-candle pattern records a session and the response to it. A three-candle pattern records something more useful: a session, a pause, and then a third session that settles the question the first two raised. That third candle is the reason three-candle patterns are slower to complete than the shorter ones, and the reason the better-known ones have the most defensible readings in the candlestick literature.
A three candlestick pattern is a formation read across three consecutive candles, where the third confirms or completes what the first two began. The best known are the morning star, evening star, three white soldiers and three black crows. Each needs a prior trend, and each is invalidated by a close back through a specific candle.
Bulkowski lists more than thirty named three-line candle patterns. Most of them are rare enough that you will never see a clean one. Six are worth knowing, and they fall into two groups: four classics, and two that are explicitly a shorter pattern plus a confirming day.
Why does the third candle matter in a three candlestick pattern?
In a two-candle pattern, the second candle makes a claim and you wait for the next session to see if it holds. The confirmation candle post covers why that wait is the step most people skip. In a three-candle pattern, that wait is already part of the definition. The pattern is not complete until the third close, and the third close is required to do something specific: close into the first candle’s body, close above the prior close, or extend a sequence.
The cost is time. By the time a three-candle pattern is complete, a good part of the move it describes may already have happened. That is the trade the shape makes: less ambiguity, later.
The four classics
The morning star
The morning star forms after a decline: a long down candle, then a small-bodied candle whose body sits below the first one’s body, then a long up candle that closes at least halfway into the first candle’s body. Read structurally, it is selling, then a stall where selling stops making progress, then buyers taking back half or more of the original down session.
- Needs: a visible decline before it. Without one there is nothing to reverse.
- Confirmed by: the third candle itself, closing past the midpoint of the first. A fourth session continuing higher adds to it.
- Invalidated by: a close below the low of the middle candle, which breaks the base the pattern is built on.
Bulkowski’s identification guidelines call for the middle candle’s body to gap below the first body and the third candle to gap above the middle one. On many daily stock charts those gaps appear; on markets that trade around the clock they rarely do, which is why the shape is often drawn without them.
The evening star
The evening star is the mirror: after an advance, a long up candle, then a small-bodied candle above it, then a long down candle that closes at least halfway down the first candle’s body.
- Needs: a visible advance before it.
- Confirmed by: the third candle closing below the midpoint of the first.
- Invalidated by: a close back above the high of the middle candle, which reclaims the stall level and says the pause was a consolidation rather than a top.
Three white soldiers
Three white soldiers are three consecutive long up candles, each opening within the previous body and closing at or near its own high. It is not a pause and a resolution; it is three sessions of the same side winning, with very little given back inside any of them.
- Needs: a decline before it (Bulkowski’s guideline is a downward price trend; a long base is the looser traditional allowance). Three strong up candles in the middle of an existing advance are an advance, not a pattern.
- Confirmed by: the third close, largely. Volume building across the three sessions and price clearing a level that previously capped it add weight.
- Invalidated by: a close back below the open of the first soldier, which retraces the whole sequence.
Three black crows
Three black crows mirror the soldiers: three long down candles after an advance, each opening within the previous body and closing at or near its low.
- Needs: an advance or the top of a range before it.
- Confirmed by: the third close, with a broken support level adding weight.
- Invalidated by: a close back above the open of the first crow.
The two that are a shorter pattern plus a confirming day
These two are worth knowing for a reason beyond themselves. Bulkowski credits both to Morris, and describes three inside up as a bullish harami with a confirming candle as the third day. Three outside up has the same structure with a bullish engulfing in place of the harami. That makes the pair a clean, if imperfect, natural experiment in what confirmation adds.
Three inside up is a bullish harami followed by a confirming day. After a decline, a tall down candle is followed by a small up candle whose body sits inside the first body (the harami), and then a third up candle closes above the second candle’s close. The harami has no drawing in the library, so here it is in words: the first two candles are a large body followed by a small body nested inside it.
- Needs: a downward price trend before it.
- Confirmed by: the third candle closing above the prior close, which is the definition.
- Invalidated by: a close below the low of the first candle, which says the selling that the harami appeared to interrupt has resumed.
Three outside up is a bullish engulfing followed by a confirming day. After a decline, a down candle is followed by an up candle whose body engulfs it, and the third candle closes higher still.
- Needs: a downward price trend before it.
- Confirmed by: the third candle closing higher than the engulfing candle.
- Invalidated by: a close below the low of the engulfing candle.
Now the numbers, with their limits stated. Bulkowski’s candlestick pages report, for each pattern, how often price broke out in the expected direction in his tests. The statistics are described as based on hundreds of trades per pattern, and the pages do not state the market or the period. On those pages, the bullish harami acts as a bullish reversal 53% of the time; three inside up, the harami plus a confirming day, 65% of the time. The bullish engulfing acts as a bullish reversal 63% of the time; three outside up, the engulfing plus a confirming day, 75% of the time.
These are separate pattern counts, not a controlled experiment, and requiring one more up close will mechanically filter out some failures. But they point the same way as the argument above. On his figures, the third candle is doing real work.
What do the classics look like on the same measure?
For completeness, on the same pages: the morning star acts as a bullish reversal 78% of the time, the evening star as a bearish reversal 72% of the time, three white soldiers as a bullish reversal 82% of the time, and three black crows as a bearish reversal 78% of the time. Each is one analyst’s count, using his own identification rules and his own definition of a breakout, and none of them transfers to another market, timeframe or set of rules. What they do not measure matters as much: how far price went, whether the move covered trading costs, or what happened on a 5-minute chart of a currency pair. The does technical analysis work post covers how to read practitioner numbers like these.
Note also the frequency ranks on the same pages. Bulkowski ranks the morning star 66th, the evening star 71st and three white soldiers 67th of 103 patterns for how often they appear. Clean versions are not common. If you are seeing one every week on the same chart, the definition being applied is looser than the one the numbers came from.
Where three-candle patterns mislead
The prior trend is assumed rather than checked. Every pattern above needs a move before it. The reversal candlestick patterns post shows the same shapes in the middle of a trend, where they mean something different or nothing.
The middle candle is too big. The star patterns depend on the middle candle being a genuine stall. A middle candle with a large body is a two-sided session, and the shape becomes something else.
The soldiers or crows are exhausted, not beginning. Three long candles after an extended run can be the end of the move rather than the start of one. The pattern carries most meaning when it emerges from a base or a decline, not when it extends a move that is already long.
The invalidation is never written down. Each pattern above has a price that ends the read. If you cannot name it before the next session opens, you have recognised a shape, not read a pattern.
For the single-candle and two-candle patterns on either side of this list, the bullish candlestick patterns and bearish candlestick patterns roundups carry the same context and invalidation for each, and the candlestick pattern library has all eighteen drawn to proportion.
Sources
Every figure above was checked against these before publishing.
- Bulkowski, Morning Star
- Bulkowski, Evening Star
- Bulkowski, Three White Soldiers
- Bulkowski, Three Black Crows
- Bulkowski, Three Inside Up
- Bulkowski, Three Outside Up
- Bulkowski, Bullish Harami
- Bulkowski, Bullish Engulfing
- Bulkowski, Three Line Candlesticks