Morning star and evening star: reading the turn three candles are supposed to show

The morning star is an early multi-candle pattern many beginners meet, and an easy one to misread. A tidy diagram labelled "bullish reversal" can encourage readers to match the shape before checking trend, location, third-candle depth and market context.

This piece takes it apart: which candle the name actually refers to, what each of the three is telling you, the conditions that separate "looks like one" from "is one", and why the textbook gap requirement needs reinterpretation on a crypto chart. Up front, as always: this describes what the candles are showing. It makes no predictions and gives no buy or sell signals.

Which candle is actually the "star"

Start with the name, because it's routinely misread. The "star" is not the group of three — it's the small middle candle on its own. In candlestick terminology a star is a candle whose body is squeezed almost flat and which sits apart from the move around it. Everything else in the pattern exists to give that one small body a context: the candle before says which way price was going, and the candle after says whether the other side took over.

Get that straight and the family of names sorts itself out. A plain morning star needs a small middle body. A morning doji star is the version where that middle candle is specifically a doji, with the open and close almost identical. An abandoned baby is stricter again: an isolated doji with a gap on either side of it. Same skeleton, tightening requirements on the middle candle. The evening star is the mirror image at the top of a move, and it has the same variants.

Our candlestick pattern guide tool lists these under "morning star" and "evening star", so if you meet the other names elsewhere you're looking at the same shape with a stricter middle candle, not a new pattern to memorise.

What each of the three candles says

Read in order, the three candles tell a story with three beats: the old direction, a stall, and the other side taking over. Candle by candle:

Candle 1 sets the scene. In a morning star it's a red candle still heading down; in an evening star, a green one still heading up. There's nothing clever about this candle — its job is to establish the direction the pattern is supposed to be turning away from. Without that background, "reversal" has nothing to reverse.

Candle 2 is the star, and the heart of it. Its body is squashed flat, open and close close together — a doji, or a small green or red body. What it says is that the side that had been pushing ran out of steam: both sides shoved, neither moved price far. Location matters too — in a morning star this small body usually pokes to the lowest point of the three; in an evening star it sits at the highest. It's worth glancing at how long its wicks are and how much got pulled back, which is the subject of what long upper and lower wicks are telling you.

Candle 3 is the one that decides it. A morning star needs a green candle that closes back above the midpoint of candle 1's body; an evening star needs a red one that drives back below that midpoint. The first two candles only build the stage — without a convincing third, nothing happened. If the third candle has little or no wick (a marubozu type), its OHLC range is concentrated between open and close; that does not prove there was no intraperiod pullback.

Read together, the three candles say something fairly plain: the side that had been pushing stalled, and the other side took the period after that. It's a description of a handover that already happened. What candle four does is not part of the deal.

Four conditions, three disqualifiers

Between "looks like one" and "is one" there are a few hard tests. All four of these need to hold:

• The location is right. A morning star has to come after a real decline, an evening star after a real advance. This condition is easy to skip and essential to whether a "turn" exists: with no prior direction, there is no turn. Work out whether you're in an uptrend, a downtrend or a range first — that's covered in how to tell what the trend is.
• Candle 2's body is genuinely small. Small enough to be obviously out of scale with the candles either side. There's no universal threshold, but the working rule is simple: if the middle body is anywhere near as long as the first one, it isn't a star.
• Candle 3 closes far enough back. Past the midpoint of candle 1's body, at minimum. The deeper it closes, the more it's worth; a token push back means the side taking over didn't have much behind it.
• The timeframe isn't tiny. A set on the daily or 4-hour is a different animal from a set on the 1-minute, where three candles cover three minutes and this shape appears constantly. For picking one, see which candlestick timeframe to use.

Any one of these three, on the other hand, and the name doesn't apply:

• No prior trend. A red-small-green sequence inside a range is just what a range looks like.
• A middle body that isn't small. No star, no pattern, however textbook the other two candles are.
• A third candle that can't reach the midpoint. That's a weak bounce, not a handover.

Working down a checklist beats memorising the picture. If you like to compare as you go, keep the pattern guide tool open on the morning star and evening star entries and hold them next to your own chart.

Why true gaps are less common in 24/7 crypto

Traditional candlestick definitions add one more requirement we haven't mentioned: the small middle body should be gapped away from candle 1's body — hanging clear of it, with no overlap. Apply that rule strictly and many candidates on a continuously traded chart will not qualify as textbook morning stars.

The reason isn't the pattern, it's the market. Stocks close and reopen, so overnight information can land in the next session's opening trade and leave a band of prices where nothing traded. Binance spot matches orders around the clock. On liquid pairs, consecutive candles usually meet end to end, so true gaps are uncommon. Uncommon is not impossible: they can still appear in illiquid markets or where bid-ask spreads are wide. Binance Academy's candlestick-pattern guide makes that same distinction and cautions against using a gap or any single pattern as a standalone signal.

This guide does not use the gap as an automatic disqualifier on a crypto chart. It puts more weight on whether the middle body is genuinely small and whether the third candle closes far enough back. If the gap evidence is absent, location, volume and follow-through have to carry more of the context.

Relaxing the rule lets more three-candle lookalikes pass the first visual check, so false-positive risk rises. This page does not define a dataset, pair list or timeframe and therefore does not assign those lookalikes a success or failure rate. The useful response is to screen for location and volume first; whether candle 3 actually had volume behind it is covered in reading volume against price.

Three ways beginners get it wrong

One: hunting for stars with a magnifying glass. Look hard enough and any range gives you a red candle, a small one, and a green one. Hunting for a shape and reading a shape are different jobs. The order that works is: decide whether this location deserves attention — is there a prior trend, is there a support or resistance level nearby — and only then look at what the candles are doing. Doing it the other way round, finding the shape first and inventing the context afterwards, is just talking yourself into something. Which levels count is covered in how to find support and resistance.

Two: calling it before candle 2 has closed. That small body only exists after the candle closes. Watch one still ticking and it may look like a perfect star right now, then grow into a long red candle ten minutes later. Calling "morning star!" early uses unfinished data. Wait for all three to close.

Three: treating recognition as a forecast. The expensive one. The pattern describes how the last three candles changed hands; it promises nothing about the fourth. The "spot a morning star, buy the bottom" posts you'll see are missing exactly the layers that make it worth anything — location, volume, follow-through — because a pattern packaged as a conclusion is what gets people to act now. We unpack that style of setup in 8 chart traps beginners misread.

Once you've spotted one, a cautious next step is to wait

A cautious next step after identifying a morning star is to do nothing yet and see whether the following candle or two pick the direction up. Three candles sort a recent handover into a category; that's a long way short of evidence a trend has actually turned. A fourth candle that carries on adds something real. A fourth candle that swallows the third back means those three were an episode, nothing more.

The other two things worth a second look are location and volume: is this sitting at a level price has tested before, or floating in mid-air? Did candle 3 come with visibly more volume than the two before it? Those answers tell you more than how textbook the shape is. Pattern, location, volume — you've read it properly when all three line up. What to do about it is not something this site advises on. To fill in the wider framework, work through the candlestick pattern guide.

A no-risk chart-replay drill

Use chart replay for a drill that involves no order. Open a daily chart, scroll back, find a promising location, and stop after the first two candles. Write down what candle 3 would have to do, such as closing above the midpoint of candle 1's body. Then move the chart one candle to the right and check each of the four conditions.

Record the condition that fails: whether candle 3 missed the midpoint, whether there was a real decline beforehand, and whether location and volume support the reading. The purpose is not to manufacture a win rate or prove the pattern predicts the next candle. It is to reduce shape-only misclassification. Combination patterns are built on single candles, so if open, high, low, close and wicks aren't second nature yet, a pass through how to read a candlestick chart first will save you time.

FAQ

What's the difference between a morning star, a morning doji star and an abandoned baby?

They're the same three-candle family with progressively stricter middle-candle rules. In a plain morning star the middle body only has to be small. In a morning doji star it must be a doji. An abandoned baby is stricter again: the doji is isolated by gaps on both sides, including the wicks. On a 24/7 crypto chart those gaps are uncommon, not impossible; they may still appear in illiquid markets or where bid-ask spreads are wide, so the pattern should not be treated as a standalone signal.

Does the middle candle have to be a doji?

No. What matters is that the body is small enough to be obviously out of scale with the candles either side of it. A doji, where open and close are almost identical, is the cleanest version, but a small green or small red body counts too. If the middle body is anywhere near the size of the first one, you don't have a star — you have three ordinary candles standing next to each other.

True gaps are uncommon on Binance charts — does a morning star still count?

It can count; the emphasis just moves. Classic pattern books ask for a gap between the middle body and the first candle's body, a rule inherited from markets that close and reopen. Binance spot trades around the clock, so true gaps are uncommon on liquid pairs, not impossible. They may still occur in illiquid markets or where bid-ask spreads are wide. For this guide, the practical checks are whether the middle body is genuinely small, whether the third candle closes far enough back, and whether location and volume support the reading. The pattern is context, not a standalone signal.

Next time someone points at three candles and announces a morning star, you've got three questions to ask before you agree: had price actually been falling into it, is that middle body genuinely small, and did the third candle close back past the halfway mark? Get through all three and the name holds. Stall on any of them and they're just three candles in a row.