Morning Doji Star — A Stronger Version of the Morning Star?

The textbook story goes like this: the morning star is good, but a morning star with a doji in the middle is better, because a doji means "deeper" indecision and therefore a stronger reversal. Sounds logical. There's just one problem: when both patterns were measured across 4.7 million candles, the "stronger variant" turned out to be slightly weaker — a 76% reversal rate versus 78% for the plain morning star, and rank 25 versus rank 12. Still a very good pattern, one of the better ones in the catalog. It's just that the legend of the doji power-up didn't survive contact with the data.

What the Pattern Looks Like

The Morning Doji Star is a three-candle bottom-reversal pattern — a morning star variant in which the middle candle is a doji:

• Trend before the pattern: downward. Without a move down, there's nothing to reverse — this is a hard requirement.
• Candle 1: a large bearish (black) candle. Sellers are fully in control; the market slides with no resistance.
• Candle 2: a doji — open and close practically at the same level. On stocks, the doji's body gaps below the body of the first candle. This is the "star": a session where the selling momentum completely dies out and the market ends in a draw.
• Candle 3: a large bullish (white) candle whose body gaps up above the doji and closes deep into the first candle's body. Buyers take over the market and erase the decline.

The sequence tells the same story as the plain morning star: sellers dominate → a complete stall → buyers take over. The difference is cosmetic — a doji instead of a small body — which is exactly why the results for both variants come out nearly identical.

The same adaptation applies to crypto as with the plain morning star: BTC/ETH have no gaps on the daily chart, so the gap conditions get loosened. We look for the sequence: a large red candle, a doji (or a candle with a negligible body) in the lower part of the range, a large green candle closing at least halfway into the first. Worth knowing the cousin too: if the doji is separated by gaps from BOTH neighbors, wicks included, that's already an abandoned baby — a separate pattern on stocks, but on crypto the distinction is purely academic.

What the Numbers Say (Honestly)

Results from Bulkowski's tests (~4.7 million daily candles, US stocks):

• Trend reversal: 76% of cases — confirmed across 932 examples. A high result, and more reliable than for many rare patterns, though still far from the 20,000 samples the author considers comfortable.
• Overall performance rank: 25/103. The top quarter of the catalog — the move after the breakout is solid; the pattern doesn't just reverse, it has follow-through.
• Frequency: 78/103 — a rare pattern. A doji in an exact spot within a three-candle sequence is a demanding configuration; you won't see it every week.
• Family comparison (the key point of this article): plain morning star — 78% reversal, rank 12; morning doji star — 76%, rank 25. Both variants sit near the top of the catalog, but the data gives ZERO support to the claim that the doji strengthens the signal. If anything, it slightly weakens it — though the difference is small enough that it's more honest to call the variants equivalent.
• From deeper in the data: the best average 10-day move was -6.25%, following a downside breakout in a bear market — that is, when the pattern failed. Before drawing conclusions: that sample was just 33 cases, and Bulkowski himself says to treat it skeptically. His more interesting qualitative observation: the best morning doji stars are the ones that end a pullback within a larger uptrend — the upside breakout then joins an existing move. Patterns near yearly highs reversed the trend most often, and tall (stretched-out) patterns beat short ones.

The standard caveat: this measurement is from US stocks, daily timeframe. On crypto — a 24/7 market, no gaps, different volatility — we carry over the structure and logic, not the percentages.

How to Trade It / How NOT to Trade It

How to trade it:

1. Best context: the end of a pullback in an uptrend. Straight from the data: a pattern ending a correction within a larger uptrend performs best, because the breakout has a tailwind — it joins an existing move. A star at the bottom of an extended bear market is fighting the higher-order trend: possible, but statistically harder.
2. Confirmation: a close above the pattern's high. The breakout is a close above the highest point of the three candles. Entering on the third candle's close shortens the distance to your stop but drops the filter the whole statistic rests on. On crypto, add rising volume on the third candle and confluence with an H4/D1 support level.
3. Selection: tall patterns, meaningful levels. Stretched-out patterns (a large high-to-low span across the three candles) outperformed tight ones. A doji landing exactly on prior support raises the quality of the setup — the level does the work, the pattern confirms it.
4. Stop loss and target. Stop below the pattern's low (usually the doji's wick) with a buffer for noise — breaking this level overturns the whole narrative of buyers taking over the market. Target: take partial profit at the nearest resistance, trail the rest with a trailing stop — rank 25 justifies giving the move time.

How NOT to trade it:

• Don't enter on the doji alone. Two candles aren't a pattern. A doji after a large bearish candle could just as easily be a stopover before another leg down — without the third candle you know nothing (and remember, a doji by itself is a statistical coin flip).
• Don't accept a weak third candle. A close barely above the doji isn't a takeover. The "at least half of the first candle's body" condition is a hard rule — without it, the statistics don't apply.
• Don't pay extra for "doji magic." Since the data doesn't distinguish the doji variant from the plain one, don't treat it preferentially — don't size up, don't loosen your stop "because it's the stronger version." It isn't.
• Don't spot the pattern inside consolidation. Without a clear downward move before the setup, three well-shaped candles are noise, not a morning doji star.

Myth vs. Measurement

Myth: "A doji in the middle makes the morning star stronger." Measurement: 76% and rank 25, versus 78% and rank 12 for the plain version. The variants are at best equivalent — and if either has an edge, it's the "plain" one. The intuition that "deeper indecision = a stronger reversal" is nice, and wrong.

Myth: "The morning doji star is a bear-market-bottom pattern." Measurement: the best-documented context is a pullback in an uptrend, and patterns near yearly HIGHS reversed the trend most often. Catching the bottom of a months-long decline is the hardest possible application.

Myth: "Without perfect gaps, the pattern doesn't count." Measurement and practice: gaps are an artifact of the stock market's trading halts. On crypto, their absence doesn't invalidate the setup — what matters is the sequence of forces: sellers, a full stall, buyers taking over. We don't carry the stock-market percentages over 1:1 anyway, so clinging to the letter of the gap definition protects nothing.

Myth: "76% effectiveness means you can trade every such pattern." Measurement: 76% applies to textbook cases on the daily chart, with breakout confirmation, on stocks. A pattern with no trend before it, a weak third candle, or on M15 on an altcoin is a different game — one without this statistic.

Quick checklist:

• Is there a clear downward move before the pattern (ideally a pullback in an uptrend)?
• Is the middle candle actually a doji — open and close practically equal?
• Does the third candle close at least halfway into the first candle's body?
• Does the pattern land on support, with volume rising on the third candle?
• Are you waiting for a close above the pattern's high instead of entering on the doji alone?
• Are you avoiding sizing up "because it's the doji variant" — since the data gives it no edge?

The morning doji star is an excellent pattern — just exactly as excellent as its plain sister. The real lesson here is broader: textbooks have spent decades repeating hierarchies ("variant X is stronger") that nobody ever actually measured. Bulkowski measured them. The variants converged, and the legend of the doji power-up turned out to be just another nice story. Trade the structure and the context — not the folklore around the middle candle's shape.

FAQ

Rafał — Krypto Bez Ściemy

Trader and founder of Strefa Tradingu. He’s been breaking crypto down on YouTube for years — no hype, no signals, with a focus on market structure and risk management.

kryptobezsciemy.plInstagramYouTubeKlub Zero