Bullish 3-CandleHeuristic1 min read
◆ The PatternLong red → small star → long green — dawn after night
The Morning Star is a three-candle bullish reversal. First, a long red candle (sellers in control). Then a small-bodied "star" (indecision — sellers exhausting). Finally, a long green candle (buyers take over). Named because it appears before dawn — the trend's darkest point.
Red (long) → Star (small, gapped) → Green (long)
The star can be a doji (called a "Morning Doji Star") — this strengthens the signal.
// Morning Star formation
Key insight: The gap between the first candle and the star, and between the star and the third candle, adds to the signal's power — but the pattern works without gaps too.
Pattern bridge: A three-candle reversal: despair, indecision, hope. In ML, warmup schedules follow the same cold-start-to-recovery arc.
Worked example
Day 1 is a long red bar from 105 to 100. Day 2 has a small body (99 to 99.3) that gaps below day 1’s close. Day 3 is green and closes at 103.5, above day 1’s midpoint of 102.5. All four conditions hold: a morning star.
The example data
The bars the example is worked on.
| Bar | Open | High | Low | Close |
|---|---|---|---|---|
| 1 | 105 | 105.5 | 99.5 | 100 |
| 2 | 99 | 99.8 | 98.5 | 99.3 |
| 3 | 100 | 103.8 | 99.8 | 103.5 |
Where it misleads
- Three bars leave many ways to bend the definition (how long, how small, gap or not), and results depend on the exact rules.
- On markets that trade around the clock the required gap rarely exists.
- Nison presents it as a reversal sign in context — at support, after a decline — not as a signal on its own.
In code
d1, d2, d3 = df.iloc[-3], df.iloc[-2], df.iloc[-1] long_red = d1.open - d1.close > 0.6 * (d1.high - d1.low) small = abs(d2.close - d2.open) < 0.3 * abs(d1.close - d1.open) gap_down = max(d2.open, d2.close) < d1.close mid = (d1.open + d1.close) / 2 strong_green = d3.close > d3.open and d3.close > mid is_morning_star = long_red and small and gap_down and strong_green
Assumes import numpy as np, import pandas as pd, and a DataFrame df with columns open, high, low and close for the data above.
Sources
- Japanese Candlestick Charting Techniques, S. Nison, New York Institute of Finance, 1991
- B. R. Marshall, M. R. Young & L. C. Rose, “Candlestick technical trading strategies: Can they create value for investors?”, Journal of Banking & Finance 30(8), 2006 — they found none on Dow Jones stocks
Last reviewed October 2026 · How this site checks its content · Report a mistake
Linked from 1