If there’s one pattern every candlestick book calls a reliable bottom, it’s the morning star. Three candles: a big down day, a small “star” that gaps lower, and then a strong up day that climbs well back into the first candle. It looks like a turn, and it’s named after the planet that shows up just before sunrise.
I expected it to do well when I ran our scanner’s three years of data through it. It did the opposite.
How it’s built
- Day one: a long down candle, sellers in control.
- Day two: a small body that gaps below day one’s body. Indecision at the lows.
- Day three: a strong up candle that closes well into day one’s body. TA-Lib, which the scanner uses, wants it to recover at least 30% of the way.
A version where day two is a doji is called a morning doji star.
The numbers
| After a… | Cases | Higher 1 day later | Higher 5 days later | Higher 20 days later | Typical 5-day move |
|---|---|---|---|---|---|
| Morning star | 1,826 | 50.8% | 49.8% | 52.3% | −0.01% |
| Morning doji star | 655 | 51.5% | 51.6% | 53.9% | +0.16% |
| Any day | 413,576 | 51.9% | 53.8% | 55.6% | +0.34% |
A morning star was followed by a gain less often than a random day at every horizon. Five days out, the stock was higher 49.8% of the time, four points below the baseline. With 1,826 cases the margin of error is about 2.3 points, so this is unlikely to be a fluke.
Maybe it needs a downtrend first?
That’s the whole point of the pattern, so here it is split by the 10 days before it:
| Before the morning star | Cases | Higher 5 days later | Any day in the same spot |
|---|---|---|---|
| Fell more than 5% | 382 | 48.7% | 54.0% |
| Moved less than 5% | 1,238 | 49.0% | 53.1% |
| Rose more than 5% | 160 | 52.5% | 53.7% |
After a real selloff, the exact situation it’s designed for, it was the weakest of all.
Waiting for confirmation didn’t save it either. When the day after the morning star closed higher again, the stock was up five days later only 48.8% of the time, against 52.7% for any up day.
Why might it fail?
My best guess: by the time all three candles are in, the stock has already bounced hard on day three. A lot of the “reversal” has happened, and short-term traders who bought the low take their profit in the following days. It’s the same thing I saw with the bullish engulfing and the piercing line: sharp one-day bounces after a selloff tend to give some back first.
What I’d do with it
Not buy it on day three. If you like the stock anyway, the data says you’ll usually get a better price over the next week. Selling a cash-secured put below the star’s low is one way to get paid while you wait for that.
Recent examples
The scanner flagged 57 morning stars in October 2026, including NTRS on October 23, WMT on October 22 and the Dow ETF DIA on October 20. Today’s list is on the candlestick scanner, and the full comparison is in Do Candlestick Patterns Work?
Data: daily prices for 559 US stocks and ETFs, October 2023 to October 2026, patterns detected by TA-Lib. Educational only, not financial advice.