The morning star is a three-candle bullish reversal; the evening star is its bearish mirror. Each is a strong trend candle, then a small indecision candle, then a strong candle in the new direction - a visual sequence of momentum stalling and then flipping.
What it looks like
A morning star forms at a bottom: a long down candle, then a small-bodied candle that gaps or stalls lower (the star), then a long up candle that closes well into the first candle's body. The evening star is the reverse at a top - up candle, hesitation, then a strong down candle. The deeper the third candle pushes back into the first, the stronger the reversal.
Anatomy at a glance
How to trade it
A candle is a trigger, not a system. Take it only where it means something - at a level, with the trend - and give it a clear stop and target:
Where traders go wrong: taking the candle anywhere. In the middle of a range this signal is noise. Wait for it to form at support or resistance or in line with the trend, and size the trade with the position size calculator so a failed signal costs a small, fixed amount.
What the measured data says
We tested both stars against a matched-risk random control - same direction, same stop distance, same bar - across 28 instruments and 698,580 bars, with real measured spreads applied. Both patterns carry a genuine edge. Neither clears the cost of trading it.
| Pattern | Timeframe | Signals | Win rate | Matched random | Edge | Significance | Profitable after costs? |
|---|---|---|---|---|---|---|---|
| Morning star | Daily | 5,059 | 47.7% | 46.0% | +1.7 pts | p < 0.05 | No |
| Evening star | Daily | 4,937 | 44.8% | 42.9% | +2.0 pts | p < 0.05 | No |
| Morning star | 4-hour | 23,808 | 43.2% | 41.8% | +1.4 pts | p < 0.001 | No |
| Evening star | 4-hour | 24,092 | 40.7% | 40.0% | +0.6 pts | not significant | No |
Read the win rate and the matched-random column together, because on its own the win rate misleads. A morning star winning 47.7% of daily 1:1 trades sounds respectable - until you see that a random entry taken long from the same bar with the same stop won 46.0% of the identical situations. The pattern itself is worth 1.7 points, not 47.7. That is a real signal, but it is a fraction of what most course material implies, and it sits below the 50% a 1:1 trade needs just to break even before spread.
Two details are worth carrying to the chart. The evening star's daily edge (+2.0 points) is slightly larger than the morning star's, which is the reverse of the usual teaching that bullish reversals are the more reliable half. And the morning star's edge survives the drop to 4-hour charts at high significance across 23,808 signals, where most of the twelve patterns we tested lose their edge entirely. Full tables for all twelve patterns, both targets and both timeframes, are in the candlestick pattern win-rate study.
Prove it before you trade it
This candle is a hypothesis about what happens next, nothing more. Replay real charts, take the pattern only at meaningful levels across a large sample, and read the win rate and risk-reward it delivers on your pairs before you rely on it.
Morning & Evening Star FAQ
What is a morning star candlestick pattern?
A morning star is a three-candle bullish reversal that forms at the bottom of a downtrend: a long down candle, a small indecision candle, then a long up candle that closes well into the first. It shows selling momentum stalling and buyers taking over, and is strongest at a support level.
Is an evening star bearish?
Yes, the evening star is a bearish three-candle reversal that forms at the top of an uptrend. It runs up candle, then a small hesitation candle, then a strong down candle that pushes back into the first. It signals buying momentum fading and sellers stepping in, especially at resistance.
How reliable are star patterns?
They are more reliable than single candles because the three-bar sequence shows a genuine shift, not a one-bar spike. Reliability still depends on where they form - at a tested level and against an extended move they are far stronger. As with any pattern, backtest it on your pairs before sizing up.