Morning Star and Evening Star: Every SPY Case Since 1993
A morning star is a three-candle pattern at the end of a decline: a long red candle, a small-bodied candle whose body sits below the first one's close, then a green candle that closes above the midpoint of the first body. The evening star is its mirror at the top of a rise and is read as bearish.
The morning star gets its name from the planet Venus, which appears just before sunrise, and the pattern is meant to show the same thing on a chart: the last dark moment before prices turn up. This page pins down the three candles in numbers, walks through a real SPY morning star from October 2023, and counts every morning and evening star on SPY since 1993.
How it forms
Day one is a long red candle in a falling market. Sellers are in control, and the close is near the low of a wide session. The candle confirms the decline rather than questioning it.
Day two is the star: a small body that sits below day one’s close. The small body means buyers and sellers ended the session near where it began. The position matters more than the color. Its body sits below the first candle’s body, separated by a gap, so the market opened lower still and then stalled.
Day three is a green candle that closes well into day one’s body. The usual test is a close above the midpoint of the first red body. By then the whole of the star’s session and a good part of the first day’s loss have been recovered.
The evening star runs the same sequence at a top. A long green candle in a rising market, a small star whose body sits above the first close, then a red candle closing below the midpoint of the first green body.
The pattern reads as a handover in three steps: sellers in control, a pause, then buyers taking over. A bullish engulfing pattern tells a similar story in two candles, without the pause in the middle.
The exact definition used here
Each test below is applied to SPY daily bars. For a morning star, with days one, two and three:
- day one is red, and its body is at least as large as the average body of the 20 sessions before it;
- day two’s body is no more than a third of day one’s body, and the top of day two’s body is below day one’s close;
- day three is green and closes above the midpoint of day one’s body;
- day one closed below the close ten sessions earlier, so the market had been falling.
The evening star uses the mirror of every rule. The outcome is SPY’s close five sessions after day three, compared with day three’s close.
The second rule, the gap between the bodies, is the easiest one to relax. Drop it and the morning star count on SPY rises from 16 to 124. Both versions are reported below, because whichever one a reader uses, the count and the base rate should travel together.
A worked example
SPY completed a morning star on Tuesday 24 October 2023. The market had been sliding, and on Friday 20 October SPY closed at $421.19, below the $429.54 close of 6 October ten sessions earlier.
Day one, 20 October: open $425.98, close $421.19, a red body of $4.79. The average body over the previous 20 sessions was $2.95, so day one clears the size rule.
Day two, 23 October: open $419.61, close $420.46, a body of $0.85. A third of day one’s body is $1.60, so the star is small enough. The top of its body, $420.46, sits below day one’s $421.19 close, so the gap rule holds.
Day three, 24 October: open $422.65, close $423.63, green. The midpoint of day one’s body is $425.98 plus $421.19, divided by 2, or $423.585. The close cleared it by less than 5 cents.
Then SPY kept falling. It closed at $410.68 on 27 October, and five sessions after day three, on 31 October, it closed at $418.20, which is 1.28% below the day-three close. The lowest close of that slide came three sessions after the pattern completed, not at it.
The original data
Across 8,472 SPY daily bars from 29 January 1993 to 25 September 2026, 16 morning stars met all four rules. Seven of them closed higher five sessions later, which is 43.8%.
Compare that with every SPY day. Across 8,467 days with a close five sessions later, 4,900 were followed by a higher one: 57.9%. On the 3,306 days that met the same falling condition, 57.0% closed higher five sessions later.
The evening star landed on its base rate. 19 met the rules, and 8 closed lower five sessions later, which is 42.1%. On all SPY days the figure was 41.8%, and on the 5,122 days after a ten-session rise it was 41.2%.
Without the gap rule the counts grow and the results move toward the averages. 124 morning stars qualify, and 69 of them closed higher five sessions later, which is 55.6%, just under the 57.9% for all days. 107 evening stars qualify; of the 106 with five sessions of data, 43 closed lower, or 40.6%.
The honest reading is that neither star separated from an ordinary day on this index. Sixteen and nineteen cases are too few to prove anything either way, and the looser version, with over a hundred cases each, lands within a few points of the base rate. The most recent evening star, completed on 7 January 2025, was followed by a 0.71% gain.
Video demand is thin. In the 24,971-video corpus this site studies, counted once per video id, 3 titles mention a morning or evening star, from 3 channels, at a median of 35 views. Every star found, with its dates and outcome, is in the SPY candle event file. The SPY bars behind every base rate on this page are in the daily bar file.
Why the gap rule matters so much
The gap rule asks for two things at once. The star must open below the first candle’s close and also close below it, so that its whole body sits clear of the first body. On SPY, 108 of the 124 loose-version morning stars failed that test.
That is why the strict pattern appeared only 16 times in 33 years. On SPY the rule filters out almost everything.
Relaxing it is a real trade-off. The loose version is common enough to study, but it is no longer the pattern the textbooks describe, and its results sat close to the base rate. The strict version is closer to the original idea, and too rare to judge.
When it fails
The first failure is the one in the example: the decline was not finished. The morning star marked a pause, and the low came days later. A buyer at the day-three close was underwater within 24 hours.
The second is a third candle that barely qualifies. In the October 2023 case, day three cleared the midpoint by less than 5 cents. A rule applied that tightly will include cases that are really a weak bounce, and a stricter reader would have skipped it.
The third is the drift working against the evening star. SPY closed higher five sessions later on 57.9% of all days, so a bearish three-candle pattern starts behind. On this data the evening star matched the base rate almost exactly.
A fourth is a shifting definition. The size of day one, the size of the star, the gap and the depth of day three are all thresholds, and sources disagree on each. A result that holds under one set of numbers may vanish under the next.
And a fifth is reading it without the trend. The definition here requires a ten-session decline before day one; drop that and the same three candles also turn up in quiet ranges, which this page did not test. Whether a trend reversal is under way is a separate question the candles cannot settle on their own.
Related
Candlestick patterns covers the wider family these stars belong to, and what any single candle can record. The bullish engulfing page shows the two-candle version of the same handover. And trend reversal explains how to tell a turn from a pause, which is the question the morning star is trying to answer.
If you use the morning star, decide up front whether the gap rule is in or out, and count your cases both ways. The strict version almost never appears on an index, and the loose version appears often enough that it is hard to call it rare or special.
— Michael Whitman
This page is educational, not financial advice. Test every idea on your own charts before risking money.