Every article about the evening star shows you the same diagram: a long green candle, a small indecisive one, then a long red candle that gives the rally back. Almost none of them tell you what happens next, on real prices, often enough to matter.
So we ran it. Across 22 US stocks and sector ETFs, over the window the free backtester covers, the pattern fired 74 times. Buying the close of the third candle and holding ten sessions, with commission and slippage on, 45 of those 74 trades finished green: a 60.8% win rate.
That is the opposite of what the pattern is sold as. The evening star is supposed to mark a top, and buying into one was profitable more often than not.
Three things that number does not say, and each matters more than the number:
- It is a small sample. 74 signals across 22 names in four and a half years is about three per stock. A 60.8% win rate on 74 trades is well inside what chance produces; it is not evidence the pattern works upside down, it is evidence it does not do what the diagram claims.
- A win rate is not a return. The same runs left the median name up 2.2% over the whole period. Which is a small number for a reason: see the next point.
- This rule is barely in the market. Holding ten days after a signal that fires three times per stock per four years means roughly 3% time invested. Everything above describes what those few days did, and nothing about it is comparable to a strategy that stays invested, so we have not put a buy-and-hold figure beside it.
The rest of this page is what the pattern is, where the context genuinely changes it, and how to check any of the above yourself.
What is an evening star
An evening star is a three-candle formation that appears after an uptrend:
- A long green candle. Buyers still firmly in control.
- A small-bodied candle, green or red, that gaps or stalls near the high. This is the hesitation, and it is the part of the pattern doing the work.
- A long red candle that closes well into the body of the first one, giving back most of the advance.
The story is a rally running out of buyers. The middle candle is where conviction thins; the third is where sellers take the session.
A variant called the evening doji star has a doji as the middle candle, where the open and close finish nearly equal. Our detector treats the two as the same pattern, because they are the same story with a stricter middle bar.
Why the middle candle is the whole pattern
Traders who dismiss candlesticks usually do so because the shapes are described without their mechanics. The evening star has one:
A long green candle means the day's buyers absorbed everything offered. A small candle immediately after means that on the following day, at those prices, they did not. Nothing has broken yet, but the bid that was there yesterday is not there today. The third candle is the confirmation, and it is also the reason the pattern is late: by the time it completes, the move it is warning about has already started.
That lateness is visible in the test. A pattern that only confirms after a full down day cannot get you out near the top, which is what people expect of it.
Where the context genuinely changes it
Our own data pushed back on the standard advice here, so it is worth being specific.
An evening star in a stock still near the top of the momentum rankings fails more often than the diagram suggests. Strong names get bought on dips, and a three-candle wobble inside a powerful trend is usually just a wobble. Where the pattern earns more respect is in a name whose relative strength has already been slipping for weeks, where it marks the point the last buyers give up rather than a pause.
The other conditions worth checking are ordinary and real:
- At a level price has already been rejected from
- Under a moving average that has turned down
- After an extended run rather than three green days
- On heavier volume than the days around it
None of these are in the backtest above. The test takes every signal the detector finds, which is deliberately the unflattering version: it is what the pattern does without a human filtering the good ones.
Test it yourself, on any stock
The backtester carries this pattern as a rule, so you do not have to take the figures above on trust. In the free backtester, open Chart setup, add Candlestick pattern, and set it to Evening star. Choose a ticker, set a hold period, and run it.
The test worth running once you are in there is the filtered one. Stack a moving-average or relative-strength condition on top of the pattern and see whether screening for the setups above actually improves the win rate, or only reduces the number of trades. That is what separates a rule from a superstition, and it is why we would rather hand you the tool than a verdict.
Its bullish mirror at the bottom of a decline is the morning star, which we tested the same way, and there is a candlestick pattern cheat sheet if you want the rest of the family in one place.
The scanning side
If you want to watch for these live rather than test them historically, this is a charting problem rather than a backtesting one. TC2000 can scan for candlestick formations and alert on them, which is what most people actually want from a pattern like this. The evening star is in its scan library, so no formula writing is involved.
Finviz has candlestick patterns in its free screener too, which is the cheaper way to see today's list. It has no alerting and no backtest, so it answers "which stocks printed one" and not "did it mean anything".
Frequently asked questions
Is the evening star a reliable reversal signal? Not on our test. Buying the completion of 74 evening stars across 22 US stocks and ETFs and holding ten days won 60.8% of the time, which is the opposite of what a bearish reversal should do. The sample is small enough that this is not proof the pattern works backwards; it is enough to say it does not reliably mark a top.
What is the difference between an evening star and an evening doji star? Only the middle candle. An evening doji star has a doji there, meaning the open and close finished nearly equal, which makes the hesitation more explicit. The story and the implication are the same, and our backtester treats both as the same pattern.
Is the evening star bullish or bearish? Bearish. It appears after an uptrend and describes buyers failing to hold a high. The morning star is its mirror image at the bottom of a down move.
Can you trade the evening star on its own? You can, but the data above is a reason not to. The pattern completes only after a full down candle, so it confirms late by construction, and on our sample the trade it implies went the wrong way more often than not. It is better used to question a position you already hold than to originate a trade.
How many candles is an evening star? Three: a long green candle, a small-bodied candle at the high, and a long red candle closing well into the first one's body.
Before you trade this pattern
The figures on this page come from one run over 22 large-cap US stocks and sector ETFs, in a window that was mostly a bull market, and the rule they describe was invested about 3% of the time. The script that produced them is in the repository (backend/scripts/research/candle_study.py) so the run can be repeated rather than trusted. The tool will show you today's numbers on any ticker you choose, and they will not match these exactly. Past prices are not a forecast, and nothing here is investment advice.