Stick sandwich candlestick: matching closes as hidden support

A stick sandwich is two black candles that close at the same price with a white candle squeezed between them. Seen from a distance it looks like a sandwich: dark slices on the outside, a light filling in the middle. The bullish reading comes from the matching closes. Sellers pushed the stock down to a price, buyers lifted it, sellers pushed it back down, and the second push stopped exactly where the first one did.

Japanese traders interpreted that repeated close as a hidden support level, a price at which demand reliably appeared. Unlike the Hammer candlestick pattern: meaning, rules and reliability, whose evidence is a single session's long shadow, the stick sandwich gets its evidence from two separate sessions agreeing on the same floor.

This page covers the formation, the tolerance TA-Lib allows for "the same close", the odd situation where Finabase can label a single candle both bullish and bearish, and what 39 stick sandwiches on large US stocks did next. Short version: they underperformed an average day.

Anatomy of a stick sandwich

1. First black candle. The decline continues with a down session.
2. White candle. The next session opens higher and trades above the first candle's close all day, then closes higher. It looks like the start of a rebound.
3. Second black candle. Sellers return and drive the stock back down, but the close lands at the same price as the first black close.

The textbook pattern appears in a downtrend, and the two black closes mark the bottom of the move. The second black candle is the one that gives traders pause: on its own it looks bearish, wiping out the white day. The pattern's message is that it could not do more than that.

The exact rules Finabase uses

Finabase uses TA-Lib's CDLSTICKSANDWICH on daily bars. The test is short:

Things the function does not check:

• No body sizes. Any of the three candles can be large or small.
• No test that the third candle reaches the white candle's open or low. It only has to end at the matching price.
• No downtrend. In our sample just 41.7% of the measurable stick sandwiches formed after a ten-session decline, so most appeared in flat or rising markets where there is no "bottom" to confirm.

The tolerance scales with volatility. For a stock that has been moving $20 a day, the two closes may differ by about $1 and still count as equal. For a $2-a-day stock, the limit is around 10 cents.

What the matching closes say

Closing prices carry extra weight in candlestick analysis because they are where the day's argument is settled. When two sessions with a rally in between both settle at the same number, it suggests that something specific happens at that price: a large buyer working an order, a cluster of stop orders already flushed, or simply a level the market has decided is fair.

The white candle in the middle matters because it shows the price was not just stuck. Buyers were able to lift the stock clear of the first close for a whole session. When the second black candle retraced all of that, it stopped at the earlier floor rather than breaking through it. That is the support argument.

The weakness of the argument is equally clear. The third candle is a strong down day. If the support fails the next morning, the pattern has done nothing but describe a sell-off with a pause in it.

Confirmation

• A close above the second black candle's high, which is the rule our statistics use for confirmation. Only 8 of the 39 detections met it.
• A close above the white candle's high, a stricter version that shows buyers regaining the ground lost on day three.
• Holding the matching close on a retest in the following days.
• Momentum support, for instance an Relative Strength Index (RSI): how to read and use it reading near oversold that turns up.

Two real stick sandwiches

Mastercard, April 16, 2025. MA had fallen 6.2% in ten sessions during the April 2025 market sell-off. On April 14 it closed at $512.46 on a black candle. On April 15 it never traded below $513.38 and closed at $517.38. On April 16 it opened at $516.05 and closed at $513.43, within 97 cents of the April 14 close, inside TA-Lib's tolerance for a stock then moving roughly $20 a day. Two sessions later the stock dipped once more, to an intraday low of $503.54, and closed at $509.32, just under the matching closes; then it turned up, and ten sessions after the pattern Mastercard was 6.5% higher. The floor bent before it held, which is typical of real support.

Meta, February 27, 2025. META had dropped 9.3% in ten sessions. It closed at $657.50 on February 25, rallied to $673.70 on February 26 without trading below $658.00, and then on February 27 opened at $682.45 and fell to $658.24, 74 cents from the first close. The same third candle was a $24 black body that swallowed the prior white body: TA-Lib's CDLENGULFING scored it −100 (bearish) on the same bar where CDLSTICKSANDWICH scored +100. The bearish reading won. Ten sessions later Meta was 10.3% lower. When a single candle produces opposite signals, the larger, more decisive candle usually deserves the attention.

What the data says

We ran CDLSTICKSANDWICH over daily bars for 60 large US stocks and ETFs from October 30, 2024 to October 29, 2026. It fired 39 times on 30 symbols; 36 had a full ten sessions of follow-up.

• Below the baseline. Fewer than half of the stick sandwiches were higher a week later, against 53.1% of random days, and the average five-session return was negative.
• Partial recovery by ten sessions. The average after ten sessions was positive but about half the baseline.
• Confirmation did not rescue it. The 8 confirmed cases were higher only 3 times in 8 after the confirmation close.

Stick sandwich versus its look-alikes

The stick sandwich is essentially a matching low with a white candle inserted. The insertion is what makes the support more convincing in theory, because the market left the level and came back to it.

An educational outline

This shows how traders usually approach the pattern. It isn't a recommendation to trade.

1. Require a decline first. Most of our detections lacked one; those are not the pattern the books describe.
2. Check for conflicts. If the third candle also qualifies as a bearish engulfing or a long black marubozu, the two readings cancel; the chart markers show both.
3. Let price confirm. A close above the third candle's high is the first evidence that the support is being bought.
4. Use the matching close as the line. A close clearly below it means the support failed.

How Finabase shows the stick sandwich

• Chart markers. The Stick Sandwich study marks the second black candle of each pattern on daily charts. Add the Engulfing study alongside it to spot bars where the two disagree.
• Scanner. TA-Lib returns +100 for every stick sandwich, so a stock completing one appears on the bullish list, even though its latest candle is a black one.
• Signals tab. On a symbol page such as AAPL's signals, the Candlestick gauge counts the most recent reading of each of its 61 patterns in the last eight bars. A stick sandwich adds a Buy vote, and a bearish engulfing on the same bar adds a Sell vote, so they offset. The score is (buy − sell) / (buy + sell + min(neutral, max(buy, sell))), with Strong Sell below −0.6, Sell below −0.2, Neutral below 0.2, Buy below 0.6 and Strong Buy above.

Frequently asked questions

Do the two closes have to be exactly equal?

No. TA-Lib allows a difference of up to 5% of the average daily range of the five candles before the pattern. On a volatile stock that can be close to a dollar.

Is the stick sandwich bullish or bearish?

It is traditionally bullish and TA-Lib always scores it +100. Its last candle is bearish, though, and in our sample the pattern underperformed an average day.

Why does the scanner show a stock with a big black candle as bullish?

Because the scanner sorts by the sign of TA-Lib's output, and the stick sandwich is always positive. Check the chart for a bearish pattern on the same bar.

Does volume matter?

TA-Lib ignores volume. Many practitioners look for lighter volume on the second black candle than on the first, which would suggest fewer sellers at the support.

Summary

The stick sandwich is a support idea expressed in candles: two black sessions closing at the same price around a white one. Mastercard's April 2025 example shows it working at a real low. Meta's February 2025 example shows the risk, a large black candle that was simultaneously a stick sandwich and a bearish engulfing, followed by further losses. In our two-year test the pattern lagged an average day, so treat it as a level to watch, not a signal, and require both a prior decline and a confirming close.