Concealing baby swallow: a rare four-candle bottom explained

The concealing baby swallow is one of the strangest entries in the candlestick catalog: a bullish reversal pattern made of four black candles. There is no white candle anywhere in it. The bullish case rests entirely on how the third and fourth candles behave after two sessions of uninterrupted selling.

The name comes from the last two candles. The third is a small "baby" that gaps lower but pokes an upper shadow back up into the previous body. The fourth is a larger black candle that "swallows" the baby whole, high and low included, so the small candle is concealed inside it. The Japanese view is that a market able to trade that high intraday, after two flat-out selling sessions, is losing its downside momentum.

You may never see one. In two years of daily bars on 60 heavily traded US stocks and ETFs, the pattern did not appear once. This page explains the shape, the exact rules Finabase's detector applies, why it is so rare, and how to think about a pattern you cannot test.

What the pattern looks like

The four candles, in order:

1. A black marubozu. The session opens at its high and closes at its low, or very nearly. Sellers controlled it from start to finish.
2. A second black marubozu, lower again. Two sessions in a row with no intraday recovery at all.
3. A small black candle that gaps down. Its body opens below the second body, but during the session price rallies far enough to leave an upper shadow that reaches into the second candle's body. This is the first sign of buying.
4. A black candle that engulfs the third, shadows and all: its high is above the third candle's high and its low is below the third candle's low.

The bullish argument is subtle. The fourth candle still closes lower, so sellers appear to win again, but the market spent the session trading above the prior high before falling. Traditional sources treat it as selling exhaustion: shorts are pressing a market that keeps bouncing.

The exact rules Finabase uses

Finabase scans for the pattern with TA-Lib's CDLCONCEALBABYSWALL:

How that compares with the textbook description:

• The engulfing test is about the range, not the body. TA-Lib checks only that the fourth candle's high and low surround the third candle's high and low. It does not require the fourth body to cover the third body, though in practice a black candle that opens near the top of that range usually does.
• "Marubozu" is relative. A shadow counts as very short if it is under one tenth of the stock's own average daily range, so a volatile stock is allowed slightly bigger shadows than a quiet one.
• No downtrend check. The source comment says the pattern is significant in a downtrend and that the function does not check for one. Two black marubozu in a row usually imply some decline anyway.

Why it almost never appears

Each of the four conditions is uncommon on its own; requiring them in sequence makes the pattern vanishingly rare.

Two marubozu back to back. On liquid US stocks almost every session leaves at least a small shadow at one end, because the opening and closing auctions rarely land on the day's extremes. A strict black marubozu is unusual; two consecutive ones are much rarer.

A body gap down after them. The third session must open below the second close (itself the day's low) and stay below it with its body, yet still trade back up into the second body. That is a gap that gets tested intraday but holds by the close.

Then a full-range engulf. The fourth candle needs a wider range than the third, in both directions, while still closing black.

Our study ran all 61 TA-Lib patterns that Finabase supports over the same 29,760 bars. Eleven of them never fired at the settings the app uses. Seven of those are patterns Finabase runs with an unusual penetration setting, which on its own rules most of them out. The concealing baby swallow is one of the four with no such excuse, alongside three stars in the south and the two kicking patterns: they are simply that rare. Shadowless candles are more common in thin or limit-bound markets than in continuously traded large-cap stocks, which may be part of the reason.

What the absence of data means

The usual sections of a Finabase pattern page, a real example and a table of what happened next, are empty here because there was nothing to measure.

That is useful information in itself. A pattern that appears once a decade on a large cap cannot carry much weight in anyone's process, and a scanner hit on it is more likely to reflect unusual data than a reliable market tell. If one does turn up, check the underlying bars for errors, a halt or a corporate action before reading anything into it.

How to judge one if it appears

Without data, the reasoning has to come from the candles themselves:

• Was there a real decline? Two marubozu in a row are themselves a sign of heavy selling. Check that the stock had already fallen meaningfully before them.
• How far did the third candle's shadow reach? The deeper it extends into the second body, the stronger the intraday buying it records.
• Where did the fourth candle close? A close near its low, below the third candle's low, is selling continuing. A close well off its low weakens the bearish case and strengthens the exhaustion story.
• Look for confirmation. A following session that closes above the fourth candle's high is the simplest evidence that buyers have taken over. Momentum turning up on the Relative Strength Index (RSI): how to read and use it from oversold is supporting evidence.

Candles it can be confused with

The ladder bottom is the closest relative: it also starts with a run of black candles and looks for exhaustion, but it ends with an actual white candle opening higher, which is why it fires dozens of times where the concealing baby swallow fires never.

How Finabase shows the pattern

• Chart markers: add Concealing Baby Swallow to a chart and Finabase will mark the fourth candle of any match on the daily bars. On the large caps we tested the study draws nothing, and an empty overlay is the expected result.
• Scanner: because TA-Lib only ever returns +100, any hit would appear on the bullish candlestick list and never on the bearish one.
• Signals tab: the Candlestick gauge on a symbol's Signals tab, for example AAPL's, tallies one vote per pattern from the latest non-zero detection in the past eight bars. With no detections, this pattern's vote is Neutral on essentially every stock, every day. It still counts toward the neutral total in the gauge formula, (buy − sell) / (buy + sell + min(neutral, max(buy, sell))), but that term is capped, so a pattern that never fires cannot dilute strong readings from the others.

Frequently asked questions

Why is a pattern of four black candles considered bullish?

Because the last two candles show sellers losing control. The third candle gaps lower but rallies into the previous body, and the fourth trades above that rally before closing. Traditional analysts read it as a selling climax that is running out of energy.

Has the concealing baby swallow ever appeared on major US stocks?

Not in our sample: zero times in 29,760 daily bars of 60 large stocks and ETFs over two years. It is possible on other stocks or timeframes, but it is exceptionally rare on liquid large caps.

Does Finabase ever flag a bearish version?

No. TA-Lib defines only the bullish form and always outputs +100, so any detection would count as a Buy vote and appear on the bullish scanner list.

Should I wait for confirmation?

Yes. With no reliable statistics, a close above the fourth candle's high, ideally on strong volume, is the minimum evidence that the reversal is real.

Summary

The concealing baby swallow is a four-candle bullish reversal made only of black candles: two shadowless sell-offs, a gapped-down baby that rallies into the prior body, and a larger black candle that swallows it. TA-Lib's version follows that description closely, always scores it +100 and does not check the trend. On 60 large US stocks and ETFs over two years it never appeared, so Finabase cannot tell you how well it works, and neither can most published sources with any confidence. Learn to recognize it, treat a sighting as a curiosity to verify, and rely on confirmation if you act on it.