Homing Pigeon Candlestick Pattern: Read the Breakout, Not the Name

The homing pigeon candlestick pattern is a quiet one. No gap, no long shadow, no dramatic engulfing bar. Just a tall red candle in a falling stock, followed by a smaller red candle that sits entirely inside the first one’s body. The textbooks file it under bullish reversals. The people who have actually counted outcomes are far less sure, and that tension is exactly why it is worth understanding before you act on it.

We will walk through how to spot it, what the published test results say, how a screener detects it, and a worked trade on a hypothetical NSE stock where the arithmetic shows why this pattern needs patience more than enthusiasm.

What the homing pigeon candlestick pattern looks like

It takes two sessions. Thomas Bulkowski’s identification guidelines on ThePatternSite lay it out simply:

1. The trend: price has been moving down into the pattern.
2. Day one: a tall black (down) candle.
3. Day two: a small black candle whose body fits inside the body of day one.

If that sounds familiar, it should. It is close to a bullish harami, except that in the harami the small second candle is white. Here both days close lower than they opened. The selling continues, but it shrinks. That shrinking is the whole story the pattern is trying to tell: sellers are still in charge, yet they are pushing with much less force than the day before.

Bulkowski makes a sharp observation about the shape. The rules only talk about bodies, not shadows. So a homing pigeon is often an inside day, but not always, because day two’s wick can poke outside day one’s range while its body stays inside. Traders who think in terms of western inside bars should keep that difference in mind.

How a screener defines it

If you run scans, the precise rule matters more than the picture. The TA-Lib reference for CDLHOMINGPIGEON, the function many charting tools and Python screeners lean on, spells it out:

• Both candles are black: each closes below its open.
• The first body is long, measured against a running average of body sizes.
• The second body is short, again against an average.
• The second body is contained by the first: its open is below the first day’s open and its close is above the first day’s close.

Two details catch people out. First, TA-Lib does not check for a prior downtrend at all. Its own notes say so. A homing pigeon flagged in the middle of a rally is technically a match and practically meaningless, so you need your own trend filter, such as price below a falling 20-day average. Second, the function only ever returns +100 when it finds the pattern. It never outputs a bearish reading. The code assumes the bullish interpretation, even though the testing data, as we will see, does not back that assumption well.

The pattern says the sellers are tiring. It does not say who wins next. Only the breakout answers that.

What the test results actually say

This is where the homing pigeon gets interesting. Bulkowski’s figures, from the same ThePatternSite page, compare theory with outcomes:

• Theory: bullish reversal.
• Tested behaviour: it acted as a bearish continuation 56% of the time. In other words, in 56% of his samples price broke out downward, in the direction of the existing fall. He calls that “near random”.
• Frequency rank: 34 out of 103 candlestick patterns, where 1 is the most common. So it turns up fairly often.
• Overall performance rank: 21 out of 103, where 1 is the best. That is a respectable ranking for what happens after the breakout.
• Best percentage meeting its price target: 68%, in a bull market with an upward breakout.
• Best average move 10 days after the breakout: a rise of 4.76%, in a bear market with an upward breakout. That ranks 33rd of 103. Bulkowski adds that a good move would be 6% or more.

Read those numbers together and a picture forms. You cannot use the homing pigeon to predict direction. A 56/44 split is barely better than a coin toss. But once price does break out, the move that follows has been decent by candlestick standards. So the pattern is better treated as a marker, a two-day range to watch, than as a buy signal.

His published tips point the same way. Homing pigeons that appear within a third of the yearly low performed best, and he suggests selecting tall candles. Both make intuitive sense. A stock that has already fallen hard toward its 52-week low has more room for sellers to be exhausted, and a tall first candle makes the shrinkage on day two more meaningful.

Defining the breakout before you trade

Bulkowski’s own example on the page shows a downward breakout, confirmed when price closed below the lowest low of the two-candle pattern. That decline lasted just a day before a doji marked a turn and price climbed in a V shape. It is a useful reminder that even the “correct” signal can be short-lived.

For our own trading, we mark two levels as soon as the second candle closes:

• Pattern high: the highest high of the two days, which in practice is usually day one’s high.
• Pattern low: the lowest low of the two days.

A daily close above the pattern high is our upward trigger. A daily close below the pattern low is the downward one. Until either happens, we do nothing. That single rule removes most of the guesswork the pattern’s name invites.

A worked example on a hypothetical NSE stock

Let’s take XYZ Ltd, a made-up stock that has slid for three weeks and sits close to its 52-week low. Every price below is hypothetical, chosen only to make the maths clear.

Day 1: opens at ₹520, trades as high as ₹523 and as low as ₹485, closes at ₹488. A body of ₹32, tall compared with recent sessions.

Day 2: opens at ₹497, high ₹499, low ₹489, closes at ₹491. Black again, with a body of just ₹6.

Check the screener rule. Day two opened at ₹497, below day one’s open of ₹520. It closed at ₹491, above day one’s close of ₹488. Both bodies are black. The small body sits inside the big one. That is a homing pigeon.

Our levels: pattern high ₹523, pattern low ₹485.

Scenario A: the upward breakout

Four sessions later XYZ closes at ₹525, above the pattern high. We buy at that close and place the stop just under the pattern low, at ₹484.

• Risk per share: ₹525 − ₹484 = ₹41.
• Trading capital: ₹5,00,000. We risk 1%, which is ₹5,000.
• Position size: ₹5,000 ÷ ₹41 = 121.95, so we round down to 121 shares.
• Money deployed: 121 × ₹525 = ₹63,525.

Now look at the reward. If we only expected a move in line with the 4.76% best-case average from Bulkowski’s data, the target would be about ₹550 (₹525 × 1.0476 = ₹549.99). That is a gain of roughly ₹25 a share against a risk of ₹41. The trade loses money in expectation unless the win rate is very high, and we already know it is not.

This is the trap with tall-candle patterns. Bulkowski says tall candles perform better, but a tall pattern also means a wide stop. The honest options are:

1. Demand a bigger target. At two times risk, that is ₹525 + ₹82 = ₹607. Only take the trade if the chart has room to get there, for instance if there is no major resistance between ₹525 and ₹607.
2. Use a tighter, structure-based stop, for example below a higher low that forms during the breakout. This reduces risk per share but raises the chance of being stopped out by ordinary noise.
3. Skip it. Plenty of homing pigeons are not worth trading, and recognising that is a skill in itself.

Scenario B: the downward breakout

Suppose instead that XYZ closes at ₹483, below the pattern low. That is the bearish continuation outcome Bulkowski found in 56% of cases. For a cash-market investor who already holds the stock, this is a clear exit signal. The market has tested whether sellers were running out of steam and the answer is no. Holding on in the hope that the “bullish” pattern eventually works is the costly mistake here.

Traders who can short through stock futures may treat it as a continuation entry, but the same risk arithmetic applies, and futures carry lot sizes and margin requirements that magnify mistakes.

A tall pattern gives a cleaner signal and a wider stop at the same time. You have to decide whether the target is big enough to pay for that stop.

Homing pigeon vs bullish harami vs matching low

Three two-candle patterns sit close together, and mixing them up is common.

• Homing pigeon: both candles black, small body inside the big body.
• Bullish harami: a tall black candle followed by a small white candle inside it. The second day actually closes up, which is a slightly stronger hint of buying.
• Matching low: two black candles that close at or near the same price. The focus is on a shared close as support, not on a shrinking body.

TA-Lib’s own page lists the harami and matching low as related functions, which tells you how close the definitions are. If your scanner throws up all three on the same stock in the same week, treat it as one story: a decline that is losing pace near a price level. Then wait for the breakout.

Mistakes we see most often

Ignoring the trend. Because the common screener definition does not check for a downtrend, many “homing pigeons” appear in sideways or rising stocks. Without a prior fall there is no selling to exhaust, so the pattern loses its meaning.

Buying on day two’s close. The second candle is still a down day. Buying it means betting on a reversal that the data shows is the less likely outcome. Wait for a close above the pattern high.

Forgetting costs and gaps. Backtested statistics assume clean entries and exits. Real NSE trades pay brokerage, STT, exchange fees and GST, and a results announcement or overnight news can gap a stock straight through a stop. A ₹41 planned risk can become ₹60 at the open. We avoid holding candlestick setups across a scheduled results date.

Trading it in isolation. A homing pigeon at a level that has held before, near the yearly low, in a stock whose broader trend is turning, is worth watching. One in the middle of nowhere is just two red candles.

A starter checklist

• Confirm a clear decline into the pattern, ideally with price in the lower third of its yearly range.
• Check that both candles are black and day two’s body sits fully inside day one’s body.
• Prefer a tall first candle, but size the trade from the full stop distance.
• Mark the pattern high and low and act only on a daily close beyond one of them.
• Take long trades only when there is room for at least twice the risk; treat a close below the low as an exit for holders.

For the wider toolkit these patterns fit into, see our guide to technical analysis for Indian stocks. The homing pigeon will rarely be the reason we buy a stock. What it does well is flag the moment a fall starts losing energy, and that is a moment worth paying attention to, as long as we let the breakout make the decision.

Disclaimer: This article is for educational purposes only and is not investment advice. XYZ Ltd and all prices in the examples are hypothetical. Trading in equities and derivatives involves risk, including loss of capital; consult a SEBI-registered investment adviser before making trading decisions.