Three Outside Down — Bearish Engulfing With Confirmation

In this series on confirmed patterns we keep saying the third candle raises the win rate: a bullish harami goes from 53% to 65%, a bullish engulfing from 63% to 75%. Three Outside Down breaks that tidy narrative. A plain bearish engulfing reverses the trend 79% of the time. A bearish engulfing with confirmation — 69%. Yes, you read that right: adding a confirming candle lowers the reversal rate. That's the best lesson this pattern has to offer — and exactly why it's worth knowing.

What the Pattern Looks Like

Three Outside Down is a three-candle bearish reversal pattern — a bearish engulfing plus a confirming candle, per Morris's definition. The conditions:

• Trend before the pattern: uptrend. No uptrend, no reversal — a setup inside an ongoing decline is continuation, not this pattern.
• Candle 1: bullish (white). Demand is still in control.
• Candle 2: bearish (black), opening above candle 1's body and closing below it. A full body engulfing — supply wipes out the entire prior session in one move and then some.
• Candle 3: closes lower than candle 2. Confirmation — sellers keep control.

The story: the market is rising, then supply suddenly seizes the session so brutally it swallows the entire previous move, and the next day it presses further. Two consecutive sessions of supply control after a period of gains is a change in market character, not a one-off event.

A note for crypto: the strict definition requires candle 2 to open above the prior candle's body — on a 24/7 market, an open always equals the previous close, so this condition literally never happens. In practice, on BTC/ETH, traders play body-engulfs-body (candle 2's close below candle 1's open). Reasonable, but the statistics were measured on the strict version — one more reason to treat the percentages as directional only.

What the Numbers Show

Results from Bulkowski's tests (~4.7 million daily candles, US stocks):

• Bearish reversal: 69% of cases. A very good result for a bearish pattern — clearly above the threshold of randomness.
• Overall performance rank: 39/103. Upper half of the table. As the study's author dryly notes, 38 patterns do better, "but who's counting."
• Frequency: 21/103. A common setup — easy to find, both in history and live.
• Measure-rule target: best 55% (bull market, upside breakout). Nearly half of all cases don't deliver the pattern's height.
• Best average 10-day move: +6.30% — a gain, following an upside breakout in a bear market. And the best 10-day rank (13/103) came after an upside breakout in a bull market.

Let's stop on that last point, because it's the crux of it. Bulkowski writes plainly: the decline 10 days after a downside breakout is pathetic — a few percent on average — while upside breakouts produced moves two to three times larger. Short-term, the pattern "works" best precisely when it fails as a bearish signal. The worst measured variant? A downside breakout in a bull market — exactly what everyone tries to play when they short a rally after a nice confirmed engulfing.

And the number we started with: a plain bearish engulfing — 79% reversals, with confirmation — 69%. Where does the difference come from? Partly from how the study is structured (the more bearish candles in the pattern itself, the more of the move has already happened, and the closer you are to "geometric confirmation"), partly from case selection. The practical takeaway isn't "confirmation is bad" — it's that confirmation isn't free and doesn't always raise the statistic. Measure it, don't assume it.

One more family comparison for the record: Three Inside Down, i.e. a harami with confirmation, reverses at 60% with rank 56 — Three Outside Down beats it on both counts (69%, rank 39). If you have to pick a bearish three-candle setup, an engulfing with confirmation has better numbers behind it than a harami with confirmation.

As always: data from US stocks, D1, laboratory conditions. Crypto — different volatility, different market. Logic, yes; percentages, no.

How to Trade It

Scenario 1 — the top of a correction within a downtrend (the data-recommended setup). The ideal setup per Bulkowski: an overarching downtrend, an upward bounce, Three Outside Down at the top of that bounce, a downside breakout returning to the existing trend. Enter on a close below the pattern's low, stop above the pattern's high, with a horizon longer than a week — short-term declines after this pattern are too meager to cover the costs.

Scenario 2 — a defensive signal for a long. A confirmed engulfing after a longer rally is a solid reason to trim a long position or tighten the stop — 69% reversals is plenty for a defensive decision, even if it's not enough for an aggressive short into a bull market.

Scenario 3 — the pattern breaking. A close above the pattern's high invalidates the signal — and per the data, the best 10-day moves followed exactly these kinds of breakouts. A broken Three Outside Down within an uptrend is an argument for going long, not for hunting a "second top."

How NOT to play it. Don't short a strong bull market on this setup alone — a downside breakout in a bull market is the worst measured variant. Don't expect a quick profit from the short — a large chunk of the decline happens within the three candles of the pattern itself, and the follow-through tends to be slow. And don't assume three candles are automatically better than two: if your edge rests on the plain engulfing, adding conditions can dilute it.

Stop loss and target. For a short: stop above the pattern's high (or, when entering at the top of a correction, above the correction's high), target at the nearest significant support with partial profit-taking. The pattern can span a wide range (engulfing plus confirmation), so if the stop ends up too far away, wait for a bounce toward the lower edge of the setup or skip the trade.

Myth vs. Measurement

Myth: "Confirmation always increases a pattern's win rate." Measurement: a bearish engulfing without confirmation — 79% reversals, with confirmation — 69%. In this case, the third candle doesn't improve the statistic. Every extra filter needs to be measured, not assumed.

Myth: "After a confirmed bearish engulfing, the decline comes fast." Measurement: 10-day moves after a downside breakout were meager — a few percent — while upside breakouts produced moves 2-3 times larger. This pattern reverses the trend often, but slowly.

Myth: "69% success rate means short in any conditions." Measurement: the worst variant is a downside breakout in a bull market. Same candle pattern, different market regime — different signal value. Context isn't an add-on to a pattern; it's half of it.

A Sample BTC Scenario

BTC has been sliding in a D1 downtrend for a month. A bounce arrives: four green candles push price up under a broken support level that now acts as resistance. At the top of the bounce, a green candle, followed by a large red candle swallowing its entire body, and on the third day another lower close. Three Outside Down exactly where the data says to look for it — at the top of a correction within a downtrend.

The plan: short on a D1 close below the pattern's low, stop above the bounce's high, target at the last trend low — with partial profit-taking and no expectation of fireworks in the first week, since this pattern declines slowly. Alternative scenario: a close above the pattern's high kills the short and — if the broader market is improving — opens a conversation about the long side, since a broken Three Outside Down statistically pulled price higher the hardest.

Quick checklist:

• Is there a clear upward move before the pattern (ideally: a correction within a downtrend)?
• Does candle 2's body fully engulf candle 1's body?
• Does candle 3 close lower than candle 2?
• Are you avoiding shorting a strong bull market on this pattern?
• Is your horizon longer than 10 days, with partial profit-taking planned?
• Do you have a plan ready for the pattern breaking (a close above the high) before entering?

Remember Three Outside Down for one reason: it's the pattern that breaks the confirmation dogma. Sometimes a third candle raises the win rate, sometimes it lowers it — and the only way to know is to measure it. Trading patterns on faith means trading someone else's legend. Checking the numbers at least means knowing the price of your own decisions.

FAQ

Rafał — Krypto Bez Ściemy

Trader and founder of Strefa Tradingu. He’s been breaking crypto down on YouTube for years — no hype, no signals, with a focus on market structure and risk management.

kryptobezsciemy.plInstagramYouTubeKlub Zero