Pattern Detail

Bearish Three Inside Down

Three-candle bearish reversal after a rally: a long up candle, a small down candle held inside it, then a third down candle that confirms.

A real Bearish Three Inside Down on NQ daily bars, Oct 18, 2018. Price then followed through 3.7% over the next 5 bars. The bright candles are the pattern; the dimmed bars are surrounding context.
A real Bearish Three Inside Down on NQ daily bars, Oct 18, 2018. Price then followed through 3.7% over the next 5 bars. The bright candles are the pattern; the dimmed bars are surrounding context.
i

How to read this

Everything here is in R, the setup's own risk. 1R is the distance from the entry (the pattern's closing price) to where it would be proven wrong — its highest high over the 3 bars that form it. So "offered 2R" means price ran twice that distance in your favor at some point before the stop. It does not assume you took profit there: a target is a strategy choice.

Room offered (≥ 1R)

46.3%

Reliable

Offered at least 1× its risk before the stop, vs 41.5% for a random short entry (+4.8 pts).

Move size vs normal

1.06×

Realized range over the next 20 bars vs a random bar. About normal.

Typical room (20-bar)

1.20R

Average run in favor (capped at 3R), vs 1.07R for a random short entry.

Summary

Offered at least 1R of room 46.3% of the time vs 41.5% for a random short entry — a 4.8-point gap, wider than the ±0.9-point margin of error from chance, and it holds across the sample. A real, if modest, tendency to offer more room than the market alone.

Room offered, this setup vs a random short entry

Outcome This setup Random entry Edge
Offered ≥ 1R 46.3% 41.5% +4.8
Offered ≥ 2R 27.2% 27.0% +0.2
Offered ≥ 3R 16.8% 18.8% -2.0
Stopped < 1R 49.7% 55.6% -5.9
Went sideways 4.0% 2.9% +1.1

10,512 occurrences · 1,162,047 random-entry controls · 20-bar horizon

A three inside down is a harami that gets confirmed. A long up candle ends the rise, then a small candle sits entirely inside that body, the buyers losing their grip. The third candle is a down candle that closes lower still, sealing the turn. The first two candles raise the warning, and the third one acts on it.

Three inside down, the confirmed bearish harami, is documented in Steve Nison’s Japanese Candlestick Charting Techniques (1991) and the candlestick literature that followed.

How to spot it

  • The market is rising into the pattern.
  • The first candle is a long up (green) candle that fits the advance.
  • The second candle is small and sits entirely inside the first candle’s body, a harami.
  • The third candle is a down (red) candle.
  • The third candle closes lower than the second, confirming the reversal.

The dashed box on the chart above marks the three candles on a real occurrence, with the advance before and the move after.

The psychology

The long up candle is the rally still pressing, buyers closing the market near its highs with the trend behind them. Then the small candle forms entirely inside that body, and the advance simply stops. Price holds in a tight range and makes no new high. That small candle is the buyers losing their grip, no longer able to push, while sellers quietly meet the bids.

The third candle turns the hint into a decision. Price closes lower than the small candle, which says sellers are no longer just holding the line, they are taking ground. The follow-through is the point: a harami alone only shows the buying paused, but the lower close shows the other side has stepped in to drive price down. Control has passed from buyers to sellers across the three bars.

Whether that confirmed turn carries any distance is what the numbers below check.

Does it actually work?

A pattern is a setup, not a trade, so the honest question is not “did it win” but “how much room did it tend to offer before it was proven wrong.” The tabs below answer that across five futures markets (Nasdaq, S&P 500, gold, crude oil, natural gas) and seven timeframes from one minute to one day.

For each occurrence we measure the room the move offered in units of the pattern’s own risk, then set it against what a random entry on the same market would have done. When the pattern offers more room more often than chance, that shows up as a real edge. When it does not, the page says so plainly.

Read it with the sample size in view. On the faster timeframes a pattern can fire thousands of times, enough to trust. On the daily chart it is far rarer, so treat those numbers as a hint rather than a verdict. Thin samples are flagged for you on the page.

How we measured it

  • Entry is the close of the final candle of the pattern.
  • One unit of risk, 1R, is the distance from that close up to the pattern’s invalidation point: the highest high of the three candles that form it. If price trades through there, the setup is wrong.
  • We then follow the next 20 bars and record how far price ran in your favor, in multiples of that risk, before the stop was hit.
  • Every figure is set against a random entry on the same market and timeframe, so the market’s own drift is accounted for.
  • No profit target and no position sizing. Where you take profit is a strategy choice; this measures only the room the pattern tends to give.

What this page does not cover

  • Volume on the pattern’s candles.
  • Whether the pattern forms at a meaningful resistance level.
  • Pairing it with a trend filter or a confirming signal.
  • A profit target or position sizing. We use the pattern’s own invalidation point as the stop to define risk, but where you take profit, and how much you put on, are strategy decisions this page leaves to you.

Sample Bearish Three Inside Down Firings (20)

Based on data through Apr 29, 2026

Time Risk (pts) Room offered Result
Apr 29, 2026, 6:20 AM CDT 17.25 0.30R Stopped
Apr 28, 2026, 2:15 PM CDT 39.75 0.94R Flat
Apr 27, 2026, 3:45 PM CDT 7.75 1.87R Ran ≥1R
Apr 24, 2026, 3:00 PM CDT 14.75 3.00R Ran ≥1R
Apr 24, 2026, 1:00 PM CDT 37.5 1.10R Ran ≥1R
Apr 24, 2026, 11:40 AM CDT 38 0.67R Stopped
Apr 23, 2026, 6:40 PM CDT 6.75 0.00R Stopped
Apr 23, 2026, 3:10 PM CDT 38.75 0.79R Stopped
Apr 23, 2026, 1:40 PM CDT 34.75 3.00R Ran ≥1R
Apr 23, 2026, 5:40 AM CDT 27.5 0.22R Stopped
Apr 23, 2026, 1:35 AM CDT 10.75 2.44R Ran ≥1R
Apr 23, 2026, 12:10 AM CDT 5.75 3.00R Ran ≥1R
Apr 22, 2026, 11:00 PM CDT 10.25 0.00R Stopped
Apr 22, 2026, 6:25 PM CDT 5.5 3.00R Ran ≥1R
Apr 21, 2026, 5:20 PM CDT 10 1.68R Ran ≥1R
Apr 21, 2026, 5:16 AM CDT 14.25 0.32R Stopped
Apr 20, 2026, 6:00 PM CDT 6.25 2.56R Ran ≥1R
Apr 20, 2026, 1:30 PM CDT 11.75 1.81R Ran ≥1R
Apr 19, 2026, 10:40 PM CDT 12.25 2.84R Ran ≥1R
Apr 17, 2026, 1:45 PM CDT 14 3.00R Ran ≥1R

Sample backtest

Real backtested runs of this pattern, with commissions and slippage. Open one for the full equity curve and metrics, or backtest it yourself on your own contract and dates.

Backtest this pattern

Run it on your contracts, timeframes, and dates.