Bullish Candlestick Patterns and Bearish Patterns

Classify the setup before learning more pattern names

Use the matrix to choose which family to study. Reversal describes a proposed change from the preceding trend; continuation describes a proposed move in the same direction. Single or multiple candles describe the size of the formation. These are separate dimensions, so a multi-candle pattern can belong to either family. The examples below are a classification map, not a ranking or an entry checklist.

Why the same candle can have a different interpretation

A hammer and a hanging man share the small-body, long-lower-shadow shape. A hammer occurs after a decline; a hanging man occurs after an advance. Labeling the shape without its context removes the distinction. TradingView lets users select trend detection based on price relative to SMA50, price/SMA50/SMA200 alignment, or no trend detection. Changing that setting can change which bars qualify. These are operational definitions, not proof that one trend filter is profitable.

• Illustrative exercise: place the same long-lower-shadow candle after a declining sequence and after a rising sequence. The preceding bars change its interpretation even though its OHLC values are identical.
• Choose the trend definition before reviewing results. Record its lookback and whether it uses the bars before the formation.
• When the selected trend rule is not met, record an unclassified shape instead of forcing a reversal label.

Separate a completed pattern from trade confirmation

Pattern completion means every required candle has closed and meets the chosen definition. A bullish engulfing needs two candles, with the second real body covering the first; it does not require engulfing the entire high-to-low range. A further close above the pattern high would be an additional entry filter. It is not part of that basic engulfing definition. TradingView documents that high, low, and close can change on an open realtime bar, so a provisional shape can disappear before the close.

• Illustrative confirmation rule: after a completed bullish setup, require a later close above its high within three bars. Cancel the pending setup if price first trades below its low. These thresholds are hypothetical and need testing.
• Define what happens when the confirmation deadline expires. An expired setup should not become a late entry without a new rule.
• Waiting for a close or a later confirmation bar delays the signal and can change the entry price. It does not eliminate losing trades.
• If both a trigger and invalidation level trade within one historical candle, OHLC alone may not establish which happened first.

A learning path for bullish and bearish candlestick patterns

Start with one shape whose meaning changes with context, then study how two candles relate. Hammer versus hanging man teaches trend dependence. Bullish versus bearish engulfing teaches body coverage and direction. Rising three methods introduces continuation: TradingView describes five candles ending with a long bullish candle that closes above the first candle close. Follow the linked pattern guides below for their full definitions; use the cheat sheet when you need a compact visual reference.

• First exercise: identify body, upper shadow, and lower shadow on closed candles. Distinguish candle direction from the direction assigned to a pattern.
• Second exercise: compare bullish candlestick formations after declines with their bearish counterparts after advances. Write down the trend rule alongside each label.
• Third exercise: compare a reversal with a continuation on the same chart interval. Ask which preceding move the pattern is supposed to reverse or continue.
• Final exercise: record every qualifying occurrence in a fixed historical window, including failed and ambiguous setups. Do not select only the attractive examples.

Avoid counting the same price move as several confirmations

Several pattern detectors can label overlapping candles. Those labels may describe the same price movement, so counting them as separate votes can overstate the evidence. For a first test, choose one pattern family and one explicit context filter. Keep the formation, confirmation, invalidation, and exit rules separate. Compare any added filter with the original rule on a later, untouched sample before treating it as useful.

• Specify one decision per setup when several selected patterns overlap. Keep the individual labels for review.
• Record symbol, chart interval, session, costs, and the exact pattern definition. A result from one combination does not establish a universal win rate.
• Use the intended price data consistently. Do not assume a result transfers between markets, sessions, or chart types.