Falling Wedge pattern

What is a Falling Wedge?

A falling wedge is a downward-sloping, narrowing price structure. Both boundary lines point down, but the line through the highs falls more steeply than the line through the lows, so the range contracts as price moves lower.

That geometry is the whole argument for the pattern. Each new low is only slightly lower than the last, while rallies keep reaching the upper line, so sellers are getting less for their effort. Even though price is still falling, downside momentum is fading. When price finally closes above the upper line, the decline has lost its structure.

Despite pointing down, the falling wedge is traditionally read as bullish. After an extended decline it is treated as a reversal pattern; inside a larger uptrend it can be a corrective pullback, and then it acts as a continuation pattern. PatternX trades it long only and labels it reversal or continuation. That label is descriptive and does not change how the wedge is detected.

Anatomy of the Falling Wedge

A falling wedge has five parts worth naming:

The falling wedge is easy to misread next to its neighbours. A descending triangle has a flat floor rather than a falling one and is read as bearish. A downward-sloping parallel channel has lines that never converge. And a bull flag is a short, roughly parallel pullback after a sharp rally. The converging, downward-sloping shape is what separates the wedge from all three.

A wedge also needs room: if the lines start almost touching, the eventual target is tiny and the trade is rarely worth taking.

How to identify a Falling Wedge

Step-by-step identification:

1. Draw the upper line through at least two lower swing highs. With more than two points, a fitted line is more objective than one forced through the most extreme wicks.
2. Draw the lower line through at least two lower swing lows.
3. Confirm both slope down and that the upper line is the steeper one. If the lower line is steeper, the lines diverge and you have a broadening pattern, not a wedge.
4. Confirm the decline is meaningful. A wedge that barely drops is really a sideways range.
5. Confirm the squeeze. The range near the end should be clearly narrower than at the start.
6. Check momentum. Many traders look for an oscillator making higher lows while price makes lower lows, a divergence that supports the fading-pressure reading.
7. Wait for a close above the upper line, ideally soon after the last swing rather than long after the wedge stopped forming.
8. Set the stop and target before entering: below the wedge's lowest low, and the opening range added to the breakout price.

Confirmation

Traders usually want three things from a falling-wedge breakout: a clear close above the upper line, a pickup in volume, and evidence that the decline has actually stopped, such as a higher low after the break. Because the upper line slopes down, the breakout level gets lower with every bar, so a late breakout can occur at a price barely above the wedge's recent lows. Some traders therefore also want price to clear the most recent swing high inside the wedge before trusting the move.

PatternX's trigger is the first close above the upper line: the previous close must not have been more than 0.2% above the line, and the last swing must be no more than 15 bars old. The breakout candle must close above its open and carry at least 1.2 times its 20-bar average volume.

Invalidation

Before a breakout, a falling wedge is undermined if price breaks down through the lower line with force. That turns a slowing decline into an accelerating one, the opposite of what the pattern describes. After a breakout, a quick fall back under the upper line and then below the wedge's lows is the clearest failure.

PatternX sets the stop at the lowest swing low used to draw the wedge, or uses your own stop-loss if that is tighter, and applies no other post-entry invalidation logic. A pullback into the old wedge does not close the trade unless it reaches that stop. Position sizing divides the risk per trade (1% of balance by default) by the stop distance, so a deeper wedge low means a smaller position, not a larger planned loss.

Volume

Volume in a falling wedge classically declines as the wedge develops, consistent with sellers losing interest, and then expands on the breakout. A breakout on rising volume is generally treated as more convincing, because it suggests new buyers are arriving rather than sellers merely pausing. Some traders also watch for heavy volume on a final low that is quickly rejected, which they read as capitulation.

PatternX applies no wedge-specific volume analysis. Its single volume requirement is universal: the breakout bar's volume must be at least 1.2 times its 20-bar average. How far it exceeds that average feeds the Volume factor of the PatternX Score, alongside a separate breakout-strength factor that looks at the candle's body.

Common false signals

Common ways a falling wedge disappoints:

• The marginal break. Because the upper line keeps falling, a weak bounce can cross it without changing the trend. Price pokes above the line, then rolls over to new lows.
• The trend that never turned. In a strong bear market, wedges form repeatedly on the way down, and each breakout produces only a brief bounce.
• The badly fitted line. Lines drawn through hand-picked wicks can produce a wedge that isn't really there, and slightly different points give a different breakout level.
• The divergent shape. If the lower line is actually the steeper one, the range is widening and the bullish reading doesn't apply.

Check the higher-timeframe trend, wait for signs that the decline is stopping, and keep position size consistent with the stop distance. The false breakout entry covers the general mechanics.

Market context

A falling wedge reads differently depending on where it appears. After a long, steep decline, it suggests selling is exhausting and a reversal may follow. Inside a healthy uptrend it is often just an orderly pullback, and a breakout resumes the trend; many traders find this version easier to trust because the larger trend is on their side. In a persistent downtrend, by contrast, falling wedges can break out repeatedly without changing anything.

PatternX does not check the trend before the wedge, so it detects all three versions. The EMA20/50 trend alignment and higher-timeframe agreement factors in the PatternX Score are where that context shows up, together with the optional AI context layer described below.

In crypto markets

Crypto produces many falling wedges, because sharp sell-offs are often followed by slower, grinding declines as leverage is flushed out. Altcoins in particular can form wedges during weak markets that break out briefly and then rejoin the broader trend, so the direction of bitcoin and the overall market is useful context.

Liquidity matters most on the smaller coins where these patterns tend to look most dramatic. PatternX's crypto picker offers the top 20 coins by 24-hour volume on each exchange, and every crypto signal must show at least $3M of 24-hour volume, a spread of 0.25% or less and a liquidity score of at least 30. That limits, but does not remove, the risk of slippage on a breakout.

In stocks, ETFs and FX

In stocks, falling wedges often appear as pullbacks after a strong earnings move or during sector-wide corrections, and on daily charts they can sit inside a longer uptrend. Two practical risks stand out. Overnight gaps can open a stock above the upper line, leaving the entry far from the stop, or below the lowest low, so the stop fills at a worse price than planned. And when the whole market is falling, single-stock wedges tend to break out only to follow the index lower.

PatternX covers 123 US stocks, 67 ETFs, 62 leveraged and inverse ETFs, 29 Bitcoin and Ether ETFs and 28 FX pairs and spot metals through your own broker connection. See stock chart pattern recognition.

Timeframes

PatternX scans for falling wedges in the latest 70 bars on each of its eight timeframes (1m, 3m, 5m, 15m, 30m, 1h, 4h and 1d). Seventy bars is about 17.5 hours on 15m, almost three days on 1h, nearly 12 days on 4h and 70 sessions on the daily chart. Because 1d is the largest timeframe, a six-month wedge on a daily chart never fits the window and is not detected anywhere.

The breakout must also come within 15 bars of the last swing, so the practical window is shorter still: 2.5 days on 4h, or three weeks of US trading on 1d. New accounts scan 5m, 15m and 1h. Each signal is compared with the trend on the next timeframe up, which only adjusts the score.

Example: the logic step by step

For example, suppose a hypothetical altcoin on a 4h chart declines inside a wedge. Its upper line falls from 0.900 to 0.812 and its lower line from 0.840 to 0.790, so the upper line is clearly the steeper one.

• Opening range: 0.900 − 0.840 = 0.060, about 7% of price.
• Closing range: 0.812 − 0.790 = 0.022, about 37% of the opening range.
• Decline of the upper line: 0.088, more than 10%.
• Breakout: with the upper line now at 0.808, the latest bar's close is 0.818, while the previous close sat below the line.
• Stop: the lowest swing low, 0.790, a risk of 0.028.
• Target: 0.818 + 0.060 = 0.878, a reward of 0.060.
• Reward:risk: about 2.1.

With PatternX's defaults, the 0.2% buffer puts the stop near 0.788, which is tighter than a 5% stop at about 0.777, and the target stays at 0.878 because a 10% take-profit (about 0.900) is further away. The ratio becomes about 2.0. This is a made-up illustration, not a trade record.

How PatternX detects the Falling Wedge

PatternX finds falling wedges with fixed, inspectable rules applied to the latest 70 bars:

1. Swing points are found with a 2-bar window on each side, so each is confirmed two bars late.
2. Two least-squares lines are fitted, one through up to the last four swing highs and one through up to the last four swing lows (at least two each). Least-squares lines are calculated, not hand-drawn, so they don't depend on which wick someone prefers.
3. Shape: both lines must slope down, with the upper line steeper.
4. Size: the opening range must be at least 2% of price, and the upper line must fall by at least 1% across the wedge.
5. Squeeze: the closing range must be no more than 68% of the opening range.
6. Breakout: the latest bar must close above the upper line while the previous close was no more than 0.2% above it, and the last swing must be no more than 15 bars old.
7. Levels: stop = the lowest swing low; target = entry + the opening range.

The match is then filtered like every other pattern: breakout candle closing above its open, volume of at least 1.2× the 20-bar average, ATR at least 0.05% of price, liquidity score at least 30, spread no more than 0.25%, order-book imbalance against the long no worse than −0.5, and crypto 24-hour volume of at least $3M. Geometry scores are capped at 85. The pattern detection methodology covers all 20 detectors.

How PatternX Classic evaluates it

PatternX Classic gives a detected falling wedge a confidence from the 13-factor, deterministic PatternX Score. Geometry (capped at 85) and a fixed prior for the pattern supply 24% of the weight. For a wedge breakout, two other factors deserve attention: breakout strength, which compares the candle's body with its range, and volume, which measures the surge against the 20-bar average. A long-bodied, high-volume breakout scores better than a hesitant one that barely crosses a falling line.

Confidence is grouped into Elite (80+), High (65+), Medium (50+) and Weak, and must reach your minimum confidence (65 by default). The stop is the tighter of your stop-loss and the lowest swing low with a 0.2% buffer; the target is the nearer of your take-profit and the measured move. Setups under 1:1 reward:risk are dropped, signals expire after 15 minutes, and the risk controls apply to every order.

What PatternX AI adds

PatternX AI (Pro and Elite) evaluates a falling-wedge long only after the engine has detected and accepted it; it cannot find wedges itself. This pattern is a real test for its context factors. A falling wedge is built from lower highs and lower lows, so at the moment of breakout the market-structure factor, which checks whether the last three swing highs and lows within 60 candles agree with the trade, may still read as bearish, and trend strength (EMA12 against EMA26) often still points down. Momentum (RSI 14) and volume quality can pull the other way. If structure scores below 35 while momentum is below 40, or the overall context score is under 48, the setup is vetoed; otherwise its confidence is re-weighted. See Classic vs AI.