A falling wedge is a chart pattern made of two trendlines that both slope down and move closer together. The upper line slopes down more steeply than the lower one. StockCharts ChartSchool calls it a bullish pattern, whether it ends a downtrend or interrupts an uptrend, and says a convincing break above the upper line confirms it. The pattern still fails often. Thomas Bulkowski tested more than 800 cases in bull markets and found that falling wedges broke upward 68% of the time. Among bullish chart patterns he ranks it 31st of 39 for upward breakouts. A falling wedge is a probability, not a signal you can rely on.
What the pattern looks like
A falling wedge forms when price sets lower highs and lower lows inside a range that keeps narrowing. The lows still break below earlier lows, but each break is shallower. ChartSchool reads this as selling pressure losing force.
ChartSchool's identification points:
- Prior trend. As a reversal pattern, the wedge should come after a downtrend that is "at least three months old". It usually takes three to six months to form.
- Upper resistance line. At least two reaction highs, ideally three, each lower than the one before.
- Lower support line. At least two reaction lows, each lower than the one before.
- Slope. The two lines converge. The support line slopes down less steeply than the resistance line. If the lines run parallel, the pattern is a descending channel, not a wedge.
Bulkowski's rules (thepatternsite.com, statistics updated 27 August 2020) add two more conditions:
- Touches. "Price should touch each trendline at least five times to outline a good pattern." That means three touches on one line and two on the other.
- Duration. "3 weeks is the minimum duration, otherwise it's a pennant."
Reversal or continuation
ChartSchool describes two settings for the pattern:
- Reversal. The wedge slopes down in the same direction as the existing downtrend and marks its final low.
- Continuation. The wedge slopes down against an existing uptrend, as a pullback before that uptrend may resume.
ChartSchool treats both as bullish.
Falling wedge and rising wedge
The rising wedge is the mirror image. Its two converging lines slope up, and ChartSchool describes it as bearish "regardless of the type (reversal or continuation)". It also says volume should ideally decline while a rising wedge forms.
| Feature | Falling wedge | Rising wedge |
|---|---|---|
| Line slope | Down, converging | Up, converging |
| Bias given by ChartSchool | Bullish | Bearish |
| Confirmation | Break above resistance | Break below support |
Confirming a breakout
Breakout and volume
ChartSchool says confirmation comes when "the resistance line is broken in a convincing fashion". For extra confirmation, it suggests waiting for a break above the previous reaction high. ChartSchool also stresses volume: "Without volume expansion, the breakout will lack conviction and be vulnerable to failure."
Bulkowski treats the pattern as valid once price closes outside one of the trendlines. In his sample, volume trended down 72% to 75% of the time before the breakout. Upward wedges with heavy volume on breakout day performed better.
Throwbacks
ChartSchool notes that after a breakout, price "can sometimes" correct to test the broken resistance line as support. Bulkowski measured how often this happens: 62% of upward breakouts in a bull market threw back. A retest is common, but not guaranteed.
Momentum indicators
- RSI. J. Welles Wilder introduced RSI in his 1978 book New Concepts in Technical Trading Systems and used 14 periods by default. ChartSchool defines bullish divergence as price making "a lower low, and RSI forms a higher low". Inside a wedge, that matches the pattern's idea of weakening selling. ChartSchool also warns that "divergences are misleading in a strong trend".
- MACD. The MACD line is the 12-day EMA minus the 26-day EMA, and the signal line is a 9-day EMA of the MACD line. A bullish crossover occurs "when the MACD turns up and crosses above the signal line". The signal line lags, and ChartSchool advises caution with crossovers at extremes and in volatile markets.
Is your chart a falling wedge yet?
Answer five questions about the lines on your chart to see whether it meets the pattern rules in the article and whether a breakout is confirmed.
Do both trendlines slope down?
- Yes, both slope down → next question
- No, both slope up → This is a rising wedge, not a falling one
Do the two lines move closer together, with the upper resistance line sloping down more steeply than the lower support line?
- Yes, they converge → next question
- No, they run parallel → Parallel lines make a descending channel
Does price touch the trendlines at least five times in total, for example three times on one line and two on the other?
- Yes, five or more touches → next question
- No, fewer than five → Too few touches to outline the pattern
Has the pattern lasted at least 3 weeks?
- Yes, 3 weeks or longer → next question
- No, shorter than 3 weeks → Too short: under Bulkowski's rule this is a pennant
What has price done at the trendlines so far?
- Closed above the upper line, and volume expanded → Breakout confirmed under both sources' rules
- Moved above the upper line only intraday, or without volume expansion → Not a confirmed breakout yet
- Closed below the lower line → The wedge broke downward
- Still trading inside both lines → The wedge is still forming
Two converging lines that slope up form a rising wedge. ChartSchool calls the rising wedge bearish, and it is confirmed by a break below support, not above resistance.
ChartSchool says that if the lines run parallel, the pattern is a descending channel, not a wedge. The lines must converge.
Bulkowski says price should touch each trendline at least five times to outline a good pattern. With fewer touches, the lines may not describe a falling wedge.
Bulkowski sets 3 weeks as the minimum duration for a falling wedge. Anything shorter counts as a pennant under his definition.
Bulkowski treats the pattern as valid once price closes outside one of the trendlines. In his sample, breakouts tended to happen about 61% to 62% of the way to the apex, and volume trended down 72% to 75% of the time before the breakout.
ChartSchool ties confirmation to a convincing break of the resistance line and says that without volume expansion a breakout lacks conviction and is vulnerable to failure. An intraday move above the line is not a close beyond it.
A close above resistance with volume expansion meets the confirmation rules in the article. A confirmed breakout can still fail: in Bulkowski's bull-market data, 26% of upward breakouts failed to reach break-even, and 62% threw back to test the broken line. A stop becomes a market order once triggered, so its fill price can differ from the stop price in a fast market.
Bulkowski found about one in three falling wedges in a bull market breaks downward. For downward breakouts his data shows a 14% average decline, a 29% break-even failure rate and a 74% pullback rate.
This checks a chart against the article's pattern rules only: it cannot read your chart, does not cover the three-month prior downtrend, RSI or MACD, and its statistics come from Bulkowski's mainly equity sample, not a forecast for any crypto asset.
Price targets and the numbers behind them
ChartSchool gives no price target method for the falling wedge. Bulkowski gives two:
- Use the highest peak inside the wedge as the target for an upward breakout.
- Measure from the highest peak to the lowest valley, multiply that height by the percentage of wedges that meet their target, and add the result to the breakout price.
Bulkowski's bull-market statistics (updated 27 August 2020):
| Measure | Upward breakout | Downward breakout |
|---|---|---|
| Break-even failure rate | 26% | 29% |
| Average move | 38% rise | 14% decline |
| Throwback / pullback rate | 62% | 74% |
| Price target met | 62% | 29% |
| Performance rank (1 = best) | 31 of 39 | 27 of 36 |
Other points from the same page:
- Breakouts tend to happen about 61% to 62% of the way to the apex.
- Tall wedges perform better than short ones.
- An upward gap on breakout day points to a better-performing wedge.
These figures come from Bulkowski's sample of mainly equity charts. They are not a forecast for any crypto asset.
Managing risk on the trade
Pattern rules do not limit losses. Stops and position size do.
- Stop placement. Common reference points are below the latest swing low inside the wedge, or below the lower support line. Bulkowski's figures show that 26% of upward breakouts still fail to reach break-even, so a stop is part of the setup itself.
- Stops are not guaranteed prices. The US SEC explains that a stop order becomes a market order once triggered, and that "the execution price an investor receives for this market order can deviate significantly from the stop price in a fast-moving market." A stop-limit order avoids an unexpected fill price, but it may not execute at all.
- Position size arithmetic. Units = the amount you accept losing ÷ (entry price − stop price). For example, accepting a $100 loss with a $4 gap between entry and stop gives 25 units. Leverage multiplies both the gain and the loss on each unit.
Common mistakes
- Treating an intraday move above resistance as a breakout. Both sources tie confirmation to a decisive break or a close beyond the line.
- Ignoring volume. ChartSchool says a breakout without volume expansion is vulnerable to failure.
- Drawing the pattern with too few touches. Bulkowski asks for at least five touches in total.
- Calling a short consolidation a wedge. Under Bulkowski's definition, anything shorter than three weeks is a pennant.
- Treating a parallel channel as a wedge. The lines must converge.
- Expecting certainty. Bulkowski's data shows about one in three falling wedges in a bull market breaks downward, and he ranks the pattern in the lower half of bullish patterns.
Limitations: the statistics come from one researcher's dataset, last updated in 2020 and built mainly on stocks, so crypto results may differ. This article explains a chart pattern and is not trading or investment advice.
