Is a descending broadening wedge pattern bullish?

Its downward tilt does not confirm a bullish outcome. A descending broadening wedge has falling boundaries and increasing width; an upside break requires a separate event rule. Price can break the lower boundary or return inside after an upside break. Keep the geometry and the subsequent outcome separate.

Are widening wedge and expanding wedge the same pattern?

They usually refer to a broadening wedge, but names are less precise than boundaries. On this page, both lines must slope in the same direction and their vertical separation must increase. Opposite slopes describe a broader megaphone structure under this classification.

How many touches confirm a broadening wedge?

There is no single touch count used by every definition. Bulkowski specifies three peaks and three valleys for ascending broadening wedges, and at least five total touches with at least two on each side for descending ones. This page deliberately uses three on each side for both variants. Touches establish the chosen shape; they do not confirm a trade.

Can I measure width by subtracting the last high and low?

Not reliably when those points occur on different bars. Evaluate both lines at the same starting bar and the same ending bar, then compare their vertical separations. Otherwise, movement along the slope can be mistaken for expansion.

Does waiting for a closed candle prevent every false breakout?

No. A closed candle settles that bar, but a later candle can return inside the wedge. It also does not remove pivot confirmation delays or every source of repainting. The worked example records completed closing breaks that fail on subsequent bars.

TradingView chart patterns

Broadening wedge pattern: widening boundaries and failed breaks

A broadening wedge pattern has two boundaries that slope in the same direction while moving farther apart. In an ascending version, the upper line rises faster than the lower line. In a descending version, the lower line falls faster than the upper line. A widening wedge pattern describes expansion, while an ordinary rising or falling wedge narrows. The shape alone does not confirm a breakout direction.

See the decision guide

Broadening versus converging wedges

Let U(t) be the upper boundary and L(t) the lower boundary at bar t on a linear price scale. Their vertical separation is W(t) = U(t) - L(t). Require W(t) to remain positive over the formation. With straight boundaries, the width changes by the upper slope minus the lower slope each bar. The table is a geometric classification, not a forecast.

An expanding wedge pattern is another name used for a broadening wedge. A formation with an upward upper line and a downward lower line is a broadening formation, but does not meet the same-direction wedge definition used here.
TradingView describes its Falling Wedge pattern as a converging channel. Do not assume that its detector identifies a descending broadening wedge simply because both patterns point downward.
These equations use arithmetic prices and bar indices. Drawing straight lines in logarithmic price space changes the geometry; keep the scale and calculation convention consistent.

Structure	Slope relationship	Width over time
Ascending broadening wedge	Both positive; upper slope is larger	Increases
Descending broadening wedge	Both negative; lower slope is more negative	Increases
Converging rising wedge	Both positive; lower slope is larger	Decreases
Converging falling wedge	Both negative; upper slope is more negative	Decreases
Parallel channel	Equal slopes	Stays constant

Define touches and a minimum expansion

A boundary needs separated swing points, not repeated candles sitting near one price. Bulkowski describes touches at minor highs and lows and excludes price cutting through a line. His published touch counts differ between ascending and descending wedges. For the custom detector below, use three confirmed highs and three confirmed lows in alternating order for either direction. This is a stricter chosen specification, not a universal minimum.

The 0.25 allowance is in price units, not percent or chart pixels. It belongs to this hypothetical example and needs a different interpretation on instruments with different prices or tick sizes.
When all checks pass, freeze the six pivots and two boundary equations for 20 completed bars starting at candidate confirmation. Evaluate no replacement candidate during that window.
An extra swing that fails the selection or touch rules is a rejection to record. Moving the line to fit that swing creates a different candidate and must not rewrite the earlier one.

Decision	Illustrative rule	Reason
Swing confirmation	Three bars on each side of an extreme; reject equal highs or lows	Each accepted point has a known confirmation time
Point selection	Evaluate the latest six alternating confirmed pivots, with adjacent pivots at least four bars apart	Avoid selecting a different historical combination after seeing the outcome
Boundary anchors	Join the first two selected highs for U and first two selected lows for L	Third touches test lines that are already defined
Third-touch tolerance	The third high and low must each be within 0.25 price units of their own line	Make near-touch acceptance measurable
Expansion	W at the last pivot divided by W at the first pivot must be at least 1.5	Reject near-parallel candidates; require a positive initial width
Formation containment	No completed close from the first pivot through candidate confirmation may exceed either boundary by more than 0.25	Reject a formation already interrupted by a qualifying break

Calculate a descending broadening wedge

Assume hypothetical swing highs at bars 10, 20 and 30 are 109, 108 and 107. The alternating swing lows at bars 15, 25 and 35 are 101.5, 98.5 and 95.5. These six points give U(t) = 110 - 0.1t and L(t) = 106 - 0.3t. Both lines fall, but the lower line falls faster. Assume the surrounding candles satisfy the stated pivot and containment rules.

The width ratio is 11 / 6, approximately 1.83, which passes the chosen 1.5 minimum. Each third pivot lies exactly on its boundary, so both touch errors are zero.
A different lower boundary of L(t) = 100 - 0.05t would narrow toward the same upper line: width falls from 9.5 at bar 10 to 8.25 at bar 35. Lower highs and lower lows can therefore occur in either expanding or contracting structures.
The final low at bar 35 is only confirmed at the close of bar 38 under the three-right-bar rule. A drawing placed at bar 35 does not make the complete candidate available three bars earlier.

Bar	Upper boundary	Lower boundary	Width
10	109.0	103.0	6.0
20	108.0	100.0	8.0
30	107.0	97.0	10.0
35	106.5	95.5	11.0

Keep repeated false breaks visible

Continue the same hypothetical candidate with its frozen equations. Define an upside event as a completed close above U(t) + 0.25 and a downside event as a close below L(t) - 0.25. Record the first eligible event, then wait for a close inside the raw boundaries before allowing another. For this audit, a return inside within the next three bars marks that event as failed. This failure label is a study rule, not a claim that price cannot resume the move.

A wick outside followed by an inside close would produce no event under this closing rule. A close outside can still fail, as the two events demonstrate.
Requiring two consecutive closes beyond the same buffered boundary would reject both one-bar breaks here. That filter also delays detection and can miss moves; the example does not establish that it improves results.
Preserve both failed events through the review window ending at bar 57. Do not widen the saved lines to absorb them or delete them when a later pattern looks clearer.
This log is not a trading strategy. Entries, exits, size, fees, slippage and gap handling would need separate rules before any performance test.

Bar and close	Boundary check	Recorded event
38: close 105.0	U = 106.2; L = 94.6	Candidate confirmed; price is inside
39: close 106.4	106.4 exceeds U + 0.25 = 106.35	Upside break
40: close 105.7	Between L = 94.0 and U = 106.0	Upside event failed; detector rearmed
42: close 93.1	93.1 is below L - 0.25 = 93.15	Downside break, assuming bar 41 stayed inside
43: close 94.0	Between L = 93.1 and U = 105.7	Downside event failed; detector rearmed

Record identification stability before testing trades

TradingView documents how a pivot plotted on a past candle can become visible only after later bars confirm it. For a broadening-wedge detector, save the pivot bar, its confirmation bar, the six selected anchors, both slopes, the width ratio and every event time. Compare that log during bar-by-bar review with the history after reloading the chart.

Separate provisional candidates from confirmed candidates. Track how often provisional anchors change, rather than presenting the final drawing as if it had always been fixed.
Check that reloads preserve events produced from the same confirmed data and settings. Investigate changes in source data, selection rules or intrabar calculations when they do not.
Ask Pineify to draft the written rules with candidate IDs, frozen boundary values and separate confirmation markers. Inspect the generated code and validate its syntax before testing the detector on TradingView.
Start with ordinary price candles on one named symbol, timeframe and session. No recognition accuracy, preferred parameter set or trading edge is established by the illustrative prices on this page.

Where Pineify fits

Pine Script AI Coding Agent

Turn your pivot, width and breakout specification into a Pine Script draft for review. Request a log of anchor changes and confirmation times. This educational page does not scan symbols or provide a tested broadening-wedge indicator.

Draft broadening-wedge rules

Related resources

TradingView chart patterns Double bottom: a fixed neckline comparison Inverse head and shoulders: a sloping neckline Bar-close alert frequency

Frequently asked questions

Educational information, not investment advice. All example prices, thresholds and event paths are hypothetical. No historical trading results or success rates are presented. A recognized shape or confirmed breakout does not guarantee direction or profit.

Sources and verification

Last verified: 2026-10-16

Published by Pineify. Pineify is an independent product and is not affiliated with TradingView. Sources support the technical explanations; examples do not establish trading performance.