Volume shapes versus triangle and double-pattern breakouts

Confirm the pattern first

A five-way volume-shape catalog, flat, receding, rising, U, and dome, was applied to a reviewed set of 38,500 chart patterns and then compared with post-breakout results. Volume-shape is the overall contour of volume across a completed chart pattern. It is distinct from volume-trend, which is the signed slope of volume from pattern start to end, typically taken from a linear fit rather than from the five named labels.

A symmetrical triangle is defined by two converging boundaries, at least two touches on each line, frequent side-to-side crossings, little unused interior space, and a breakout when price closes outside a boundary. An ascending triangle compresses price between a flat upper boundary and a rising lower boundary, with repeated touches, side-to-side travel, and limited unused space. A broadening formation expands between diverging boundaries, entered from above as a bottom or from below as a top, with repeated touches and filled interior movement. A breakout is a close outside a pattern boundary that starts the post-pattern measurement window.

Scarce volume shapes were pooled

Flat volume across a symmetrical triangle is described as rare, occurring in only 1 percent of those patterns, so other volume contours are easier to find first. In a reviewed set of 1,092 ascending triangles, the breakout was upward 70 percent of the time, and receding volume, higher at the start than at the end, is also described as uncommon.

Rising and receding volume shapes are treated as scarce enough that they were pooled into a random-volume-shape bucket. Random-volume-shape is a leftover volume contour after scarce rising and falling shapes are pooled because they do not appear often enough to stand alone. That leftover class is not the same classification as a rising or falling volume-trend slope from linear regression.

Post-breakout averages did not favor one shape

Post-breakout averages were computed to the ultimate high or low before a 20 percent trend change. That ultimate-move is used here as a historical yardstick rather than a forecast. Expenses were ignored, and cells with fewer than 30 observations were labeled too few samples to treat as accurate.

Across the tabulated patterns, no single volume shape produced a consistent post-breakout advantage. Dome and U shapes were the most common, and large-sample triangles showed little difference between those two shapes.

Pattern-specific historical associations were mixed: dome-shaped volume with broadening tops; U-shaped volume with broadening bottoms and double tops; random volume shapes with ascending triangles; rising volume trends with ascending triangles; and receding volume trends with broadening formations and symmetrical triangles.

The source conclusion is that volume shape is a weak predictor of results after a chart-pattern breakout, so any remaining edge is only a same-pattern comparison among already-identified setups.