Wedge Pattern: Rising and Falling Wedges on 76 Years of Data
A wedge pattern is a price shape bounded by two trend lines that slope in the same direction and draw closer together. A rising wedge climbs while narrowing and is read as bearish; a falling wedge slides while narrowing and is read as bullish, each confirmed when price closes outside the line it is expected to break.
A wedge is one of the few chart shapes where both sides agree on direction and still disagree on speed. This page defines the rising and falling wedge precisely enough for a computer to find them, shows the S&P 500’s falling wedge of June 2022, and reports what followed every break since 1950.
How it forms
Start with two lines. One joins the swing highs, the other joins the swing lows. In a wedge, both lines point the same way, up or down, and they are not parallel: they move toward each other.
A rising wedge has both lines climbing, with the lower one climbing faster. Each dip is shallower than the last, but each new high adds less than the one before. Price is still rising, yet each push gains less ground, and that loss of momentum is what the bearish reading rests on.
A falling wedge has both lines descending, with the upper one falling faster. Each new low undercuts the last by a smaller amount, while the rallies fade more quickly. Selling is still in control, but it is running out of distance, which is why the pattern is read as bullish.
The break decides it. A rising wedge confirms on a close below its lower line; a falling wedge confirms on a close above its upper line. A close out the other side, through the upper line of a rising wedge, cancels the reading.
Wedges differ from triangles only in slope. A triangle has one flat side or two sides pointing in opposite directions. A wedge leans, and the lean is the reason its expected break goes against it.
The exact definition used here
The scan uses S&P 500 daily closes. A swing high is a close that is the highest of the 11 sessions centered on it; a swing low is the lowest. A wedge is four swings in a row that alternate high, low, high, low (in either order), within 120 sessions, where:
- rising: both highs and both lows step up, and the line through the lows rises faster, measured in percent per session, than the line through the highs;
- falling: both highs and both lows step down, and the line through the highs falls faster than the line through the lows;
- the gap between the two lines at the last swing is no more than 75% of the gap at the first.
The break is the first close outside the expected line from the sixth to the 40th session after the last swing, because a swing is only known five sessions after it forms. A close through the other line first cancels it. Two wedges of the same kind may not overlap. The outcome is the close 20 sessions after the break.
That wait matters. In 106 of the 163 rising wedges and 57 of the 77 falling wedges, the index had already closed past the expected line within those first five sessions. The rule ignores those closes, because nobody could yet have known the last swing was a swing.
A worked example
Between May and June 2022 the S&P 500 drew a falling wedge. The four swings were a low close of 3,900.79 on 19 May, a high of 4,176.82 on 2 June, a lower low of 3,666.77 on 16 June and a lower high of 3,911.74 on 24 June.
Check the slopes. The upper line fell from 4,176.82 to 3,911.74 over 15 sessions, about 0.42% per session. The lower line fell from 3,900.79 to 3,666.77 over 19 sessions, about 0.32% per session. The highs were falling faster, so the lines were closing in.
Check the narrowing. Extended back to 19 May, the upper line sat 435.08 points above that day’s close. By 24 June the gap was 306.55 points, 70% of the starting width and inside the 75% limit.
The break came on 5 July 2022. The upper line projected to that day stood at 3,805.71, and the index closed at 3,831.39, above it. It had closed just above the line twice before, on 27 June and 1 July, but both closes fell inside the five-session wait for the last swing, so the rule dates the break to 5 July.
Then the follow-through. Twenty sessions later, on 2 August, the index closed at 4,091.19, up 6.78% from the break. The usual projection adds the wedge’s starting width, 435.08 points, to the breakout close, which gives 4,266.47. The index closed above that on 12 August 2022, at 4,280.15.
This is the most recent falling wedge the scan found, and the rise it is read for did follow. The data section below shows how typical that was.
The original data
Across 19,305 S&P 500 daily closes from 3 January 1950 to 25 September 2026, the rules found 163 rising wedges. 111 of them broke down through the lower line; in the other 52, the index closed above the upper line first.
Of the 110 rising-wedge breaks with 20 sessions of data after them, 48 closed lower 20 sessions later, or 43.6%. Across every 20-session window in the same period, the index closed lower 38.2% of the time. The median change after a rising-wedge break was a gain of 0.44%.
Falling wedges were rarer and resolved more often: 77 found, 58 broke upward. Of those 58, 40 closed higher 20 sessions later, which is 69.0%, against 61.8% of all 20-session windows. The median change was a gain of 2.17%.
The projected level was reached less often than it was missed. Within 60 sessions of the break, 34 of 110 rising wedges and 23 of 58 falling wedges got there.
Read the two results together. Both came in five to seven points above their base rates. On 110 and 58 cases, gaps that size are well within what chance alone produces, so neither result separates from an ordinary 20-session window, and both samples are small.
Few videos cover it, and two carry most of the views. In the 24,971-video corpus this site studies, counted once per video id, 5 titles mention a wedge, from 5 channels, at a median of 527 views. Two of them passed 20,000 views. Every wedge, its swing dates and its outcome are listed in the chart pattern event file. The daily closes the scan ran on are in the S&P 500 close file.
Why the bearish wedge is the harder read
It is a bearish shape drawn entirely out of higher highs and higher lows. Anyone looking only at trend structure sees an uptrend, and in 52 of 163 cases the index broke out above the upper line before it ever closed below the lower one.
The index drift works against it as well. The S&P 500 rose over 61.8% of 20-session windows, so a pattern that asks for a decline starts behind. A rising wedge that preceded a lower close 43.6% of the time still preceded a higher or flat close more often than not.
The falling wedge has the easier job. It asks for the direction the index already leans, so most of its 69.0% is that lean: 61.8% of all 20-session windows closed higher anyway.
When it fails
The first failure is the break the wrong way. In 52 of the 163 rising wedges, price closed above the upper line first, and a trader positioned for the downside was holding into strength.
The second is a mechanical wedge that no chartist would draw. The scan’s rising wedge of spring 2026 began with a close of 6,946.13 on 25 February, fell 8.7% to 6,343.72 by 30 March, then climbed to a higher high. Four swings met every rule, but the shape was a sharp drop and recovery, not a slow narrowing climb. Its break on 2 June 2026 was followed by a 1.66% fall over 20 sessions, the fall the pattern is read for, on a shape that was barely a wedge.
The third is line-drawing freedom. Two swings define a line exactly, but which swings to use is a choice. Pick different ones and the break date moves, and with it the outcome. Deciding the swing rule before looking is the only defense.
A fourth is a false break. Price closes just outside the line, then returns inside it the next day. The false breakout page covers why those snap-backs happen at obvious lines.
And a fifth is treating the projection as a destination. It was reached in about a third of rising wedges and about two in five falling wedges. The rest of the time it was only a number on a chart.
Related
The named chart patterns page sets the wedge beside the head and shoulders and cup and handle, each as a checkable condition. Trend lines explains how the two sides of a wedge are drawn and why the choice of swing points matters. And the triangle pattern page covers the converging shape whose sides do not lean the same way.
Draw both lines before you decide what the wedge means, and write down the close that would break each one. If you cannot say which close breaks the pattern, you do not have a pattern yet, only two lines that happen to lean the same way.
— Michael Whitman
This page is educational, not financial advice. Test every idea on your own charts before risking money.