Triangle chart patterns: ascending, descending and symmetrical

A triangle is a stretch of trading in which the swings get smaller. Highs and lows squeeze toward each other until price runs out of room and breaks out of one side. The lines through the highs and the lows form the triangle, and the direction of each line gives the pattern its name: ascending (flat top, rising lows), descending (falling highs, flat bottom) or symmetrical (both lines converging).

Triangles are usually taught as continuation patterns, pauses in which a trend catches its breath before resuming. The ascending triangle is said to lean bullish and the descending triangle bearish. In practice, triangles break in either direction often enough that the breakout itself, not the shape, is what most traders act on.

This guide explains how to draw each type, where the breakout and the measured target are, how volume behaves, and four real examples on US stocks, including an ascending triangle that broke down and a symmetrical triangle whose breakout failed.

The three triangles

Ascending triangle

Sellers hold a fixed ceiling, perhaps a round number or a prior high, while buyers step in at higher and higher prices on each pullback. The flat top is resistance and the rising line underneath is the trendline of higher lows. The textbook reading is bullish: demand is increasing while supply is static, so the ceiling eventually gives way.

Descending triangle

The mirror image. Buyers defend a fixed floor while each rally stops lower. The flat bottom is support, and the falling line on top connects lower highs. The textbook reading is bearish: sellers are becoming more aggressive while buyers only hold one price.

Symmetrical triangle

Both lines slope toward each other: lower highs and higher lows. Neither side is gaining ground. Edwards and Magee treated it as a pause that most often resolves in the direction of the trend that preceded it, while acknowledging plenty of exceptions. Many traders simply treat it as neutral until it breaks.

Rules of construction

• Touches. At least two highs on the upper line and two lows on the lower line. Five alternating touches give a well-defined pattern.
• A shrinking range. Each swing should be smaller than the one before. If the swings stay the same size, you are looking at a consolidation range or a rectangle, not a triangle.
• Flat means flat. For an ascending or descending triangle, the horizontal side should be close to level, within a percent or two. If both lines slope the same way, the pattern is a wedge.
• Volume. In the classical description, volume tends to shrink as the triangle narrows and expands on the breakout. A breakout on light volume is treated with more suspicion.

Breakout, target and the apex

The apex is where the two lines would meet. Many practitioners prefer breakouts that come about one-half to three-quarters of the way from the triangle's start to its apex. A triangle that drifts all the way to the apex has usually lost its tension. Price simply leaves the pattern because the lines have run out, and the "breakout" carries little information.

Retests. After a breakout, price often returns to the broken line. An ascending triangle's old ceiling may act as support on the way back. Whether to wait for that retest is a trade-off: a better entry when it happens, a missed move when it does not.

Four real triangles

The examples were found by mechanical rules (described below) on daily bars. The orange lines run through the first and last swing high and the first and last swing low. The shaded column marks the session the outcome was decided.

Cisco, 2024–2025: the textbook case. CSCO's rallies stalled at $60.23 on December 5, 2024 and $60.06 on December 26, while its pullbacks bottomed at $56.35 (November 19), $57.27 (December 20) and $58.17 (January 13, 2025). On January 15 it closed at $59.98, just above the slightly declining upper line. The height at the start was $3.88, for a target near $63.86. The stock reached it on February 13, twenty sessions later.

Deere, 2025: an ascending triangle that broke down. After a 32% rally from its April 9 low, DE stalled three times near $531 (May 19, June 6 and June 18) while its lows rose from $501.02 to $507.73. That is a classic ascending triangle. On June 25 the stock closed at $506.74, below the rising lower line, and the measured target of about $475 was reached on August 14, 35 sessions later. A trader who bought in anticipation of an upside breakout would have been on the wrong side of the move.

FedEx, 2025: a descending triangle. FDX held a floor near $197 three times (February 4, February 21 and March 6, 2025) while its rallies weakened from $217.12 to $213.54. On March 11 it closed at $195.78, below support. The target was about $177.68, 9% lower. The stock drifted just below support for a week. On March 21 it gapped down from a $198.40 close to a low of $175.04, reaching the target in one session. News can complete a pattern as easily as it can break one.

T-Mobile US, 2025: a breakout that failed. TMUS made lower highs at $247.55, $243.28 and $238.16 and higher lows at $220.00 and $225.28 between June and July 2025. On July 24 it gapped above the upper line and closed at $247.50, up 5.8% on the day. The target was about $275. The stock reached $261.56 on August 20, roughly halfway, then faded, and on October 6 it closed below the last swing low of $225.28. The pattern failed even though the breakout had looked strong.

What our data says about triangles

We searched the same 60 large US stocks and ETFs used in our double top test with a 3% ZigZag, looking for five alternating swing points whose ranges shrank and whose lines met our definitions (a "flat" line changes by less than 1.5% over the pattern). The rules found only 37 completed triangles in two years, and when we varied the thresholds the counts and the breakout directions shifted a lot. Our sample was too small and too sensitive to the definition to support a reliable statistic, so we do not publish one here.

Two qualitative lessons held up across every setting we tried:

• The "bias" is weak. Ascending triangles broke down and descending triangles broke up often enough that betting on the textbook direction before the breakout was not supported by our data.
• Swing detection uses hindsight. A swing point is only confirmed after price moves away from it, so any rule-based count includes information a trader would not have had at the time. Results from simple pattern scanners, ours included, should be read with that in mind.

For large-sample statistics, Bulkowski's Encyclopedia of Chart Patterns catalogs breakout direction, failure rates and average moves for each triangle type. Those figures depend on his definitions and his data, from earlier decades.

Common mistakes

• Trading the bias instead of the breakout. The Deere example shows why. Wait for a close outside the lines.
• Drawing lines to fit. Two touches on each side is the minimum. A line that cuts through several bodies is not a boundary.
• Ignoring the apex. A break that comes after the lines have nearly met is often a drift, not a decision.
• Forgetting that news overrides geometry. Two of the four examples turned on overnight gaps. A pattern does not know when earnings are due. Check the earnings calendar before relying on one.

Triangles compared with similar patterns

How Finabase helps you draw and test triangles

Finabase does not detect triangles automatically. What it gives you:

• The trend line tool. Every symbol's chart has a trend line tool anchored to two bar dates and prices. Draw one through the highs and one through the lows. Signed-in users' drawings are saved per symbol, so you can come back after the breakout. Try it on a one-year CSCO chart.
• The ZigZag study. To choose touch points by rule rather than by eye, add the ZigZag with a small deviation, such as the 3% ZigZag on FDX. The ZigZag indicator: filtering swings, and why the last leg repaints guide explains why its last leg can change.
• The volume pane under price, to check whether volume dried up inside the triangle and expanded on the breakout.
• The Scanner. The Scanner lists stocks making 52-week highs or gapping sharply. An ascending triangle under a 52-week high that finally breaks will often show up there.

Frequently asked questions

Is an ascending triangle always bullish?

No. It is bullish only if price breaks out above the flat top. Ascending triangles also break down, as Deere's did in June 2025. The shape describes a balance of buying and selling. It does not decide the outcome.

How many touches does a triangle need?

At least two on each line, so four in total. Five or more alternating touches make the pattern clearer. Fewer than that, and almost any two converging swings could be called a triangle.

What is the difference between a triangle and a pennant?

A pennant is a small, short symmetrical triangle that forms right after a steep, almost vertical move (the pole). It usually lasts a few days to a few weeks. Triangles are larger and do not require a pole. The Flags and pennants: trading the pause after a sharp move guide covers them in detail.

Where should a stop go after a triangle breakout?

A common choice is beyond the last swing on the opposite side: below the last low for an upside breakout, above the last high for a downside one. Our examples used that rule to judge failure.

Summary

Triangles show a market narrowing its range until one side gives way. Ascending, descending and symmetrical triangles each tell a slightly different story about buyers and sellers, but the story does not decide the ending. Our examples included a textbook success, a bullish triangle that broke down, and a strong-looking breakout that failed. Draw the lines carefully, wait for a close outside them, measure the target from the widest point, and place your stop where the breakout would be proven wrong.