Symmetrical Triangle pattern

What is a Symmetrical Triangle?

A symmetrical triangle forms when buyers and sellers both become less aggressive at the same time. Each rally stops below the previous high and each decline holds above the previous low, so a falling upper trendline and a rising lower trendline squeeze price toward a point called the apex. The slopes don't need to match exactly; what matters is that one line falls and the other rises.

Unlike the ascending and descending triangles, it has no flat edge, and so no built-in bias. It describes indecision and shrinking volatility, and periods of compression are often followed by expansion. Classical texts treat it mainly as a continuation pattern that tends to break in the direction of the prior trend, but it can just as well mark a reversal, which is why PatternX labels it reversal or continuation.

PatternX treats it as bilateral. It waits for price to break one of the lines and trades that direction: long on a break above the upper line, short on a break below the lower one.

Anatomy of the Symmetrical Triangle

A symmetrical triangle has two sloping edges and no horizontal one.

Two measurements carry most of the information. The opening range sets the measured-move target, and the ratio of closing range to opening range shows how far the coil has tightened. A pattern whose lines barely converge is closer to a channel than a triangle.

Many traders also watch where along the pattern the break happens. Classical charting often describes breaks in the later part of the triangle, but well before the apex, as the cleanest; once price drifts right up to the apex, the pattern has usually lost its tension.

How to identify a Symmetrical Triangle

How to identify a symmetrical triangle, step by step:

1. Find at least two lower highs. Connect them; the line must slope down.
2. Find at least two higher lows. Connect them; the line must slope up.
3. Check that the lines converge. If both lines slope the same way, you are looking at a rising wedge or a falling wedge instead.
4. Check the scale. The opening range should be wide enough to beat routine noise. A coil spanning a fraction of a typical day's range is rarely worth trading.
5. Check the tightening. The latest range should be clearly narrower than the opening one.
6. Note the trend that preceded it. The pattern has no bias of its own, so the prior trend and the higher timeframe are your main clues about which break is likelier.
7. Prepare for both outcomes. Decide in advance what you will do on a break up and on a break down, including where each stop and target would sit.
8. Rule out a pennant. A pennant is a small triangle that follows a sharp, fast move; a symmetrical triangle needs no such pole and usually takes longer to form.

Confirmation

Because the pattern does not choose a side, confirmation is mostly about the quality of the break. Traders look for a decisive close outside one of the lines, a rise in volume, and continuation over the following bars. Some also wait for a return to the broken line, which then acts as support or resistance from the other side. A break in the direction of the prior trend is generally treated as more trustworthy than one against it.

PatternX acts on the latest bar. A close above the upper line produces a long signal and a close below the lower line a short signal, and either one must pass the universal checks, including a breakout candle whose body points the same way and volume of at least 1.2 times the 20-bar average. The latest bar may still be forming, so a break can fade before it completes.

Invalidation

A symmetrical triangle has an unusual invalidation profile because both sides can trigger. Before a break, the pattern simply ends if price drifts into the apex without breaking out, or if a new swing interrupts the sequence of lower highs and higher lows. After a break, the trade is in trouble if price comes back through the opposite line, because the market has rejected one side and broken the other.

PatternX's stop for this pattern sits at the opposite line's value at the breakout bar (or at your own stop-loss if that is tighter), and that is the only invalidation it applies once a trade is open; there is no post-entry logic beyond the stop level. Near the apex the two lines are close together, so that stop is tight and can be reached by ordinary noise.

Volume

Volume typically contracts as a symmetrical triangle forms, which is one reason the pattern is called a coil: activity and range shrink together. Traders then look for a clear expansion in volume on the breakout. Because the pattern can break either way, volume is one of the few clues to which side has real participation behind it, and a quiet break in one direction followed by a high-volume move the other way is a classic head-fake sequence.

PatternX does not measure volume inside the triangle. It applies only the universal rule that the breakout bar's volume must be at least 1.2 times its 20-bar average, whichever direction the break goes. The size of that surge also contributes to the Volume factor in the score.

Common false signals

The symmetrical triangle's main weakness is that its signal can come from either side, so a false break in one direction is often followed by a real move in the other.

• The head fake. Price breaks one line, draws traders in, then reverses through the opposite line. It is this pattern's defining failure.
• The apex drift. Price wanders sideways into the apex, and the eventual break is small and directionless.
• The news break. Compression ahead of a scheduled event resolves on the event, often with gaps or long wicks through both lines.
• The noise break. On low timeframes the lines sit close together, and a single wide bar can cross one of them.

Defences include waiting for a clear close beyond the line, giving more weight to breaks in the direction of the prior trend, and avoiding entries just before known events. None of them removes the risk of a false breakout.

Market context

Since the pattern itself is neutral, context does most of the work. A symmetrical triangle inside a strong uptrend is usually read as a pause, with an upside break as the conventional expectation; in a downtrend the reverse applies. At the end of an extended move, the same coil can mark exhaustion and a reversal.

Compression also tends to build ahead of scheduled events such as earnings, central-bank decisions or major economic releases, and the event often decides the breakout direction. That makes the pattern a poor guide in the hours before such news.

PatternX scores this context rather than filtering on it. Trend alignment (EMA20 against EMA50) and higher-timeframe agreement are factors in the PatternX Score, so a break against the prevailing trend loses points on both.

In crypto markets

In crypto, symmetrical triangles frequently form while the market waits for a catalyst, such as a macro data release, an exchange listing or a token unlock, and they can resolve abruptly. Perpetual futures make both directions equally tradable, which suits a bilateral pattern. Positioning still matters: if funding shows one side heavily crowded, a break against that crowd can be amplified by forced liquidations.

Weekends and quiet hours bring thinner order books, and thin books make false breaks more likely. PatternX's universal filters reject setups where the spread is wider than 0.25%, the liquidity score is under 30 or the order book leans against the trade worse than −0.5, and crypto setups also need $3M of 24-hour volume. More on scanning crypto is in crypto chart pattern recognition.

In stocks, ETFs and FX

In stocks and ETFs, symmetrical triangles often form in the days before earnings, when traders are reluctant to commit. A breakout that happens on the earnings gap is an event, not a pattern, and the gap can open beyond both the trigger and the stop. FX pairs show similar coils ahead of interest-rate decisions and major data releases, and the break can come as liquidity arrives at a major session open.

PatternX can scan these markets through your broker connection, across 309 curated instruments that include 28 FX pairs and spot metals. A symmetrical triangle can produce a short signal on a stock, so check that your account allows short selling, or switch short trades off in the risk settings. See traditional markets for what is covered.

Timeframes

PatternX looks for symmetrical triangles in the last 75 bars on each timeframe you scan, from 1m through 3m, 5m, 15m, 30m, 1h and 4h to 1d. On a 5m chart that is just over six hours; on 1h it is a little over three days; on 1d it is 75 sessions. New accounts start with 5m, 15m and 1h.

Low timeframes generate many small coils, and because the lines sit close together, a stop on the opposite line can be very tight relative to normal movement. Higher timeframes produce fewer, larger triangles with wider stops. Every breakout is also compared with the trend on the next timeframe up, which nudges the score but cannot block a signal. Coils larger than 75 bars are not detected on that timeframe.

Example: the logic step by step

For example, take a hypothetical stock near 50 on a 30m chart. Its swing highs fall from 52.00 to 50.80 and its swing lows rise from 49.00 to 50.20.

• Opening range: about 52.00 − 49.00 = 3.00, roughly 6% of price.
• Closing range: 50.80 − 50.20 = 0.60, a fifth of the opening range.
• At the breakout bar: the upper line is at 50.70 and the lower line at 50.30.

Upside case. Price closes at 50.95, above the upper line, so the signal is long. Stop at the lower line, 50.30 (risk 0.65); target 50.95 + 3.00 = 53.95; reward:risk about 4.6.

Downside case. Price instead closes at 50.05, below the lower line, so the signal is short. Stop at the upper line, 50.70 (risk 0.65); target 50.05 − 3.00 = 47.05; the same ratio.

The large ratio comes from the tight stop near the apex, and that stop is also easy to hit. With PatternX's defaults, the 0.2% stop buffer widens the risk to about 0.75, for a ratio near 4.0, and both targets sit inside a 10% take-profit. These numbers are illustrative only.

How PatternX detects the Symmetrical Triangle

PatternX's symmetrical-triangle detector is the only one of its 20 detectors whose direction is set by the breakout. On each scan it works through the latest 75 bars:

• Swings. Swing highs and lows use a 2-bar window on each side, so each one is confirmed two bars after it forms.
• Lines. It fits an upper line through recent swing highs and a lower line through recent swing lows. The upper line must slope down and the lower line must slope up.
• Size. The opening range must be at least 2% of price.
• Convergence. The closing range must be no more than 68% of the opening range.
• Direction. A close above the upper line on the latest bar produces a LONG; a close below the lower line produces a SHORT.
• Stop. The value of the opposite line at the breakout bar: the lower line for a long, the upper line for a short.
• Target. Entry plus the opening range for a long, entry minus it for a short.

The universal filters then apply in the direction of the trade: the breakout candle's body must point that way, its volume must be at least 1.2× the 20-bar average, ATR at least 0.05% of price, liquidity score at least 30, spread at most 0.25%, order-book imbalance against the trade no worse than −0.5, and crypto 24-hour volume at least $3M. Geometry scores are capped at 85. The reversal-or-continuation label is descriptive only.

How PatternX Classic evaluates it

Because the pattern has no bias, the context factors of the PatternX Score do much of the judging in PatternX Classic. The score has 13 deterministic factors, and detector geometry plus a fixed prior for the pattern account for 24% of the weight. Trend alignment (EMA20 against EMA50) and multi-timeframe agreement ask whether the side that broke matches the trend; breakout strength and volume ask how forceful the break was. A live historical win rate for that exchange and symbol counts once there are at least 10 closed trades; until then that factor stays neutral.

The resulting confidence (Elite 80+, High 65+, Medium 50+, Weak below 50) must reach your minimum, 65 by default. The opposite-line stop and the measured target are combined with your stop-loss and take-profit, signals below 1:1 reward:risk are dropped, and signals expire after 15 minutes. The risk controls apply before every order.

What PatternX AI adds

On Pro and Elite, PatternX AI adds a context check after a symmetrical-triangle setup has been detected and accepted. It doesn't detect the triangle or choose its direction; the breakout already did that. Instead it asks whether the context agrees with the side that broke. Trend strength (EMA12 against EMA26) and confluence, the two most heavily weighted of its eight fixed factors, are especially informative for a pattern without a bias, alongside market structure, volume quality and momentum. The layer can veto the setup or re-weight its confidence, and it leaves stops, targets and size untouched. On Pro, AI mode allows up to 5 trades a day. See the AI context layer.