Bullish Pennant — Small Pattern, Big Move?

The bullish pennant is the bull flag's younger sibling: the same steep pole, but instead of a rectangle the consolidation takes the shape of a small triangle — two converging lines, price squeezing like a spring. The story is tempting: "the tighter the spring, the harder the breakout." The pennant has a reputation as a day-trader favorite and a sign that you're only halfway through the move.

The numbers say otherwise. In Thomas Bulkowski's database (over 1,600 perfect trades), pennants have a 54% failure rate after an upside breakout — the majority(!) of confirmed signals don't even reach 5%. Average gain: 7%. The measure-rule target is hit in 35% of cases — two-thirds of trades never reach their goal. And the famous "half-mast" rule? It holds up about 30% of the time.

For comparison, flags: 44% failures, +9%, target in 46% of cases. The conclusion is simple and unromantic: flags are statistically better than pennants in every single column. A tighter spring doesn't mean a stronger breakout — it just means a tighter spring. Let's see what to do with that.

How to Identify a Bullish Pennant

Identification criteria per Bulkowski:

1. A flagpole is mandatory. An unusually steep, near-vertical up move lasting several days. Without a pole, converging lines are just a small symmetrical triangle — not a pennant.
2. Converging trendlines. The upper line follows declining highs, the lower one rising lows — the shape of a small symmetrical triangle sitting on top of the pole.
3. Three weeks, max. A longer pattern is a symmetrical triangle or a wedge — different patterns, different statistics. A pennant is a sprint, not a marathon.
4. Volume shrinks during the formation — in as many as 86% of cases. This is the pennant's most consistent trait; consolidation on rising volume should raise doubts about the classification.
5. Confirmation: a close outside the line. The breakout goes upward in 57% of cases — barely better than a coin flip.

An extra nuance from the data: slope matters. A pennant sloping with the trend (upward, in an uptrend) performs worse than one sloping against the trend or horizontally.

A distinction that saves money: a pennant without a pole is not a pennant. Converging lines after a slow, multi-week rise are a small symmetrical triangle — a two-sided pattern where the breakout direction is only decided after the fact. The pennant's statistics (and its measure rule) apply strictly to setups with a vertical impulse at the base.

What the Numbers Say (Bulkowski, Encyclopedia of Chart Patterns)

Definitions: failure rate — how often price, after a confirmed breakout, doesn't even travel 5% in that direction; average move — measured on the short swing (which is why pennants, like flags, don't get an overall rank in the catalog); target % — how often price reaches the measure-rule goal.

A separate measurement dismantles the "half-mast" legend. Average pole before a pennant: +19% in 11 days. Average continuation after a pennant: +14% in 10 days. Similar duration, clearly smaller extent — and the move after the pattern equals or exceeds the move before it only 30% of the time. A pennant more often hangs at two-thirds of the pole than at its midpoint.

The worst column, though, is the failure rate: 54%. Most confirmed breakouts from a pennant don't even travel 5%. That's not an edge — it's a slightly negative coin that needs to be improved through selection.

Where does this gap between the flag and the pennant come from, given that the two patterns look almost identical? A likely mechanism: the pennant narrows toward its apex, so it self-extinguishes volatility — a breakout from a very tight range can be technical, forced by geometry rather than a real return of demand. The flag keeps a constant trading range, and its breakout more often requires actual buyer commitment. This is an interpretive hypothesis, not a measurement — but the numbers it's trying to explain (54% vs 44% failures, 35% vs 46% target rate) are hard data.

The other thing worth seeing in the numbers: the pennant isn't a direction-forecasting pattern. An upside breakout 57% of the time is close to a coin flip. All the information a pennant carries boils down to "a pause inside a strong move is happening, resolution is near" — direction is decided only by the breakout, and nothing else.

How to Trade a Bullish Pennant (Measure Rule)

Entry. A close above the pennant's upper line. With a 54% failure rate, the retest-and-confirmation variant makes even more sense here than with flags — you give up part of the move but filter out some duds.

Stop. Below the low of the pennant (safer) or below the broken line with a buffer. The pattern is small, so the stop is naturally tight — that's its biggest practical advantage: even at a weak hit rate, the reward-to-risk ratio can still hold up.

Target — measure rule. The height of the swing from the start of the move to the top of the pole × 35%, added to the pennant's lower edge. Yes, just 35% — the full-flagpole projection lands in one out of three trades, so as a base-case target it's fantasy. A modest goal plus a tight stop is the only math that makes the pennant work.

Filters that improve results:

• A tight pennant beats a loose one — no meandering, no poking outside the lines.
• Slope against the trend or horizontal — a pennant sloping in the trend's direction hurts results.
• A flat base — a pennant sitting above a flat base foreshadows a bigger move.
• The pole must be fresh and furious. The more vertical and consistent the pole, the better — a pennant after a slow, choppy rise is a statistical false friend.

Position management. With an average move of +7% and a target hit rate of one in three, there's no room for "hold to the moon." Take part off at 1:1 and trail the rest with a tight stop behind the lows. If price returns inside the pennant after the breakout and closes below its low, the signal is broken — you're out. Trapped buyers from a broken pennant hand fuel to the other side anyway, so a break can be more valuable information than the pattern itself.

Time-based invalidation. A pennant that drags to the apex without a breakout loses its spring — breakouts right at the tip are statistically weaker across the whole family of converging patterns. And a consolidation beyond 3 weeks is already a symmetrical triangle: recalculate against its stats, because they're different.

Myth vs Measurement — the Spring That Doesn't Fire

Where does the popularity of a pattern with a 54% failure rate come from? First, a pennant looks clever: a narrowing consolidation suggests "energy building up," and the spring metaphor sells better than a table of failure rates. Second, a small shape is easy to find everywhere — on the M15, several pennants can be drawn every day, and the more drawings, the more hits to brag about. Third, narrative survivorship: pennants that took off go viral on screenshots; the ones that fizzled after 2% don't exist in collective memory.

The measurement says: the bullish pennant is the weakest link in the continuation family — worse than the flag across every column and successful in delivering its target only one time in three. If you have a choice, play the flag. If you play the pennant, play it tight: a small stop, a 35%-projection target, zero waiting for a miracle.

There is one thing the pennant's numbers don't invalidate, though: its value as a context marker. A tight pennant on fading volume after a vertical pole tells you momentum is still alive and supply hasn't taken control — even if the breakout trade itself is a bet on a modest few percent. Many experienced traders use pennants not as a signal, but as confirmation of trend strength, looking for entries elsewhere. That's a more honest use than believing in the spring.

And remember the disclaimer: Bulkowski's measurements are US stocks, daily timeframe, bull market — nobody has rigorously confirmed these numbers on crypto or intraday.

FAQ

What is the success rate of a bullish pennant? A 54% failure rate on an upside breakout, an average move of +7%, measure-rule target hit in 35% of cases (Bulkowski, n>1,600, US stocks, daily). Flags outperform in every column.

How is a pennant different from a symmetrical triangle? A pennant lasts up to 3 weeks and must sit on a steep, near-vertical pole. Converging lines without a pole, or a pattern longer than 3 weeks, is a symmetrical triangle — a different pattern with different statistics.

Does a pennant appear at the halfway point of a move? Only about 30% of the time does the move after the pattern match the move before it. On average: +19% for the pole vs +14% for the continuation. "Half mast" is a starting point for estimating a target, not a promise.

FAQ

Rafał — Krypto Bez Ściemy

Trader and founder of Strefa Tradingu. He’s been breaking crypto down on YouTube for years — no hype, no signals, with a focus on market structure and risk management.

kryptobezsciemy.plInstagramYouTubeKlub Zero