Get 20% off all challenges

Pennant Pattern: How to Trade Bull and Bear Pennants

Pennant pattern explained: bull pennant and bear pennant rules, pennant vs flag vs symmetrical triangle, Bulkowski's statistics and a worked two-target trade.

Gianluca Pizzituti••14 min read
Share article
Pennant Pattern: How to Trade Bull and Bear Pennants

One account, every market we offer

Explore all markets

A pennant pattern is a short continuation pattern made of a sharp price move, called the flagpole, followed by a brief consolidation between two converging trendlines that looks like a small triangle. The pennant pattern is meant to mark a pause in the middle of a move, with price leaving the triangle in the same direction it entered.

It is one of the first patterns momentum traders learn, and one of the most over-trusted. Thomas Bulkowski's data on more than 1,600 pennants shows that the textbook target is reached far less often than most guides imply. This article covers the bull pennant and the bear pennant, how a pennant differs from a flag, a symmetrical triangle and a wedge, and how to plan the trade, with a worked Solana example.

Quick answer: A pennant pattern is a continuation pattern: a steep flagpole, then a small symmetrical triangle that forms on shrinking volume and lasts no more than about 3 weeks on a daily chart. A bull pennant follows a sharp rally and breaks upward. A bear pennant follows a sharp drop and breaks downward. Traders enter on a close outside the pennant, place the stop on the other side, and project the flagpole's length for a target.

Highlights of this article

  • A pennant is a steep flagpole followed by a small triangle with converging lines and falling volume
  • A bull pennant follows a rally; a bear pennant follows a drop; the rules are mirror images
  • Converging lines make it a pennant, parallel lines make it a flag, and more than about 3 weeks makes it a triangle
  • In Bulkowski's data, only 35% of upward and 32% of downward breakouts reached the full flagpole target
  • Tight pennants beat loose ones, and a two-stage target fits the statistics better than one distant objective

The 3 parts of a pennant

Every pennant has the same three components, in the same order.

1. The flagpole. A fast, near-vertical move on heavy volume. StockCharts' ChartSchool says that without a sharp advance or decline the formation's reliability becomes questionable. Bulkowski asks for a flagpole that is unusually steep and lasts several days on a daily chart. A slow, grinding trend is not a flagpole.

2. The pennant. A small consolidation where each high is lower than the last and each low is higher, so the two trendlines converge. StockCharts notes the slope is usually neutral. Volume contracts while it forms. Bulkowski found a downward volume trend in 86% of the pennants he studied.

3. The breakout. A candle closes outside one of the two lines. In a textbook pennant, that close is in the direction of the flagpole and volume expands again.

The logic: after a sharp move, traders who caught it take some profit and traders who missed it wait for a better price. Neither group is large enough to reverse the move, so the range narrows until the original side pushes again. The statistics below show how often practice matches the theory.

Bull pennant

A bull pennant forms after a sharp rally. Price stalls, prints a series of lower highs and higher lows, and volume fades. The pattern completes when a candle closes above the upper trendline.

What to check:

  • The rally before it stands out on the chart. It often starts with a break of a prior high, which is the setup described in the breakout trading guide.
  • The pennant holds in the upper part of the flagpole.
  • The pennant is flat or tilts slightly down, against the rally. Bulkowski found that performance suffers when the pennant slopes in the direction of the prevailing trend.

Bear pennant

A bear pennant is the same structure upside down. A sharp drop forms the flagpole. Price then bounces weakly, with lower highs and higher lows squeezing into a point, on declining volume. The pattern completes when a candle closes below the lower trendline.

In Bulkowski's data, the average decline after a downward breakout was 6%, against an average rise of 7% after an upward breakout, so the two directions performed about the same.

One caution on direction. Across all pennants in Bulkowski's sample, 57% broke upward, so a large minority broke down. Wait for the close. Do not buy inside a bull pennant on the assumption that it will resolve the right way.

A bitcoin price chart climbing steeply on a dark trading screen
The flagpole comes first: a steep, obvious move that stands out from the candles before it.

Pennant vs flag vs symmetrical triangle vs wedge

These four patterns are all consolidations drawn with two trendlines. Two questions separate them. Are the lines parallel or converging? And how long did the pattern take to form?

Pattern Trendlines Needs a flagpole? Typical length on a daily chart Read as
Pennant Converging, roughly symmetrical Yes 3 weeks or less Continuation
Flag Parallel, tilted against the trend Yes Less than 3 weeks Continuation
Symmetrical triangle Converging, symmetrical No Longer than a pennant Breakout either way, often with the trend
Wedge Converging, both sloping the same way No Weeks to months Often a reversal

Pennant vs flag. The flagpole is identical. The difference is the consolidation. A flag is a small channel with parallel lines that drifts against the trend. A pennant is a small triangle whose lines meet. Both are traded with the same logic. The flag's channel rules and entries are in the bull flag pattern guide.

Pennant vs symmetrical triangle. A pennant is a small symmetrical triangle with a flagpole attached. Size and time are the difference. Bulkowski draws the line at 3 weeks: patterns longer than that are symmetrical triangles or wedges. A triangle is a larger structure that does not need a sharp move before it, and its target is measured from its own height. The triangle patterns guide covers the symmetrical, ascending and descending types.

Pennant vs wedge. In a wedge, both lines slope in the same direction, up or down. In a pennant, one line falls and the other rises. A rising wedge after a rally is usually read as a warning, not a continuation. See the wedge patterns guide.

For where the pennant sits among the other continuation and reversal formations, the chart patterns guide has the full map.

What the statistics say

Bulkowski's pennant figures were last updated in August 2020 and cover more than 1,600 pennants in US stocks during bull markets. He measures the short-term swing after the pattern, not the move to an ultimate high or low.

Measure Pennants (up / down breakouts) Flags (up / down breakouts)
Break-even failure rate 54% / 54% 44% / 45%
Average rise or decline 7% / 6% 9% / 8%
Reached the full flagpole target 35% / 32% 46% / 46%
Upward breakouts 57% 60%

Three points stand out.

  1. More than half barely move. Bulkowski's break-even failure rate counts patterns that fail to rise or fall at least 5% after the breakout. For pennants it is 54% in both directions.
  2. The full target is the exception. Bulkowski notes that the trend after the pennant equals or exceeds the trend before it just 30% of the time. The idea that a pennant marks the halfway point of a move holds in a minority of cases.
  3. Flags tested better. On every measure in the table, flags outperformed pennants in his data.

None of this makes the pennant untradable. A plan that needs the full flagpole projection is relying on the one-in-three case. A plan that takes some profit early and lets the rest run fits the numbers. These are daily US stock figures, so test your own market and timeframe.

A 5-point check before you trade a pennant

  1. Is the flagpole obvious? It should be among the steepest moves on the visible chart. This is a momentum trading setup.
  2. Is the pennant tight? Bulkowski found that tight pennants perform better than loose ones. A loose pennant is one where price meanders, pokes outside the trendlines or looks jagged.
  3. Is it short? On a daily chart, 3 weeks or less. On other timeframes, a working rule is about 15 candles or fewer. That translation is a convention, not a tested figure.
  4. Did volume fall? Volume should shrink through the pennant and return on the breakout.
  5. Is the tilt neutral or against the trend? A pennant that leans in the direction of the flagpole tested worse.

If two or more answers are no, skip it.

How to trade a pennant pattern

Entry. Wait for a candle to close outside the pennant in the direction of the flagpole. A wick through the line is not a breakout. Stop orders gather just beyond obvious lines, and price often spikes through and returns. The liquidity sweep guide explains how to tell that spike from a real break.

Stop. The tighter stop goes beyond the last swing inside the pennant, for a bull pennant the most recent higher low. The wider stop goes beyond the far side of the whole pennant. Because a pennant narrows, the tight stop is often very close, which is the pattern's main attraction. The stop loss guide covers how to size from either.

Targets. Two methods exist, and they give very different numbers.

  • The classic measured move. Take the flagpole's height and project it from the breakout point. StockCharts describes this method for flags and pennants.
  • Bulkowski's adjusted measure. Multiply the flagpole's height by the share of pennants that reach the full target, 35% for upward breakouts, and add the result to the bottom of the pennant. For downward breakouts, use 32% and subtract from the top of the pennant.

A two-stage plan uses both: the adjusted figure as a first target and the classic projection as a second.

Worked example: a bull pennant on Solana

Prices are illustrative. On the 1-hour chart, Solana breaks above a prior high and rallies from 180 to 204 in 6 candles on heavy volume. The flagpole is 24 dollars.

Over the next 10 candles, price prints highs of 204, 203.1 and 202.4 and lows of 198, 199.2 and 200.1. The lines converge, the pennant holds in the top quarter of the flagpole, and volume falls each hour. The eleventh candle closes at 202.9, above the upper line, on the highest volume since the rally.

Item Value How it is derived
Entry 202.9 Close above the upper trendline
Stop 199.6 Below the last higher low at 200.1
Risk per SOL $3.30 202.9 minus 199.6
Target 1 206.4 35% of 24 is 8.4, added to the pennant low of 198
Target 2 226.9 The full 24 added to the breakout at 202.9
Reward to risk at target 1 About 1.1R 3.5 divided by 3.3
Reward to risk at target 2 About 7.3R 24 divided by 3.3
Size at $150 risk 45.4 SOL, about $9,200 notional 150 divided by 3.3

The plan: sell half at target 1, move the stop on the rest to the entry price, and hold the second half for target 2.

Three outcomes cover most cases:

  • Stopped out. Price falls back through 199.6. Loss: $150.
  • Target 1 only. Half is sold at 206.4 for about $79. The rest is stopped at the entry price for nothing. Result: about $79.
  • Both targets. About $79 on the first half and about $545 on the second. Result: about $625.

With only about a third of pennants reaching the full projection, the middle outcome will be common. It is small, and it is positive. Whether the mix is worth trading depends on your own win rate, which the risk to reward guide shows how to work out.

A bear pennant in brief

GBP/USD drops from 1.3100 to 1.2980 in one morning, a 120-pip flagpole. It bounces to 1.3012, then prints a lower high at 1.2998 and a higher low at 1.2986. A 15-minute candle closes at 1.2969, below the lower line.

The short entry is 1.2969. The stop goes above the last lower high, at 1.3003, for 34 pips of risk. The classic projection subtracts 120 pips from the breakdown and gives 1.2849, about 3.5R. A first target at 1R, 1.2935, plays the role of target 1 above.

A laptop showing a candlestick chart in front of a large bright screen with a falling price chart
Draw the two lines first, then wait for a candle to close outside them before acting.

Why pennants fail

  • No real flagpole. A small triangle after an ordinary move is noise. Without the sharp move, there is no momentum to continue.
  • Breakout without volume. A close outside the line on thin volume often reverses within a few candles.
  • The apex. If price drifts all the way to the tip of the triangle without breaking, skip it.

Using the pennant in a prop firm challenge

Pennants suit evaluations because the stop is close and defined. The risk is the fast conditions around them, where it is easy to enter larger than planned.

Velotrade runs simulated, educational evaluations across crypto, forex, stocks, index ETFs and commodities on DXtrade. On a $50,000 PRO 1-Step account, the daily loss limit is 3% and the maximum drawdown is 3%. The drawdown is static, so the floor is fixed at $48,500 for the life of the account. See static maximum drawdown explained for how that works.

The Solana example risked $150, or 0.3% of the account. At that size, the tenth full loss in a row is the one that reaches the floor. Failed pennants in a row are normal, so risk per trade has to be small against a 3% limit. Funded traders keep an 80% profit split.

This article is educational and is not investment advice. A pennant describes a pause. The stop and the size are the parts you decide.

Sources

  • Thomas N. Bulkowski, Encyclopedia of Chart Patterns (Wiley), with statistics published on ThePatternSite: pennants and flags: identification guidelines, failure rates, average moves, share meeting the price target, the measure rule and the tight versus loose finding. Statistics updated 27 August 2020.
  • Thomas N. Bulkowski, ThePatternSite glossary: definition of the break-even failure rate.
  • StockCharts ChartSchool, "Flag, Pennant": the sharp move requirement, pennant shape, volume behaviour and the flagpole target method.

Frequently Asked Questions

About the author

Gianluca Pizzituti

Gianluca Pizzituti

Chief Executive Officer

Formerly on the derivatives desk at Dresdner Kleinwort in London, then founded and ran a proprietary HFT firm in FX and equity indices out of Singapore.

Velotrade was founded in Hong Kong in 2016 by a team from JP Morgan, Dresdner Kleinwort and Bank of America, has facilitated over $2.5 billion in trade-finance payments, and has been featured in Bloomberg, the Financial Times, the Wall Street Journal and on Nasdaq. Its prop trading evaluations are simulated and educational.

View author page

Ready to trade with
$200,000 capital?

Up to 90% profit split

Keep most of what you earn

No consistency rule

News trading allowed

Payouts within 24 hours

Once approved