Chapter 6 - Continuation Patterns

{fullwidth}

Technical Analysis of the Financial Markets — John J. Murphy

Macro Overview & Strategic Value

Chapter 6 completes the pattern-recognition toolkit by covering the second category from Chapter 5’s framework: continuation patterns, which signal a pause in the prevailing trend rather than its reversal. Murphy’s core thesis is that these patterns are shorter-duration, less consequential interruptions — triangles, flags, pennants, wedges, and rectangles — that resolve, more often than not, in the direction of the pre-existing trend, giving traders a mid-trend confirmation and re-entry framework rather than a reversal-detection one.

The strategic value for a practitioner is dual: first, correctly reading a continuation pattern lets a trend-follower add to or re-enter a position mid-trend with a defined risk/reward setup (using the same height-based measuring logic from Chapter 5); second, recognizing that categories are “usually” but not always accurate — a triangle can act as a reversal, a head and shoulders can act as a continuation — reinforces the book’s recurring caution against rigid rule application without corroborating volume and structural evidence.

The chapter closes by formally introducing confirmation and divergence, a conceptual pair that Murphy flags as recurring “again and again throughout the book” — meaning this short closing section is actually a preview of a much larger analytical framework (multi-indicator agreement) used extensively from Chapter 7 onward.

Core Concepts & Mechanics

• Triangle taxonomy (symmetrical/ascending/descending) — Symmetrical triangles are directionally neutral continuation patterns (resolve with the prior trend); ascending and descending triangles carry inherent bullish/bearish bias regardless of where they appear, since they reflect an explicit imbalance between buyers and sellers.
• Triangle apex time rule — Breakout should occur between 2/3 and 3/4 of the horizontal distance to the apex; if price drifts past that window without breaking out, the pattern loses forecasting reliability, giving traders an explicit time-based invalidation rule beyond just price.
• Broadening formation (megaphone top) — A rare, diverging (not converging) pattern with expanding volume and price swings, reflecting an emotionally out-of-control market; almost always bearish and appears near major tops, making it a red-flag pattern rather than a tradeable setup.
• Flags and pennants — “half-mast” measuring rule — Both require a steep prior “flagpole” move and form near the move’s midpoint; the subsequent leg after breakout is expected to roughly duplicate the flagpole’s length, giving an immediate reward projection from a highly reliable, low-ambiguity pattern.
• Wedge slope-direction rule — A wedge that slants against the prevailing trend (falling wedge in an uptrend, rising wedge in a downtrend) is the expected continuation form; critically, a rising wedge is always bearish and a falling wedge always bullish, regardless of whether it appears mid-trend or at a reversal point — a rare pattern where the interpretation rule never flips.
• Rectangle range-trading duality — Traders can either await a confirmed breakout (trend-following approach) or trade the range itself (buying support, selling resistance) since well-defined horizontal boundaries create low-risk entries — but this range-trading tactic works only until the eventual breakout, at which point the trader must reverse rather than fade it.
• Measured move (swing measurement) — A well-defined trend is assumed to divide into two roughly equal legs separated by a corrective wave retracing 1/3-1/2 of the first leg; projecting the first leg’s height from the correction’s endpoint gives a target for the second leg — directly extending the retracement logic from Chapter 4.
• Continuation head and shoulders — An upside-down version of the reversal pattern from Chapter 5 (middle trough lower than shoulders in an uptrend, or middle peak higher in a downtrend), structurally distinct enough to avoid confusion with the reversal variant despite superficial similarity.
• Confirmation and divergence — Confirmation means multiple indicators/signals agree on direction; divergence means they disagree — introduced here as a chart-pattern concept but flagged as a book-wide analytical principle that recurs through every later indicator chapter.

Technical Terminology & Reference Table

Term	Operational Definition
Symmetrical triangle	Converging trendlines (one descending, one ascending); neutral continuation pattern
Ascending/descending triangle	Flat line + sloped line; inherently bullish/bearish regardless of position in trend
Apex	Point where a triangle’s two trendlines converge; also acts as support/resistance post-breakout
Broadening formation	Diverging trendlines with expanding volume; rare, bearish, found at major tops
Flag	Small parallelogram sloping against the trend, following a steep “flagpole” move
Pennant	Small symmetrical-triangle-shaped consolidation following a flagpole move
Wedge	Converging trendlines with a pronounced slope; rising = bearish, falling = bullish
Rectangle	Sideways trading range between two horizontal lines; also called a trading range or Dow Theory “line”
Measured move	Pattern where a trend’s second leg duplicates the first leg’s size after a corrective retracement
Confirmation	Multiple technical signals agreeing on the same directional conclusion
Divergence	Technical signals failing to agree; treated as an early reversal warning

The Author’s Market Philosophy

Murphy continues to treat chart patterns as recordings of crowd psychology rather than arbitrary geometry — the ascending/descending triangle’s fixed directional bias reflects a genuine, observable imbalance between buyer and seller aggression, not a coincidence of shape. His edge-generation model here emphasizes volume as the tie-breaker whenever pattern classification is ambiguous (e.g., is this rectangle a continuation or a hidden triple top?) — he consistently defers to “which side has heavier volume” as the deciding behavioral signal over pure pattern shape. He explicitly builds a probabilistic, exception-tolerant framework — repeatedly using “usually” rather than absolute rules — reflecting an assumption that markets are self-similar but imperfectly so, and that a competent technician must corroborate any single pattern with time constraints, volume behavior, and now, cross-indicator confirmation, rather than trading a single signal in isolation.

Systemic & Portfolio Integration

The confirmation/divergence framework introduced at the chapter’s close becomes the connective tissue for the book’s later oscillator, volume, and multi-indicator chapters — meaning every subsequent systematic signal is meant to be filtered through this same agreement/disagreement lens. The measured-move and half-mast flag/pennant projections extend Chapter 5’s height-based expectancy calculations into mid-trend continuation setups, giving trend-following and momentum systems explicit reward targets for adding to winning positions rather than only for initial entries.

Important Formulas, Data, or Initial Examples

• Triangle apex timing: breakout expected between 2/3 and 3/4 of the pattern’s horizontal width (e.g., a 20-week-wide triangle should resolve between weeks 13-15); Dell example: an 18-week triangle broke out in week 13, just past the two-thirds mark.
• Triangle height projection: measure the vertical base height and project it from the breakout point (alternative: draw a parallel channel line from the top of the base).
• Flag/pennant “half-mast” rule: measure the flagpole (prior move from original breakout to pattern start) and project that same distance from the flag/pennant’s breakout point.
• Rectangle height projection: measure top-to-bottom range height and project it from the breakout point.
• Measured move worked example: a prior upleg (AB) of 20 points, with the correction (BC) bottoming at 62, projects a target of 82 for the second leg (62 + 20 = 82).
• Pattern duration guidance: triangles, wedges, and rectangles typically run 1-3 months (intermediate); flags/pennants run only 1-3 weeks (shorter in downtrends, often 1-2 weeks).

Active Recall Evaluation

1. Why do ascending and descending triangles carry a fixed directional bias regardless of where they appear in a trend, while the symmetrical triangle does not?
2. Explain the significance of the two-thirds to three-quarters apex rule — what happens to the pattern’s reliability if price fails to break out within that window, and why?
3. Why is a rising wedge always interpreted as bearish and a falling wedge always bullish, even though wedges usually function as continuation patterns?
4. How does the “half-mast” concept for flags and pennants connect back to the retracement principles introduced earlier in the book?
5. What is the practical risk in treating a rectangle purely as a range to fade (buying support, selling resistance) rather than waiting for a confirmed breakout?

Answer Key (spoiler)

1. The ascending and descending triangles are built from one flat trendline and one sloped trendline, which directly encodes a real, ongoing imbalance between buyer and seller aggression (rising lower line = buyers stepping in more eagerly at ever-higher prices); the symmetrical triangle’s two converging, opposing slopes reflect balanced indecision, so its resolution direction depends entirely on the pre-existing trend rather than any inherent bias in the pattern’s shape.
2. If price remains inside the triangle beyond the three-quarters mark, the pattern is considered to be losing its potency — the converging trendlines are running out of room, and price is more likely to simply drift out through the apex without a decisive directional move, meaning the setup no longer offers a reliable, tradeable signal.
3. Because a wedge’s sloped shape itself encodes a specific directional warning independent of context: a rising wedge shows a market advancing on progressively narrowing, weakening momentum (a bearish exhaustion signal) while a falling wedge shows a decline running out of downside momentum (a bullish exhaustion signal) — this makes the wedge one of the few patterns whose interpretation stays consistent whether it appears mid-trend or at a reversal point.
4. Flags and pennants typically appear at the halfway point of a market move, which echoes the same tendency of markets to retrace a third to a half of a prior trend discussed with percentage retracements — both concepts reflect the idea that trends pause for a proportionate corrective phase before resuming, letting a trader use the size of the already-completed leg to project the remainder.
5. Range-fading works only as long as the rectangle holds; if a breakout occurs, a trader still positioned according to the fading strategy (e.g., long near support, right before a downside breakdown) is now on the wrong side of a new trend and must not only exit immediately but reverse the position — meaning the fading approach carries the specific risk of being caught by surprise exactly when the pattern finally resolves.