Chart Patterns
The named shapes traders actually use - head and shoulders, double tops and bottoms, triangles, flags, wedges, cup and handle. How each forms, the psychology underneath, the confirmation rules, and how to avoid pattern-matching yourself into a bad trade.
Chart patterns are the named shapes that recur on charts because human behavior recurs. A head and shouldersHead and shouldersA three-peak reversal pattern. Middle peak (head) is highest; flanked by two lower peaks (shoulders). Confirms on neckline break.Read in glossary → in 1970 looks like one in 2026 for the same reason old chess traps still work: the psychology hasn't changed. Patterns compress everything you've learned so far - trends, support and resistance, breakouts, volume - into a single compound structure that carries its own bias. This lesson covers the two categories (continuation and reversal), the named patterns every trader needs to recognize, the specific confirmation rules for each, and the common traps that turn pattern-matching into a losing strategy.
The two categories
Continuation patterns
The existing trend paused, consolidated, and is expected to resume. The pattern describes the shape of the pause.
Examples: flags, pennants, ascending and descending triangles, symmetrical triangles (usually), rectangles, cup and handle.
Reversal patterns
The existing trend is showing signs of exhaustionExhaustionThe aggressive side of a move running out of fuel. New extremes print on lower volume and lower delta.Read in glossary → and is expected to reverse. The pattern describes the shape of the turn.
Examples: head and shoulders (and inverse), double tops and bottoms, triple tops and bottoms, rising and falling wedges, rounding tops and bottoms.
One pattern can appear in both categories depending on context - a rectangle in a trend is continuation; a rectangle at the top of an extended rally is often a top. Context matters more than the shape itself.
Reversal patterns
Head and shoulders
The most famous reversal pattern in trading, appearing at tops.
Structure:
- Left shoulder - an initial peak after the trend; price pulls back.
- Head - a higher peak than the left shoulder; price pulls back again.
- Right shoulder - a peak lower than the head, ideally near the height of the left shoulder.
- Neckline - a line drawn connecting the lows between the two pullbacks.
Confirmation: price breaks below the neckline on expanding volume.
Psychology: buyers push to a higher high (head), fail to sustain the new peak, and can only manage a lower high on the next attempt. Demand is drying up.
Measured move target: project the distance from the top of the head to the neckline downward from the breakdown point.
Inverse head and shoulders - the mirror at a bottom. Same rules, inverted.
Double top and double bottom
Simpler reversal patterns based on two failed attempts at a level.
Structure (double top):
- Price makes a high in an uptrend, pulls back.
- Price rallies to approximately the same high, fails to exceed.
- Neckline = the pullback low between the two peaks.
Confirmation: price breaks below the neckline (double top) or above the neckline (double bottom) with volume.
Psychology: twice-tested resistanceResistanceA price level where sellers have historically stepped in with size. Acts as a ceiling until it breaks.Read in glossary → holding - the second attempt failed on less conviction than the first. Sellers defending effectively.
Triple top / bottom - three tests instead of two. Rarer but stronger - three failed attempts signal deeper exhaustion.
Rising wedge (bearish)
Structure: upward-sloping, converging trendlines. Both highs and lows rise, but the highs rise slower than the lows - the channel narrows.
Appearance: most commonly at a top as a slow-rolling reversal sign, or as a counter-trend rally inside a larger downtrend.
Confirmation: break of the lower trendline on volume.
Psychology: buyers are still pushing but doing so with shrinking ambition; each up leg accomplishes less than the last.
Falling wedge (bullish)
Structure: downward-sloping, converging trendlines. Lows fall slower than highs, channel narrows.
Appearance: often at a bottom or as a pullback within a larger uptrend.
Confirmation: break of the upper trendline on volume.
Psychology: sellers losing steam - each down leg is shorter than the last.
Continuation patterns
Flag and pennant
The two most common continuation patterns. Near-identical in psychology; different in shape.
- A sharp, near-vertical move (the flagpole).
- Brief consolidation between parallel trendlines (the flag), sloped against the direction of the pole.
Structure (pennant): same as flag but consolidation is a small symmetrical triangle instead of a parallel channel.
Confirmation: break of the flag/pennant in the direction of the flagpole, on expanding volume.
Psychology: a fast move attracted profit-taking and contrarian fading. The dip is absorbed, weak hands shake out, and the trend resumes with stronger hands holding.
Measured move target: length of the flagpole projected from the breakoutBreakoutPrice closing decisively through a resistance level on expanding volume. Often followed by retest and continuation.Read in glossary → point.
Ascending triangle
Structure: flat resistance at the top, rising support at the bottom, converging to a point.
Bias: bullish continuation.
Psychology: each pullback is shallower than the last - buyers stepping in earlier every time. Sellers hold a fixed defensive line at the resistance, but eventually their size runs out.
Confirmation: break above the flat resistance on volume. Target: height of the triangle projected upward.
Descending triangle
Structure: flat supportSupportA price level where buyers have historically stepped in with size. Acts as a floor until it breaks.Read in glossary → at the bottom, falling resistance at the top.
Bias: bearish continuation (usually).
Psychology: each rally makes a lower high - buyers getting less aggressive. Sellers defending a fixed line, knowing the eventual break will run.
Symmetrical triangle
Structure: both highs and lows converge toward a point. Neither side decisively dominates.
Bias: neutral until break. Direction follows the break; most often continues the prior trend.
All three triangles share a working rule: the break in the final third of the triangleTriangleConverging trendlines forming a symmetrical, ascending, or descending shape. Usually resolves in direction of prior trend.Read in glossary → is most reliable; breaks too early are often fakeouts. And the apex (where the trendlines meet) is a dead zone - by the time price reaches the apex without breaking, the pattern has lost its tension.
Rectangle
Structure: price oscillates between parallel horizontal support and resistance.
Bias: depends on prior trend. In an uptrend, often a continuation pattern (bullish rectangle); in a downtrend, bearish. At extremes of extended trends, can be a distribution/accumulation zone instead.
Confirmation: break of either level on volume.
Trading approach: many traders fade the range (buy near support, sell near resistance) while the pattern is intact, then flip to trading the eventual breakout.
Cup and handle
A longer-horizon bullish continuation pattern popularized by William O'Neil (CAN SLIM methodology).
Structure:
- Cup - a rounded, U-shaped base spanning weeks or months. Price declines, bottoms smoothly, recovers to the prior high.
- Handle - a small, short-duration pullback near the prior high. Usually shallow (no more than ~1/3 of the cup's depth).
Confirmation: break above the high on volume.
Psychology: extended accumulation in the cup clears out weak hands and builds a durable base; the brief pullback (handle) is one more stop-hunt / distribution check before the advance.
Inverse cup and handle - bearish variant. Rare but valid.
The reliability hierarchy
Not all patterns are equal. Rough ranking from most- to least-reliable (approximate, context-dependent):
- Head and shoulders (both directions) - well-defined, widely watched.
- Double tops and bottoms - simple, confirmable, frequent.
- Flags and pennants - high base rate, clear continuation bias.
- Ascending / descending triangles - strong psychology, clear levels.
- Cup and handle - slow to form but durable when it does.
- Symmetrical triangles - neutral; reliability depends on breakout confirmation.
- Rectangles - valid but generic; depends heavily on context.
- Wedges - real but often misidentified; purists only.
- Rounding tops/bottoms - slow, subtle; hindsight-prone.
- Exotic patterns (diamond, three drives, gartley, etc.) - declining marginal value.
A working trader is well-served focusing deeply on the top five. Exotic patterns have fans, but the signal/noise degrades rapidly down the list.
Volume behavior by pattern stage
Every pattern has a characteristic volume profile. Violating it is a red flag.
| Pattern | Formation | Confirmation |
|---|---|---|
| Head and shoulders | Volume fades on each peak | Expands on neckline break |
| Double top/bottom | Second peak/trough on lower volume than first | Expands on neckline break |
| Flag/pennant | Volume contracts during consolidation | Expands on the break |
| Triangles | Volume contracts toward apex | Expands on breakout |
| Cup and handle | Light volume through cup; very light through handle | Expands on breakout of prior high |
| Wedges | Volume diverges against the wedge's slope | Expands on break |
A pattern that "looks right" but has the wrong volume character is usually broken. Trust volume over geometry.
The confirmation rules that matter
Every pattern entry should pass these five checks. Skip any and reliability drops sharply.
- Is the pattern fully formed? - Wait for the final leg. Patterns in progress are guesses.
- Is there a clear confirmation level? (neckline, flag top, triangle edge) - If you can't name it, you can't trade it.
- Did volume confirm? - Most patterns specify volume expansion on the break.
- Is the higher timeframe aligned? - A daily-chart pattern against a weekly-chart opposite trend is the weakest setup.
- Is the measured-move target realistic? - If the target sits exactly on the next major resistance, the setup has less room.
Five-of-five yes = high-quality setup. Three-of-five or fewer = pass.
Common misreads
Seeing patterns where none exist
The pareidolia trap. The brain pattern-matches aggressively; if you look for head and shoulders, you'll see one on every chart. Discipline: patterns are there only if they meet formation rules precisely, not if they "sort of look like" one.
Trading before confirmation
Entering inside a triangle before the break, or after the right shoulder but before the neckline break, is trading on prediction, not signal. The confirmation bar is the trade - before it, there is no pattern yet.
Ignoring the prior trend
A "bullish continuation" pattern in a downtrend is not bullish. Continuation patterns continue the prior trend. Context selects the right bias.
Forgetting volume
Every canonical pattern has a volume specification. The pattern without correct volume is a failed pattern, not a late one.
Measured-move worship
The measured-move target is a guideline, not a rule. Price often exceeds it; often falls short. Scale out along the way; don't hold to the exact number.
A worked pattern read
AAPL daily chart, last 4 months.
- Prior trend: uptrend from $170 to $200 over three months.
- Last 3 weeks: a clear symmetrical triangle between $198 resistance (lower highs) and $188 rising support.
- Volume: contracted from 80M shares average to 45M during the triangle.
- Apex zone approaching next 5 sessions.
- Higher timeframe: weekly uptrend still intact, well above rising 50-week MAMoving averageThe average price over the last N bars. Used as dynamic support/resistance and trend filter. EMA weights recent data heavier.Read in glossary →.
Read: continuation triangle in an established uptrend. Bias is breakout up. Volume contraction during formation is textbook. Apex coming soon, so a resolution is expected within 1-2 weeks.
Plan:
- Long entry on break above $198 with volume expansion (>75M shares).
- Stop at $186 (below triangle lows).
- Target 1: measured move up = $208. Target 2: prior swing high $215 if momentum continues.
- Invalidation: break below $188 on volume - flip to reversal-triangle thesis.
That's the whole workflow. Pattern identified, psychology understood, volume behavior checked, confirmation defined, trade planned.
Common questions
Which patterns have the best statistical edge? Flags and pennants (in established trends) and head-and-shoulders (at clean trend ends) have the most consistent positive expectancyExpectancyExpected R-multiple per trade: (WinRate × AvgWinR) − (LossRate × AvgLossR). Positive = edge. Negative = bleed.Read in glossary → in published studies. But studies often cherry-pick clean examples; real-world execution matters at least as much as pattern choice.
Can I trade patterns on intraday timeframes? Yes, with caveats. The patterns are identical; reliability is lower on shorter timeframes because the samples are noisier. Intraday patterns work best when aligned with a higher-timeframe pattern or level.
What about harmonic patterns (Gartley, Bat, Crab)? They have a following, but they rely on precise Fibonacci measurements that degrade quickly under real-market noise. Not essential for foundational pattern trading. Explore them later if you're curious; don't start there.
Do patterns work in crypto? Yes - arguably better than in equities at times, due to reflexive retail participation and round-the-clock trading. Same rules, different volatility.
How long should I wait for a pattern to confirm? As long as the pattern's structure requires. A daily H&S might take weeks to complete. A 5-minute flag completes in 20-30 minutes. Don't try to hurry it. If the confirmation doesn't come, there's no trade.
What if I see multiple patterns simultaneously? Go with the higher-timeframe pattern. A daily double top overrides a 5-minute flag. If they're aligned (same direction), treat it as confluence. If they contradict, the higher timeframe wins.
Key takeaways
- Chart patterns are named shapes that recur because trader psychology recurs. Same setups; same outcomes.
- Two categories: continuation (trend resumes) and reversal (trend turns). Context - prior trend - selects which bias applies.
- The working set: head and shoulders, double top/bottom, flags and pennants, ascending/descending/symmetrical triangles, cup and handle. Master these before exotic patterns.
- Every pattern has: a formation structure, a confirmation level (neckline / flag / triangle edge), a required volume behavior, and a measured-move target.
- Volume behavior is non-optional. Patterns with wrong volume are broken, not late.
- Five-point confirmation: pattern complete, clear confirmation level, volume confirmed, higher timeframe aligned, realistic target.
- Trading patterns before confirmation is prediction, not signal. The confirmation is the trade.
- Pareidolia is the enemy. Patterns exist when they meet formation rules precisely - not "sort of."
- Say the full sentence: "This is a ___ pattern forming in a ___ trend, on a ___ chart, with volume ___, and confirmation would be ___." Clean sentence = real pattern.
- Combine pattern with level context (from the S/RR-multipleThe dollar amount risked on a trade. Every outcome is measured in R: a 2R winner made twice the risked amount.Read in glossary → lesson) and trend context (from the trends lesson) for highest-quality setups.
Up next: Volume Analysis - the other half of every pattern's confirmation. What volume actually measures, the four volume-price relationships, and how to read it without fooling yourself.
Related lessons
Technical Analysis Foundations
Why studying the chart works (and when it doesn't), the three Dow principles every trader still uses, the split between price action and indicators, and the honest limits of reading past price to forecast future price.
Price Action Fundamentals
The seven building blocks of pure price reading - candles, support and resistance, trends, gaps, breakouts, chart patterns, and volume. How each one carries signal, the common misreads, and the working order you should learn them in.
Reading Candlesticks
Body, wick, open, close - and every named pattern built from them. Doji, pin bar, engulfing, hammer, morning star, three white soldiers, and the reading framework that stops you from trading isolated patterns without context.
