The Big Picture: Why Charts "Work"
The lines on a chart have no predictive power of their own. Charts are worth reading because the behavioral patterns of the participants behind those lines repeat.
Price Chart Candlesticks & Volume Composite Result of All Participants' Trades expectations, fear, stop-loss, profit-taking... 4 Reasons Behavior Repeats 1. Self-Fulfilling everyone believes & acts 2. Behavioral Bias human psychological habits 3. Institutional physical necessity of order splitting 4. Information Lag differences in reaction speed Order flow becomes skewed -> source of a probabilistic edge Thought Experiment If only one person in the world watched charts, nothing happens -> the lines have no magic When many participants watch the same line, concentrated action moves price -> this is the essence The essence of chart analysis: read the behavioral patterns of participants behind the lines, not the lines themselves
Charts work when "many participants recognize the same pattern and take similar actions." There are four mechanisms that satisfy this condition. Explore each one in the tabs below.
Mechanism 1: Self-Fulfilling Prophecy
The pattern didn't "come true" on its own. It became reality because everyone believed in it and acted accordingly.
Support Trader A "Buy here" Trader B "Buy here" Trader C "Buy here" Buy orders concentrate demand spikes at the same level -> price actually bounces 1 Price approaches support line 2 3 Bounce
The wider technical analysis textbooks are read, the more traders recognize the same support and resistance levels. It is not that "the prediction came true" — rather, "the prediction aligns behavior," and that alignment produces the outcome. This is the self-fulfilling prophecy mechanism.
Mechanism 2: Behavioral Bias (Human Tendencies)
Investors sitting on unrealized losses sell when price recovers. Those who missed a breakout panic-buy. These tendencies repeat.
Bought at peak Investor X: bought at $120 Enduring the loss -15%... "I'll sell when it recovers" Relief Selling "Finally back... sell!" -> the source of overhead resistance 1 2 3 3 Recurring Behavioral Patterns FOMO (Fear of Missing Out) new high -> panic-buy -> fuel for breakouts Loss Aversion holding unrealized losses -> source of overhead selling Panic Selling sharp drop -> panic-sell -> capitulation exhaustion pattern
As behavioral finance research (Kahneman & Tversky, 1979) demonstrates, humans react roughly twice as strongly to losses as to equivalent gains. This asymmetry manifests repeatedly as recognizable "shapes" on a chart. Even without anyone watching charts, these patterns persist as long as human psychology remains unchanged.
Mechanism 3: Physical Constraints of Large Investors
Buying $100 million all at once would spike the price against the buyer. So institutions split orders over many days. This constraint is one structural cause of trends.
Price Day 1 Day 2 Day 3 Day 4 Day 5 Day 6 Buy Buy Buy Buy Buy Buy Institution: $100M split over 6 days ~$17M/day -> gradual uptrend forms This structure emerges regardless of whether anyone is watching the chart
As the optimal execution theory of Almgren & Chriss (2000) shows, large orders are spread over time to minimize market impact. This physical constraint structurally generates part of what we call a "trend" — a mechanism that exists independently of technical analysis.
Mechanism 4: Information Takes Time to Spread
Even strong earnings don't trigger everyone's reaction at once. This time gap is one cause of momentum.
Strong Earnings Day 0 Pros & Institutions same day or next day algorithms & specialist analysts Day 0-1 Retail Investors react a few days later brokerage reports via investment blogs Day 3-5 Late Movers react 1-2 weeks later TV news social media buzz Day 7-14 Price Impact Momentum Price Time ->
As the Hong & Stein (1999) model shows, information does not reach all market participants simultaneously. This difference in diffusion speed manifests in prices as "momentum." Jegadeesh & Titman (1993) empirically demonstrated that a strategy of buying the past 3-12 month winners and selling losers generates significantly positive returns.
The Right Question to Ask About Charts
Ask not about the pattern name, but "does an asymmetric order flow follow this shape?"
✗ Wrong Mindset

"It's a double bottom, so buy"

"A golden cross means uptrend"

"A triangle breakout goes up"


Memorizing pattern names and trading mechanically when they appear. No hypothesis for why it should work.

✓ Right Mindset

"Does an asymmetric order flow follow this shape?"

"Will the participant behavior driving that order flow repeat this time?"


Re-read patterns as "hypotheses about the expected order flow."

Translating Patterns into Order Flow Hypotheses
Ultimately, edge is not determined by win rate alone. It is the product of gain when right × loss when wrong × frequency × costs. Treating patterns as "hypotheses" rather than certainties lets you use them without overconfidence or blind faith.