C.W.K.
Stream
Lesson 02 of 06 · published

Behavioral biases — loss aversion, herding, recency

~30 min · behavioral

Level 0Numeracy Apprentice
0 XP0/68 lessons0/14 achievements
0/100 XP to next level100 XP to go0% complete

The systematic ways your brain gets investing wrong

Daniel Kahneman and Amos Tversky's research (later expanded by Richard Thaler — three Nobel laureates between them) established that humans aren't rational decision-makers. We have systematic, predictable biases. In investing, these biases consistently cost money. Five big ones:

1. Loss aversion

Losses hurt about 2x as much as equivalent gains feel good. So a 50% loss followed by a 50% gain feels like a wash; the math says you're at 75% of where you started.

Investing implication: investors hold losing positions too long ("hoping it'll come back") and sell winning positions too early ("locking in gains"). The math says do the opposite — cut losers, ride winners. Loss aversion fights you on both.

2. Overconfidence

People consistently overestimate their abilities. About 80% of drivers think they're above-average drivers. Same with investors: most think they can beat the market, even though only ~20% of active funds do.

Implication: people pay for unnecessary active management, trade too often (one study found that retail traders who traded most underperformed the most), and ignore evidence that they should be passive.

3. Herding / FOMO

The instinct to follow the crowd. When everyone's piling into a stock, you feel pressure to join. When everyone's selling, you feel pressure to sell.

Implication: buying tops, selling bottoms. The crowd's behavior is exactly what drives bubbles and crashes — joining it consistently is buy-high-sell-low.

4. Anchoring

Decisions get anchored to arbitrary reference points. The price you bought a stock at becomes a psychological anchor — "I'll sell when it gets back to my purchase price." But the market doesn't care what you paid.

Implication: people refuse to sell losers above their purchase price even when current evidence says the company is broken. The anchor traps capital in losing positions.

5. Recency bias

We weight recent events more heavily than old ones, even when statistical evidence suggests we shouldn't. After a strong year, investors expect another strong year. After a crash, they expect another crash.

Implication: retail money flows into asset classes that just went up (mean-reverting strategies suffer). Asset bubbles get extended because recent gains feel like the new normal.

How to fight back — pre-commitment

You can't out-think these biases in real time. The fix is pre-commitment — making decisions ahead of time, when you're calm, that override emotional decisions in the moment.

  • Set asset allocation in advance. Rebalance mechanically. Don't react to month-to-month moves.
  • Use automatic contributions (dollar-cost averaging). Buying during crashes feels terrible; the rule says buy anyway.
  • Have a written investment policy. Review only when something material changes (life event, not market move).
  • Limit how often you check accounts. Daily checking amplifies recency bias and FOMO.

Behavioral biases are the human-evolution defaults. Investing requires deliberately overriding them.

The takeaway

Five major biases: loss aversion, overconfidence, herding/FOMO, anchoring, recency bias. Each costs money in predictable ways. Pre-commitment (rules set in advance) overrides emotional decisions in the moment. The fix isn't being smarter — it's being structurally less reactive. Lesson 10-3 covers the macro version of this: cycle psychology.

Exercise

  1. How does loss aversion lead to holding losers too long while selling winners too early?
  2. Why does overconfidence persist among retail investors despite the data showing most underperform?
  3. Give a personal example (real or hypothetical) of recency bias affecting an investing decision.
  4. Name one pre-commitment rule that would help against herding/FOMO.

Progress

Progress is local-only — sign in to sync across devices.
Spotted a bug or have feedback on this page?Report an Issue

Comments 0

🔔 Reply notifications (sign in)
Sign inPlease sign in to comment.

No comments yet — be the first.