# Prospect Theory > Understand how people evaluate risk irrationally through loss aversion and reference points - predict decisions by recognizing pain of loss outweighs equivalent gain - Skill: `lev-os/prospect-theory` (Agent Skill) - Install (CLI): `npx skillmds@latest add lev-os/prospect-theory` - Raw SKILL.md: https://api.skillmd.com/api/skills/lev-os/prospect-theory/raw - Safety review: pending - Works with: Claude Code, Claude.ai, OpenAI Codex - Category: Coding & Dev Tools - Author: lev-os (https://skillmd.com/u/lev-os) - Updated: 2026-09-10 - Page: https://skillmd.com/skills/lev-os/prospect-theory --- # Prospect Theory ## Overview Prospect Theory, developed by Daniel Kahneman and Amos Tversky in their landmark 1979 paper, revolutionized economics by describing how people actually make decisions under uncertainty - as opposed to how perfectly rational agents would behave. The theory's core insight: humans evaluate outcomes relative to a reference point (not absolute terms), losses loom psychologically larger than equivalent gains (loss aversion), and we're risk-averse with gains but risk-seeking with losses. This work earned Kahneman the 2002 Nobel Prize and launched behavioral economics. The framework predicts systematic deviations from rational choice theory: people will reject a 50/50 bet to win $110 or lose $100 (rational expected value = +$5) because the pain of losing $100 exceeds the pleasure of gaining $110 by roughly 2x. ## When to Use - Designing pricing, framing, or incentive structures (marketing, product, negotiation) - Predicting how customers, employees, or voters will react to changes - Understanding why people hold losing investments too long or sell winners too early - Structuring offers to account for reference point anchoring - Explaining seemingly irrational behavior (why people buy insurance then play lottery) - Making personal decisions involving risk (career, investments, health) ## The Process ### Step 1: Identify the Reference Point People don't evaluate outcomes in absolute terms - they compare to a reference point (usually their current state, but can be shifted by framing). Same objective outcome feels like gain or loss depending on reference. **Example:** Receiving a $5,000 bonus feels great. Receiving a $5,000 bonus when you expected $10,000 feels like a $5,000 loss. Objective outcome identical, reference point changes everything. ### Step 2: Apply Loss Aversion Asymmetry Losses hurt approximately 2-2.5x more than equivalent gains feel good. Use this to predict behavior: people will work harder to avoid losing $100 than to gain $100. **Example predictions:** - Employees will fight harder against 5% pay cut than they'll celebrate 5% raise - Customers will churn faster from price increase than they'll sign up from price decrease - Investors will hold losing stocks (avoiding realization of loss) longer than rational ### Step 3: Frame Outcomes Relative to Reference Point How you frame an option as gain or loss (relative to reference) dramatically changes acceptance, even with identical math. **Classic example (Tversky & Kahneman):** **Gain frame:** "This treatment saves 200 of 600 patients." (72% acceptance) **Loss frame:** "This treatment results in 400 of 600 patients dying." (22% acceptance) Same outcome, different framing, 3x difference in acceptance. ### Step 4: Predict Risk Preference Reversal People are risk-averse in gains (prefer guaranteed $50 over 50% chance of $100) but risk-seeking in losses (prefer 50% chance of losing $100 over guaranteed loss of $50). Use this to design choices. **Application:** If selling a premium product (gain domain), emphasize certainty and safety. If fixing a customer problem (loss domain), they'll gamble on risky solutions to avoid certain loss. ## Example Application **Situation:** SaaS company considering annual subscription pricing change from $1,200/year to $120/month (mathematically identical). **Application:** - **Reference point shift**: Customers anchor on "$1,200" as reference. Monthly pricing creates new "$120/month" reference. - **Loss aversion insight**: Customers perceive annual change as one large decision ($1,200 at risk). Monthly feels like smaller, reversible commitment ($120 risk). - **Framing strategy**: Position monthly as "try for just $120" (gain frame: get access for small amount) vs. annual "commit $1,200 upfront" (loss frame: big money at risk). **Outcome:** Conversion rate increased 34% with monthly pricing despite higher total cost. Loss aversion made upfront $1,200 feel riskier than 12� $120 payments. Prospect Theory predicted customer behavior correctly. ## Example Application 2 **Situation:** Hospital reducing elective surgery cancellations (patients ghost appointments, wasting OR time). **Application:** - **Standard approach**: Remind patients of appointment (neutral framing). Cancellation rate: 23%. - **Prospect Theory approach**: Reframe as loss - "By not calling to cancel, you're causing the hospital to lose $500 and preventing another patient from getting care." - **Reference point**: Shifted from "my appointment" to "resource I'm taking from someone else." **Outcome:** Cancellation rate dropped to 9%. Loss framing (you're causing loss to others) more effective than gain framing (please confirm to help us). ## Anti-Patterns - L Assuming people evaluate absolute outcomes rationally (ignoring reference dependence) - L Framing only in gain terms when loss framing would be more persuasive - L Treating loss aversion as a "bug" rather than predictable pattern to design around - L Ignoring that reference points can be manipulated by anchoring - L Using Prospect Theory to manipulate unethically (dark patterns, predatory pricing) - L Forgetting that you're also subject to these biases in your own decisions ## Related - system-1-system-2 (loss aversion is System 1 automatic response) - anchoring (reference points heavily influenced by anchors) - endowment-effect (owning something shifts reference point) - sunk-cost-fallacy (loss aversion makes past losses loom large) - framing-effects (outcome presentation changes reference perception)