inversion

name: inversion description: Solve problems by identifying what would cause failure instead of optimizing for success - reveals critical failure modes that forward thinking obscures

Overview

Inversion flips conventional problem-solving by asking "What would cause failure?" instead of "How do I succeed?". Popularized by Charlie Munger (who famously said "All I want to know is where I'm going to die so I'll never go there"), this framework reveals critical failure points that forward thinking obscures. Rather than optimizing for progress, you systematically identify and avoid paths to disaster.

The principle originates from mathematician Carl Jacobi's maxim "Invert, always invert" and has been applied across domains from WWII meteorology to business strategy. Munger demonstrated its power over 60+ years at Berkshire Hathaway, using inversion to make decisions that prioritized avoiding catastrophic mistakes over chasing brilliant wins.

When to Use

You're stuck on a forward-thinking solution that isn't breaking through
High-stakes decisions where avoiding errors is more critical than optimizing gains
Multiple psychological biases could cloud direct judgment
Long-term success depends on consistency and survival over occasional brilliance
You need to identify critical failure modes before they manifest
Planning requires understanding what absolutely must not happen

The Process

Step 1: Reframe Your Goal as a Negative Question

Instead of "How do I achieve X?", ask "What would guarantee I fail at X?" or "What would make X impossible?"

Example: Rather than "How do I build a successful product?", ask "What would guarantee this product fails?"

Step 2: List Specific Failure Conditions

Enumerate concrete, actionable failure scenarios. Be specific - avoid vague fears.

Example for product failure:

No user research before building
Solving a problem that doesn't exist
Ignoring customer feedback for 6+ months
Building features competitors already do better
Burning runway without validating core assumptions

Step 3: Convert Failures into Avoidance Principles

Transform each failure condition into a guardrail or system that prevents it.

Example conversions:

"No user research" → "Talk to 5 users weekly minimum"
"Solving fake problem" → "Require 10 customer interviews before any build"
"Ignoring feedback" → "Monthly review of all support tickets + user interviews"

Step 4: Apply Two-Track Analysis

Evaluate both rational AND psychological failure modes. What rational mistakes could you make? What psychological biases could derail you?

Rational track: Technical failures, market miscalculations, resource constraints Psychological track: Overconfidence, sunk cost fallacy, confirmation bias, fear of competition

Example Application

Situation: Munger was tasked with improving WWII flight safety through better weather prediction.

Application: Instead of asking "How do I predict weather better?", he inverted to "When should flights absolutely be grounded?" He identified specific atmospheric conditions that correlated with pilot deaths and created hard rules: no flights when those conditions exist.

Outcome: Prevented multiple pilot deaths. The inversion revealed that perfect prediction was unnecessary - identifying lethal conditions was sufficient.

Example Application 2

Situation: Physical puzzle - head stuck in metal trap, can't pull through.

Application: Inverted from "How do I pull my head out?" to "What if I push my body through instead?"

Outcome: Instant solution. Forward thinking created tunnel vision on one approach; inversion revealed the opposite worked.

Anti-Patterns

❌ Listing fears without converting them to actionable avoidance principles
❌ Ignoring psychological biases in your two-track analysis (rational failures only)
❌ Using inversion exclusively without combining with forward thinking
❌ Treating inversion as purely theoretical - it must yield concrete guardrails
❌ Confusing what to avoid with what to do (inversion shows avoidance, not direction)

via-negativa (knowledge through subtraction, removing what doesn't work)
pre-mortem (project failure analysis before launch)
second-order-thinking (analyzing consequences of consequences)
circle-of-competence (knowing boundaries of your knowledge)
margin-of-safety (building buffers against failure)