MENTAL MODELS

Break a problem down to its basic truths (axioms) and build up from there, rather than reasoning by analogy ("doing what others do").

80% of effects come from 20% of causes.

Instead of thinking about how to achieve success, think about how to avoid failure.

Don't just look at the immediate effect of an action. Look at the effect of the effect.

Every “yes” is also a “no” to the next-best alternative you could have chosen.

Evaluate choices by probability × payoff, not by the most vivid outcome.

Update beliefs gradually as new evidence arrives instead of flipping opinions emotionally.

Know what you know, know what you don’t know, and make high-stakes bets only inside your competence.

Outcomes are produced by interacting parts, feedback loops, delays, and constraints—not single causes.

Behavior is shaped by reinforcing loops (more leads to more) and balancing loops (stability mechanisms).

Assume your estimates are wrong and build buffers so you survive error.

Small improvements repeated over time produce non-linear gains.

A system's output is limited by its most constrained step.

Among competing hypotheses, prefer the one with the fewest assumptions.

Never attribute to malice what can be adequately explained by incompetence or accident.

Small inputs can produce disproportionately large outputs when applied at the right point.

The best bets have limited downside and unlimited upside.

Extreme outcomes tend to move back toward the average over time.

A product or system becomes more valuable as more people use it.

People make better decisions when they bear the consequences of those decisions.

Some systems don't just resist shocks—they get stronger from disorder.

Past investments (time, money, effort) should not influence future decisions—only future costs and benefits matter.

Think in likelihoods and ranges rather than certainties.

Models and abstractions are useful but always incomplete representations of reality.

Work expands to fill the time available for its completion.

Things that have survived a long time are likely to survive even longer.

Addition by subtraction—removing the negative is often more effective than adding the positive.

Rare, unpredictable events with massive impact happen more often than our models suggest.

A claim is only meaningful if it can be proven wrong. Unfalsifiable claims are useless.

You must keep running (improving) just to stay in place relative to competitors.

We only see the winners and forget the countless failures, skewing our perception of success.

People respond to incentives—if you want different behavior, change the incentives.

Small obstacles dramatically reduce behavior; small enablers dramatically increase it.

The best choice in a narrow context may not be the best choice overall.

Observe, Orient, Decide, Act—and repeat faster than competitors.

Distinguish between reversible and irreversible decisions—treat them differently.

In many domains, a small number of inputs produce the majority of outputs.

The immediate cause is not always the underlying cause. Fixing symptoms doesn't fix problems.

We create simple stories to explain complex, often random, outcomes.

Each additional unit of input eventually produces less additional output.

Shared resources get overused and depleted when individuals act in self-interest.

People often hide their true preferences to conform socially, masking real sentiment.

Make decisions by imagining your 80-year-old self looking back—minimize future regret.

Once people commit to something, they feel compelled to act consistently with that commitment.

Novices overestimate their ability; experts underestimate theirs.

Limitations force focus and innovation. Unlimited options often lead to paralysis.

Starting requires more energy than continuing. The hardest part is beginning.

Many things concern us, but we can only influence a subset. Focus on the latter.