Mental Model Selector

Instead of asking how to succeed, ask: what would guarantee failure? Then avoid those.

And then what? Trace consequences two and three steps past the obvious first effect.

Strip the problem to physics-level truths. Rebuild from primitives, not analogies.

Someone takes every option you are weighing off the table. What would you do then? That answer is usually the one worth having.

"Either I stay quiet or I blow this up." That's a false pair. Ask what it'd take to get both things you want.

Stop arguing about which option is right. Ask what each one needs to be true, then go check the shakiest one.

Among competing explanations, the one needing the fewest assumptions usually wins.

Before removing something, understand why it was put there. Don't reform what you can't first justify.

Imagine the decision failed. Pick a point in time that fits the plan, then walk backward: what killed it?

Build buffer for the case where you're wrong. Don't price assumptions to perfection.

Write down now the signal that would tell you to stop. Decide it while you can still think straight about it.

Plan for the unmodeled rare event. The most damaging surprises come from outside the distribution you considered.

When the downside is catastrophic and the upside is only moderate, or the other way around, the math alone won't save you. Weigh which side is bigger.

Spell out all three. Can you live with the worst case? If yes, the upside is mostly free.

When the fix creates the very problem it was meant to solve, or does more damage than the thing it was treating. What could this cure break?

Reversible decisions deserve speed. Irreversible ones deserve patience and process.

Prefer choices that keep future doors open at low cost. Pair a safe floor with a few small bets that gain from surprise, and skip the mediocre middle.

Today's options are constrained by yesterday's choices. Which doors does this decision close on your future self?

Decide when you have ~70% of the info. Waiting for 90% is usually slower than course-correcting.

Spend the first third of your search learning what's out there. After that, take the first option better than everything you've already seen.

Observe, orient, decide, act, then go again faster than the situation changes. When iteration speed beats getting it right the first time.

Work expands to fill the time given. Shrink the timebox and the decision gets sharper.

You underestimate what your own plans will cost and how long they'll take, even when you know about this. It runs long even when you pad it for running long. Use what similar things actually took.

Name the one belief this whole thing rests on. Find the cheapest way to test it this week instead of debating it.

A belief you can't imagine ever being wrong is not a conclusion, it's an identity. Say out loud what would change your mind.

Build the strongest version of the opposing view, then argue with that. You're a scout mapping the ground, not a soldier defending it.

Ask "why" five times in succession to drill from the surface symptom to the actual root cause.

Stop asking "will it work?" Start asking "what are the odds?" and act on the distribution.

What's the historical batting average for decisions like this one? Anchor there before the story.

Start with a prior. Update incrementally as evidence arrives. Hold no belief with 100% confidence.

Multiply outcomes by probabilities. Choose the bet with the best long-run math, not the best feeling.

Facts narrow the field, they don't pick for you. Separate what the research actually says from what you actually want, then decide.

You're studying the winners. The full picture also requires the silent losers you'll never hear from.

Even a positive-EV bet ruins you at the wrong stake size. Size the bet to the edge and the odds, not to your conviction.

Extreme results — good or bad — tend to drift back toward average. Don't extrapolate a hot streak.

Among multiple goals, find options that can't be improved on one axis without sacrificing another. Stop optimizing inside dominated choices.

Score by Impact × Confidence × Ease (and Reach). Forces you to commit to numbers, not vibes.

Hungry, angry, lonely, tired. Fix whichever one is true before you decide anything you can't undo.

You give better advice than you take. What would you tell a friend sitting exactly where you're sitting?

Know the perimeter of what you actually understand. Decisions inside it are cheap, outside it expensive. Feeling certain on unfamiliar ground usually means you're past the line.

Plans fail for lack of counsel and succeed with many advisers. Who have you actually asked, and did you ask anyone likely to tell you no?

Separate what's in your control from what isn't. Spend zero effort outside the line.

System 1 is fast and emotional. System 2 is slow and deliberate. Big decisions deserve System 2; small ones don't need it.

Write down what you decided, why, and what you expected. Review later. Separates good decisions from lucky outcomes.

Money and time already spent can't be won back by spending more. Decide on what's ahead, but keep what the past taught you.

Losses feel about twice as strong as the same size gain. Are you saying no out of fear, not strategy?

We default to "no change" even when change is the better play. The cost of doing nothing is real.

You weigh evidence that fits your hypothesis more heavily. Actively hunt for what would prove you wrong.

The first number or framing sets the reference point. Reset to absolutes before judging "good" or "bad."

You overvalue what you already own and what you personally built. Would you actively choose this today if you didn't already hold it?

You over-weight the near-term reward against the long-term payoff. It's why "future you" keeps getting betrayed by "present you."

We want what others want, not what we actually want. Strip the social signal: would you still want this if no one knew you chose it?

You explain other people's behavior by character and your own by circumstance. Are you reading the situation, or just reading the person?

Vivid recent examples feel more probable than they are. Are you over-weighting one memorable story?

"I knew it all along" rewrites the past so the outcome looks inevitable. Distorts the lessons you draw from your last big decision.

What would have happened anyway, without your action? Subtract the noise to measure your real contribution.

Never attribute to malice what is adequately explained by incompetence or oversight.

Pull up a chair for whoever isn't in the room but has to live with this. Say the decision out loud to them and see if it still sounds right.

Would you choose this if you didn't know which role you'd play in the outcome? A fairness stress-test.

Picture this printed on the front page with your name on it. Still fine? If you'd explain it one way in public and another in private, that's your answer.

Trust forecasts only from people who bear the downside if they're wrong. Whose advice survives if it backfires on them too?

Other players are optimizing too. What's their best response to your move? Reason about where it settles.

When a measure becomes the target, it stops being a good measure. Watch what your metric is quietly distorting.

Individually rational behavior produces collectively destructive results. If everyone in your position did the same thing, what breaks?

Without communication, what would each party naturally converge on? Use focal points to coordinate.

Groups burn disproportionate time on trivial details they can easily opine on. Are you debating the easy 5% to avoid the hard 95%?

The real cost of a "yes" is every "no" it forces. What are you giving up?

Roughly 80% of effect comes from 20% of causes. Find that 20% before anything else.

Improvement often comes from removal, not addition. What should you stop doing instead of layering on something new?

Set what good enough looks like, then take the first option that clears it. People who hold out for the very best end up less happy with what they get.

The raise, the house, the title all stop feeling good faster than you expect. Count how long the good part actually lasts.

Past a certain point, more money and more status stop buying anything. Have you named your number, or are you still running a race with no finish line?

Sort by urgent and important. Do the important-not-urgent before it becomes the urgent crisis.

If it's not a "hell yeah," it's a no. The middle is where commitment quietly drains out of life.

Project yourself to 80 and ask which choice you'll regret not having tried. For smaller ones, ask how it'll feel in 10 minutes, 10 months, and 10 years.

Some choices change who you are, so the person deciding isn't the person who has to live with it. You can't run the numbers on becoming someone new.

One list gets read at your funeral, the other gets read by recruiters. Which list does this decision add to?

Name who's better off if this goes well, besides you. Work aimed only at your own advancement runs out of fuel.

Small advantages, repeated, beat large advantages, abandoned. What is the version of this decision that compounds?

You don't rise to your goals, you fall to your systems. Which option leaves you with a better routine, not just a better outcome?

Many small pushes in one direction compound into momentum. Which version of this builds a flywheel instead of starting from rest every time?

Focus where what you love, what you're best at, and what pays intersect. A filter for opportunity-rich, attention-poor decisions.

For ideas and non-perishables, the longer it's been around, the longer it'll likely stay. Bet with time.

Your model is a simplification. Watch where the model breaks against the actual terrain.

Diagnose the domain — simple, complicated, complex, chaotic — before picking a method. Most bad decisions are a domain mismatch.

Don't reduce a role to its narrowest function. The doorman opens the door; he also signals luxury, safety, and ceremony.