Mental Models for Better Decisions

Every decision you make is based on a model. Not a formal model, usually. An internal representation of how things work. These models are often invisible, inherited from experience, culture, and habit. And when your model is wrong, your decision will be wrong, no matter how carefully you reason.

Mental models are tools for thinking. The more models you have, the more perspectives you can apply to any situation. Charlie Munger calls it a "latticework of mental models." The goal isn't to master one model. It's to collect enough models that you're rarely stuck seeing a problem from only one angle.

First principles thinking

Most decisions are made by analogy. "This is similar to that other thing, so I'll do what worked there." Analogies are fast and usually good enough. But they break down when the situation is genuinely new.

First principles thinking strips a problem down to its fundamental truths and reasons up from there. Instead of asking "how has this been done before?" you ask "what do I know to be true, and what can I build from that?"

Elon Musk's often-cited example: instead of accepting that batteries cost $600/kWh because that's what they cost, he broke batteries down to their raw materials (cobalt, nickel, lithium, carbon) and found the material cost was about $80/kWh. The gap between $80 and $600 was a design and manufacturing problem, not a physics problem.

You don't need to be reinventing industries to use this. When you're stuck on a personal problem, ask: "What's actually true here? What am I assuming just because it's always been done this way?"

Inversion

Instead of asking "how do I succeed?" ask "how would I guarantee failure?" Then avoid those things.

Inversion reveals blind spots that forward thinking misses. If you want a good relationship, list everything that would destroy one: dishonesty, contempt, neglect, keeping score. Now don't do those things. You haven't figured out everything a good relationship needs, but you've eliminated the most common failure modes.

For complex decisions, inversion is often more productive than forward planning. There are many ways to succeed but a smaller number of reliable ways to fail. Identifying and avoiding failure modes is simpler than optimizing for success.

Second-order thinking

First-order thinking asks: "What happens if I do this?"

Second-order thinking asks: "And then what?"

Most bad decisions are first-order correct and second-order disastrous. Eating junk food is first-order pleasant (tastes good) and second-order harmful (health consequences). Avoiding a difficult conversation is first-order comfortable and second-order destructive (resentment builds).

The practice is simple: for any significant decision, run the chain forward three steps. If I do X, then Y happens. And if Y happens, then Z. Most second-order effects are predictable if you bother to look.

The map is not the territory

Every model simplifies reality. That's what makes models useful, they reduce complexity. But it also means every model is wrong in some way. The question is whether it's wrong in a way that matters for your current decision.

This model reminds you to check your abstractions against reality. Your model of a person, a market, a relationship, or a project is a simplification. When decisions aren't producing expected results, the problem is often that your map has diverged from the territory. Update the map.

Circle of competence

You know more about some domains than others. Inside your circle of competence, you can make well-calibrated decisions. Outside it, you're operating on incomplete mental models, which means your confidence exceeds your accuracy.

The skill isn't expanding your circle (though that helps over time). The skill is knowing where the boundary is. Inside the circle, decide confidently. Outside it, either learn before deciding, consult someone whose circle includes this domain, or acknowledge the uncertainty explicitly.

Most costly mistakes happen when someone operates outside their circle of competence without realizing it. They apply models from their domain to a new domain where those models don't hold.

Probabilistic thinking

Most things aren't certain. They're probable. Thinking in probabilities instead of certainties produces better decisions because it matches reality.

Instead of "this will work" or "this won't work," try "there's about a 70% chance this works and a 30% chance it doesn't." Now you can make decisions that account for both possibilities. You can prepare for the 30% scenario while pursuing the 70% one.

Probabilistic thinking also protects against outcome bias. A good decision can produce a bad outcome, and a bad decision can produce a good outcome. Judging decisions by outcomes alone leads you to overcorrect good processes and reinforce bad ones.

Using mental models in practice

You don't need to memorize a list. You need to internalize a few and practice applying them.

Start with these five. For the next month, when you face a significant decision:

1. Ask what's actually true (first principles)
2. Ask what would guarantee failure (inversion)
3. Ask what happens after what happens (second-order effects)
4. Ask whether your model matches reality (map vs territory)
5. Ask whether you're inside your circle of competence

One of these will be useful almost every time. And the practice of pausing to apply a model, any model, before deciding is itself the most valuable mental model of all.