In the days before a hurricane reaches the coast, two people stand in front of the same satellite loop. The first, asked whether the storm will hit the city, says: “Well, it’s possible. It could go either way. Better safe than sorry.” The second says: “Roughly a 30% chance it makes landfall here, 70% it curves north and misses us.” Both are looking at identical data. Only one of them has said anything you can actually use. With “it’s possible,” you can’t choose between boarding up your windows and going to work. With “about 30%,” you can.
That gap — between a mind that flips a yes/no switch and a mind that turns a dial — is the entire subject of this course. Almost everyone defaults to the switch. They ask “is it possible?”, get the answer “yes” (because nearly everything is possible), and then either panic or shrug. The edge, in forecasting, investing, medicine, business, and your own ordinary decisions, belongs to the people who skip that useless question and go straight to “how likely?”
The move, in one breath
Here’s the whole idea before we spend five lessons unpacking it. Thinking in probabilities means replacing the true/false switch in your head with a dial. Instead of sorting claims into will happen and won’t happen, you attach each one a number between 0 and 1 — a probability — and ask “how likely?” rather than “is it possible?” A 0 means it can’t happen; a 1 means it’s certain; almost everything interesting in life lives somewhere in the messy middle, and that middle is where all the information is.
It sounds almost trivially obvious. It is also the single habit most reliably skipped under pressure, because the human brain loves certainty. A clean “yes” or “no” feels like an answer; “about 30%” feels like dithering. So we round our beliefs to the nearest extreme — it’ll be fine or we’re doomed — and throw away the precise, uncomfortable number that was the actual answer.
Before you read — take a guess
A friend refuses to fly because 'a plane crash is possible — it could happen to me.' By that exact reasoning, what else should they also refuse to do?
Why “is it possible?” is almost a useless question
Look at what trapped that friend. They asked a binary question — is a crash possible? — and got the only answer that question ever really gives: yes. Possibility is binary: something either can happen or it can’t, and the bar for “can” is absurdly low. A meteor through your roof: possible. Winning the lottery twice: possible. Your flight going down: possible. Once you notice that nearly everything clears that bar, you realize the word “possible” sorts the world into one giant pile labelled “yes” and a tiny one labelled “physically impossible.” That’s not a useful sort.
Probability is the real information. It doesn’t ask whether something is in the “can happen” pile — it asks where on the dial it sits. Flying and driving are both in the “possible to die” pile, but one is at 0.0000001 per trip and the other is hundreds of times higher. The possibility question can’t see that difference. The probability question is made of that difference. That’s why “how likely?” beats “is it possible?” every single time: it’s the question whose answer actually changes what you should do.
The one-sentence version
Thinking in probabilities = stop asking whether something is possible (almost everything is) and start asking how likely it is — a number on the 0-to-1 dial — because that number is the only part that tells you what to do. If you remember nothing else from this course, that sentence already separates you from most decision-makers.
Which of these statements actually carries decision-useful information, rather than just confirming something is in the 'can happen' pile?
Why this is a mental model, not just “be less certain”
“Don’t be so black-and-white” is advice everyone has heard and almost no one can act on, because it doesn’t tell you what to do instead. Thinking in probabilities is the operational version. It gives you a specific, repeatable move — attach a number (or a range) between 0 and 1 to the belief — a specific thing to look for — the actual likelihood, not just whether it’s possible — and a specific failure to expect, which we’ll meet again and again: people who hear “30%” and mentally round it to “won’t happen,” then feel betrayed when the 30% lands. (It was supposed to land three times in ten. That’s not the forecast being wrong; that’s you mis-hearing the dial as a switch.)
And like every good model, it’s portable. The same dial that turns “will it rain?” into “60% chance of rain” turns “is this investment safe?” into “how likely is each outcome, and what’s it worth?”, turns “is the patient sick?” into “given a positive test, how probable is the disease really?”, and turns “was that a good call?” into “was it likely to work, regardless of how it happened to land?” It isn’t tied to a domain. It’s a move you can make about any uncertain claim, anywhere, and it reliably surfaces the information that yes/no thinking throws away.
A weather app says '70% chance of rain' on Tuesday. Tuesday is dry. Which reaction shows someone actually thinking in probabilities?
The map of the course
Five short teaching lessons, then one exam you can’t undo. The route:
- Possibility vs. Probability — the core distinction made precise: why “it’s possible” is binary and nearly free, why probability is a number on 0–1, the “it’s possible, therefore…” trap, and how to speak in calibrated language instead of vague hedges.
- Base Rates & the Outside View — how likely something is before you look at the specifics, why ignoring that starting number (the base-rate fallacy) wrecks predictions, and how to pick the right reference class. You’ll explore an interactive base-rate visualizer here.
- Expected Value — the workhorse formula
EV = Σ(probability × payoff), worked out in full tables; why a low-probability bet can still be worth taking, asymmetric bets, and how variance differs from EV. You’ll drive an interactive expected-value calculator here. - Thinking in Ranges — why a single point estimate lies, the danger of false precision, the “flaw of averages,” and how to reason with ranges and confidence intervals instead of one tidy number.
- Decision vs. Outcome — the final and most counter-intuitive move: judging a choice by whether it was likely to work, not by how it happened to land (“resulting”), and separating luck from skill. Ends with a whole-course recap quiz.
Then a Final Exam — graded, one question at a time, one-way: once you answer, it locks. No back button, no retries. You’ll be ready.
How to use this course
One rule does most of the work: guess before you peek. When you hit an exercise, commit to an answer in your head before revealing anything. The small sting of being wrong is what makes the idea stick — a smooth, nodding read-through teaches you almost nothing. The exercises are the lesson; the prose just sets them up.
Next up: lesson 1, where we pin down exactly what separates possibility from probability — because right now you have “how likely?” as a slogan, and a slogan is not yet a tool.