Someone says “but it could happen” and the room goes quiet, because nobody knows how to argue with it. And they’re right — it could. A meteor could come through your roof tonight. You could win the lottery tomorrow. A shark could be waiting at the beach next summer. Every one of those is genuinely, truthfully possible. The trouble is that “possible” is true of almost everything, which makes it nearly worthless as a guide to what to actually worry about or do. This lesson is about the difference between could it happen and how often does it happen — and why confusing the two is one of the most expensive mistakes in thinking.
Before you read — take a guess
A friend refuses to fly because 'planes can crash' — then drives four hours to the destination instead. What's the cleanest reason this reasoning is broken?
The two different questions
Possibility and probability sound like cousins, but they answer two completely different questions, and they even come from different fields.
Possibility asks: could this happen at all? It’s a yes/no, on/off, binary verdict. There are only two answers — yes (it’s possible) or no (it’s impossible) — and its home turf is logic: a thing is possible if it doesn’t contradict itself or the laws of the world. “Could a coin land heads?” Yes. “Could a coin land on the number 7?” No. That’s the entire range of answers possibility offers.
Probability asks a richer question: how often does this happen? It’s not yes/no — it’s a number on a continuous scale from 0 to 1 (or, if you like, 0% to 100%). Its home turf is probability theory and statistics, the branch of math built specifically to put a number on uncertainty. “A coin lands heads” isn’t just possible; it happens with probability 0.5. That number is doing work the word “possible” simply cannot.
The relationship is one-directional and worth pinning down: everything with a probability above 0 is possible, but knowing something is possible tells you nothing about where on that 0-to-1 scale it sits. Possibility opens the door. Probability tells you how often anyone actually walks through it.
The tell
Possibility is a switch (on/off). Probability is a dial (anywhere from 0 to 1). Whenever a claim only flips the switch — “it could happen” — your next move is to ask for the dial: how far is it turned?
Where this lies to you: because both words feel like statements about risk, the brain treats “it’s possible” as if it carried weight. It doesn’t. A switch that’s merely on tells you nothing about how bright the room is.
”It’s possible” is almost no information
Here’s the uncomfortable truth: the set of things that are possible is enormous — so enormous that membership in it barely narrows anything down. Winning the lottery is possible. Being struck by lightning is possible. A shark attack, a meteor strike, a stranger leaving you a fortune in their will — all possible. If you sorted the world into “possible” and “impossible,” nearly everything interesting would land in the same gigantic bin, useless for ranking.
That’s the core failure of possibility as a decision tool: it can’t rank anything. Two events can both be possible while differing in likelihood by a factor of a million, and the word “possible” applies identically to both. You cannot prioritize, budget, plan, or worry sensibly using a label that fits almost everything equally.
Worked example — the shark and the car. Consider two ways an American could die this year, both unquestionably “possible”:
| Cause | Roughly how possible? | Roughly how probable (per year, U.S.) |
|---|---|---|
| Killed by a shark | Possible | About 1 in 4,000,000 |
| Killed in a car crash | Possible | About 1 in 9,000 |
Both sit in the “possible” bin, indistinguishable by that label. But the car is on the order of 400 times more probable per year. The person who scans the surf for fins before a beach trip — having just driven there at 70 mph — has ranked the wrong risk because they ranked on possibility, where the two look the same, instead of probability, where they’re separated by a factor of hundreds.
The misconception: “If it’s possible, it’s worth worrying about.” No — worth worrying about is set by probability (and stakes), not by mere possibility. Almost everything clears the low bar of “possible.” Very little clears the bar of “probable enough, at a big enough cost, to act on.”
Probability is a number between 0 and 1
So if possibility is a switch, let’s get precise about the dial. A probability is a number from 0 to 1 that measures how likely an event is:
- 0 means impossible — it never happens.
- 1 means certain — it always happens.
- Everything real and uncertain lives strictly between those two ends.
- 0.5 is the point of maximum uncertainty — a true coin flip, where you have the least possible information about which way it’ll go.
You can read the same number as a percentage by multiplying by 100: 0.5 is 50%, 0.05 is 5%, 0.999 is 99.9%. Same dial, different label.
The catch is that people rarely speak in numbers. They say “likely,” “a real chance,” “almost certain,” “probably not” — and those words are dangerously elastic. The classic cautionary tale comes from intelligence analysis: in a 1951 U.S. National Intelligence Estimate, analysts wrote that an attack was a “serious possibility.” One author meant by it roughly a 65% chance; when colleagues were polled, their private interpretations of that exact phrase ranged from about 20% to 80%. The same three words meant a near-certainty to one reader and a long shot to another. That ambiguity is exactly what attaching a number kills.
| Vague word | What people think it means | What it might actually mean |
|---|---|---|
| ”A real chance" | "Pretty likely — maybe 60%?” | Anywhere from 10% to 70% |
| “Probably" | "Around 75%“ | Anywhere from 55% to 90% |
| “Almost certain" | "Like 95%“ | Anywhere from 85% to 99%+ |
| “Unlikely" | "Maybe 20%“ | Anywhere from 5% to 35% |
Which of these statements about probability is true?
When to reach for the number
The instant a decision actually depends on an uncertain event, translate every vague word into a number — even a rough one. “Likely” is fine for chit-chat; it’s a liability when you’re choosing between a treatment, an investment, or a plan. Force the question: if I had to bet, what number between 0 and 1 would I put on this? You’ll often find the people in the room disagree by 40 percentage points while nodding along to the same word.
Odds vs. probability
You’ll constantly meet probability wearing a disguise: odds. They describe the same uncertainty but count differently, and mixing them up is a common, costly slip.
A probability compares the favorable outcomes to all outcomes. Odds compare the favorable outcomes to the unfavorable ones. So if an event has probability 0.2 (it happens 1 time in 5):
- As a probability: 0.2, or 1 in 5, or 20%.
- As odds: 1 to 4 (“1:4”) — for every 1 time it happens, there are 4 times it doesn’t.
The conversions, worked once:
- Probability → odds: odds = p : (1 − p). For p = 0.2, that’s 0.2 : 0.8 = 1 : 4.
- Odds → probability: p = (first number) ÷ (sum of both). For odds 1 : 4, that’s 1 ÷ (1 + 4) = 1 ÷ 5 = 0.2.
The misconception: treating “5 to 1” as if it meant a 1-in-5 (20%) chance. It doesn’t — odds of 5 to 1 against mean 5 losing outcomes for every 1 winning one, so the probability is 1 ÷ 6 ≈ 0.17 (about 17%), not 20%. The two scales never line up except at the extremes; always check whether a quote is odds or probability before you reason with it.
A weather model says rain tomorrow has odds of 3 to 1 in favor. What is the probability of rain?
The “anything can happen” trap
Now the failure mode this whole lesson is built to defuse. Once people learn that an event isn’t impossible, they tend to make one of two opposite errors — and both come from refusing to look at the actual number.
Error one — the coin-flip inflation. Treating any nonzero probability as if it were roughly a coin flip. “There’s a chance the flight gets cancelled, so I genuinely don’t know if I’ll make the wedding” — when the real cancellation rate is around 2%. A 1-in-50 risk is being lived as a 1-in-2. Mere possibility gets mentally rounded up to “could really go either way.”
Error two — the dismissal. Waving away a real, well-quantified risk because it’s “only a possibility.” “Sure, smoking could cause cancer, but anything could happen” — collapsing a probability that’s enormous into the same “merely possible” bin as a meteor strike. Here possibility-talk is used to shrink a genuine danger down to background noise.
Both errors share one root: they stop at the switch and never read the dial. The cure is the same in both directions — calibrated thinking attaches the actual number. Not “it could happen,” but “it happens about X% of the time, and the cost if it does is Y.” Once the number is on the table, the inflation deflates and the dismissal collapses, because you can finally see whether 2% or 40% is what you’re dealing with.
Where the trap lies to you
The dangerous phrase is “anything can happen.” It’s literally true and almost always misused — deployed to make a tiny risk feel like a coin flip, or to make a huge risk feel like background noise. Both moves work by erasing the number. The instant someone (including you) reasons purely from “it’s possible,” the question to fire back is: yes — but how probable, in numbers, and at what cost?
When to reach for it
Reach for the possibility-vs-probability distinction the moment an argument leans on the word “could.” “But it could happen” / “you can’t prove it won’t” / “anything’s possible” — these are possibility-claims masquerading as risk-claims. The disciplined response is never to deny the possibility (you’ll lose, because they’re right) but to redirect to the dial: granted it’s possible — now, how likely is it, and how much would it cost if it occurred? That single redirection turns an unwinnable “could it?” standoff into a decidable “how often, at what price?” question.
Recap
Check yourself: possibility vs. probability
What is the key difference between possibility and probability?
Check your answer to continue.
You now have the foundation the rest of the course is built on: stop ranking the world by what could happen and start ranking it by how often — a number from 0 to 1, with the stakes attached. But a number alone can still fool you if you pull it out of thin air or ignore how common the thing is to begin with. That’s the next lesson, “Base Rates,” where you’ll learn why the starting frequency of an event often matters far more than the vivid evidence sitting right in front of you.