We can't predict exactly what will happen, but we can predict how often it will happen. Probability is logic applied to uncertainty.
The foundation is simple: probability = favorable outcomes ÷ total possible outcomes, always landing between 0 (impossible) and 1 (certain). A fair coin lands heads with probability 1/2; a die shows a six with probability 1/6. From that one ratio, an entire science of prediction unfolds — the same math that prices insurance, powers weather forecasts, and runs casinos profitably.
Two Rules Cover Most Problems
- The AND rule (multiply): For independent events, P(A and B) = P(A) × P(B). The chance of two heads in a row is 1/2 × 1/2 = 1/4.
- The OR rule (add): For events that can't happen together, P(A or B) = P(A) + P(B). The chance a die shows 1 or 2 is 1/6 + 1/6 = 1/3.
Worked Example: The "At Least One" Trick
What's the probability of rolling at least one six in four rolls of a fair die? Counting the ways directly is messy — one six, two sixes, three, four... The professional shortcut: compute the opposite and subtract from 1.
- Define the complement. "At least one six" fails only when ALL four rolls avoid six.
- One roll avoids a six with probability 5/6.
- Four independent rolls all avoid it: (5/6)⁴ = 625/1296 ≈ 0.482.
- Subtract: P(at least one six) = 1 − 0.482 = 0.518, about 52%.
Slightly better than a coin flip — which surprises most people, who intuitively guess 4 × (1/6) = 67%. That naive addition is wrong because the rolls can overlap (you might roll two sixes); the complement method handles it exactly. Whenever a problem says "at least one," reach for the complement.
The Normal Distribution
Also known as the Bell Curve. It appears everywhere: height, IQ, test scores. 68% of data always falls within 1 standard deviation of the mean.
Extend that: about 95% falls within 2 standard deviations, and 99.7% within 3 — the famous 68-95-99.7 rule. If exam scores average 70 with a standard deviation of 10, then roughly 68% of students score between 60 and 80, and a score above 90 puts you in about the top 2.5%. The bell curve arises whenever many small independent effects add together — that's the Central Limit Theorem, and it's why the same shape shows up in biology, finance, and measurement error alike.
Expected Value: The Long-Run Average
Multiply each outcome by its probability and add them up, and you get the expected value — what you'd average per play over many repetitions. A lottery ticket that pays $500 with probability 1/1000 has an expected value of 50 cents; if it costs $2, you lose $1.50 per ticket on average, which is precisely how lotteries and casinos guarantee their profit. Insurance works the same way in reverse. Expected value is the single most practical idea in probability: it turns uncertain outcomes into a concrete number you can compare against a price.
Common Mistakes to Avoid
- The gambler's fallacy. After five heads in a row, tails is NOT "due." The coin has no memory; the next flip is still 50/50. Independent events never balance out short-term.
- Adding probabilities of overlapping events. P(A or B) = P(A) + P(B) only when A and B can't co-occur. Otherwise you must subtract the overlap: P(A) + P(B) − P(A and B).
- Multiplying when events aren't independent. Drawing two aces from a deck without replacement isn't (4/52)², because the second draw depends on the first: it's 4/52 × 3/51.
- Probabilities outside [0, 1]. If your calculation yields 1.3 or −0.2, an error is guaranteed — often a symptom of one of the mistakes above.
Frequently Asked Questions
What's the difference between probability and odds?
Probability compares favorable outcomes to all outcomes; odds compare favorable to unfavorable. A 1/4 probability equals odds of 1:3 ("one to three in favor"). Sportsbooks quote odds; scientists quote probabilities — and converting carelessly between them is a classic slip.
What does "independent" actually mean?
Two events are independent when knowing one outcome tells you nothing about the other — coin flips, separate dice. Card draws without replacement are dependent, because each draw changes the remaining deck. The multiplication shortcut P(A)×P(B) is only valid for the independent kind.
Why is the normal distribution so common in nature?
Because of the Central Limit Theorem: when a quantity is the sum of many small, independent influences — genes and nutrition on height, dozens of question-level slips on a test score — the total tends toward a bell curve, regardless of what each individual influence looks like. It's a mathematical guarantee, not a coincidence.