Odds vs Probability: Two Ways to Say the Same Chance

What does probability actually measure?

Probability measures how large a slice one outcome occupies within every possible outcome. It equals favorable outcomes divided by total outcomes, a number from 0 to 1 or 0% to 100%, describing the long-run share of times an event is expected to occur.

On a European roulette wheel there are 37 pockets, numbered 0 through 36. A bet on a single number has exactly one favorable pocket out of 37 total, so probability equals 1 divided by 37, roughly 2.70%. That figure never changes from spin to spin; the wheel has no memory, and each pocket carries the same weight every spin.

Probability is a share of the whole; odds compare two parts of it.

What are odds, and how do they differ from probability?

Odds compare the ways an event can fail against the ways it can succeed, usually written as ways-against to ways-for. They describe the exact same chance as probability, but as a ratio between two counts rather than a single share of the whole.

Take that same single number on the European wheel. Out of 37 pockets, 36 are losing and 1 is winning, so the true odds are 36 to 1 against. Probability and odds describe the identical 37-pocket reality; probability names the winner's share of the total, while odds compares the 36 losers to the 1 winner.

How do you convert between odds and probability?

The two formats convert with simple arithmetic in either direction, using only the counts already sitting on the table. Odds against equal one minus probability, divided by probability. Probability equals the for side of an odds ratio, divided by the sum of both sides of that ratio.

Both directions use the same two ingredients: the count of favorable outcomes and the count of unfavorable outcomes. Once those two counts are fixed, probability and odds are simply two sentences describing the same reality.

• Probability to odds: subtract the probability from 1, then divide by the probability. For 1/37 (about 0.0270), that is 0.973 divided by 0.027, which rounds to 36 to 1 against.
• Odds to probability: add the two sides of the ratio, then divide the for side by that sum. For 36 to 1, that is 1 divided by 37, the familiar 2.70%.
• Sanity check: 36 to 1 against always describes a smaller chance than 3 to 1 against. The larger the first number, the rarer the event.
• Either direction, both numbers must describe the same wheel and the same 37 pockets. Conversion never changes the chance, only its notation.

Why does the wheel say 1 in 37, but the table pays 35 to 1?

Because the payout board shows payout odds, chosen by the house, not the true odds of the bet. True odds on a single European number are 36 to 1 against; the house pays 35 to 1, one unit short of a fair payout.

A fair bet would pay exactly what the true odds demand: 36 units for every 1 staked, on a wager that only wins once in 37 spins. Instead the payout is 35 to 1. That single missing unit, repeated across every spin the wheel ever takes, is the mechanism by which the house earns a return. Odds in a payout line describe what a player is paid, not the chance of winning; treating 35 to 1 as the true rarity of the event is the most common misreading of a payout board.

A payout ratio tells you what you win. It does not tell you how rarely you win.

What exactly is house edge, and where does the 2.70% figure come from?

House edge is the casino's average profit as a percent of every bet placed, produced entirely by the gap between true odds and payout odds. On European roulette that gap works out to 2.70%, about 2.7 ENT per 100 staked, collected slowly across many spins.

The arithmetic is direct. Stake 1 unit on 37 spins of a single number, in perfect proportion to probability, and the number wins once. A fair payout would return 36 units on that win, leaving the player exactly even against 37 staked. The house instead returns 35 units, a shortfall of 2 across 37, or 2.70%. House edge is an average, not a per-spin outcome; any single spin wins or loses in full, and the edge only becomes visible once the same bet repeats many times, the way probability itself only becomes visible over a long run.

How can a player check whether any payout is fair?

Compare the payout odds printed on the felt to the true odds implied by probability. If the payout odds are shorter than the true odds, meaning the house pays less than a fair share, the house keeps that difference as edge.

The check is the same conversion used throughout this piece, run in reverse. Take the probability of the event, convert it to true odds, and set that figure beside the odds actually paid. On a single European number, true odds are 36 to 1 against; the paid odds are 35 to 1. Thirty-five is shorter than thirty-six, and that one-unit gap is the edge, expressed plainly rather than hidden inside a percentage. This comparison works on any bet in any fair game, not only roulette; wherever a payout ratio sits below the true odds, edge is present.

Does knowing the probability tell you what happens on the next spin?

No. Probability and odds both describe the long run, the average shape of many repetitions. Every individual result in a fair game is independent, so a 2.70% probability says nothing about whether the very next spin wins or loses at all.

Independence means each spin is unaffected by every spin before it. A single number that has not appeared in 40 consecutive spins is not due; its probability on the 41st spin is still exactly 1 in 37, the identical 2.70% it always was, because the wheel carries no memory of its own history. Probability and odds are long-run descriptions of what a bet costs on average, not forecasts of a single upcoming result, and mistaking one for the other is the error behind most gambling misconceptions.