The wheel gives you everything you need
Roulette is one of the few casino games where every possible outcome is visible before you place a bet. The wheel has a fixed number of pockets, the ball must land in exactly one of them, and the payout for each winning bet is printed on the table layout. That means the probability of any bet can be calculated exactly — not estimated, not modelled, but counted. The same cannot be said for a card game where the remaining deck composition changes hand to hand, or a slot where the reel mapping is hidden inside software. Roulette's transparency is what makes it the standard teaching example for how casino odds work.
The calculation itself is simple enough to do on paper. For any bet, you count how many pockets win, count how many pockets lose, and divide the winning count by the total number of pockets. A European wheel has 37 pockets and a single zero; an American wheel has 38 pockets with a double zero. The zero pockets are the mechanism through which the operator's advantage is built, and they change every probability on the table.
Where the house edge lives
The house edge in roulette is not a fee added to your bet. It is a mismatch between the true odds of an event and the payout odds the table offers. If a bet covered all the pockets that exist, a winning payout would have to be 36 to 1 on a 37-pocket wheel for the game to be perfectly fair. Instead, a straight-up number pays 35 to 1. The missing unit is the house edge, expressed as a percentage of each wager over the long run.
This is why the wheel type matters more than the bet type. The edge on most roulette bets is the same across the layout, because the payout structure is built around the same mismatch. The exception is the American wheel's extra double-zero pocket, which increases the edge on every bet compared with a single-zero wheel. The relationship between a game's return to player (RTP) and its house edge is fixed: for the same bet under the same rule set, RTP and house edge sum to 100%. If a game returns a given percentage to players over a very large number of spins, the operator keeps the remainder, and that remainder is the edge.
| Wheel type | Pockets | Zero pockets | Effect on probability | Effect on house edge |
|---|---|---|---|---|
| European (single zero) | 37 | 1 | Each number has 1 chance in 37 | Lower edge on every bet |
| American (double zero) | 38 | 2 | Each number has 1 chance in 38 | Higher edge on every bet |
Because the zero pockets are losing pockets for most outside bets, they are where the operator's advantage is concentrated. A bet on red loses on the zero and double zero even though red and black split the remaining pockets evenly. The payout stays even money, but the probability of winning is no longer one in two. That gap is the entire business model of the game.
Calculating a bet's probability, step by step
Pick any bet on the layout and count the winning pockets. A straight-up bet on a single number wins on exactly one pocket. A split bet covering two adjacent numbers wins on two pockets. A street covering three numbers wins on three. A red bet on a European wheel wins on the 18 red pockets and loses on the 18 black pockets plus the single zero. The probability is the winning count divided by the wheel total.
As an example, suppose you are playing on a 37-pocket wheel and you place a bet on a single number. The probability of winning that spin is 1 divided by 37. The probability of losing is 36 divided by 37. If the wheel has 38 pockets, the same bet wins with probability 1 divided by 38 and loses with probability 37 divided by 38. These fractions are exact. They do not change because of previous spins, dealer behaviour, or how much you have wagered.
What changes with the bet type is how often you win, not the edge attached to each unit wagered. A bet that covers more pockets wins more often but pays less, and the two effects are calibrated so that the operator's edge stays roughly constant across most of the layout. The exception is the five-number bet on the American wheel, which covers 0, 00, 1, 2 and 3 and carries a worse edge than the rest of the table because its payout is not adjusted for the extra pocket it covers.
Why short sessions look nothing like the average
A probability is a long-run statement. It says what happens over a very large number of independent spins, not what happens in the next hour. In a short session, the variance dominates. You can win several spins in a row on a bet with a low probability, or lose repeatedly on a bet that wins nearly half the time. Neither outcome changes the underlying probability of the next spin.
This is the most common misreading of roulette math. A player sees a run of black numbers and concludes that red is now more likely. The wheel has no memory, and the pocket count has not changed. The probability of red on the next spin is exactly what it was on the previous spin: the number of red pockets divided by the total number of pockets. The past sequence is irrelevant to the next outcome, which is why no observation of the wheel can improve your odds.
The same logic explains why a betting progression cannot turn a negative-expectation game into a positive one. Doubling your bet after a loss changes the size of your wagers, not the probability of any outcome. It redistributes when wins and losses land: a progression typically produces many small wins and occasional large losses, while the reverse pattern produces the opposite. The total expected loss over a fixed number of spins is the same either way, because the edge applies to every unit wagered regardless of how the units are sized.
Expected value as the honest arithmetic
Expected value is what you get when you multiply each possible outcome by its probability and add the results. For a roulette bet, there are two outcomes: you win the payout, or you lose your stake. The calculation shows what happens per unit wagered, on average, over enough spins that the variance washes out.
Suppose, as an example, you wager 10 units on a single number on a 37-pocket wheel. The probability of winning is 1 in 37, and the payout is 35 to 1, so a win returns 350 units. The probability of losing is 36 in 37, and a loss costs 10 units. The expected value of the bet is the winning outcome multiplied by its probability, plus the losing outcome multiplied by its probability. The result is a small negative number per spin, and that negative number is the house edge expressed in units rather than as a percentage.
The same arithmetic works for any bet on the layout. What changes is the win probability and the payout, not the method. A bet with a higher win probability pays less, and the product of the two is calibrated so that the expected value stays negative. No combination of bets changes this, because every bet on the table is priced with the same mismatch between true odds and payout odds.
What you control, and what you do not
The probability of any roulette outcome is fixed by the wheel and cannot be changed by the player. What is under the player's control is the choice of wheel, the choice of bet, the number of spins per hour, and the total amount wagered. Each of these changes the total expected loss in a session, but none changes the edge per unit.
Choosing a single-zero wheel over a double-zero wheel lowers the edge on every bet, which means the expected loss per unit wagered is smaller. Choosing a bet with the same edge but different variance changes how the session feels: a high-variance bet produces longer losing runs and occasional larger wins, while a low-variance bet produces more frequent small outcomes. Neither choice changes the long-run expectation.
The number of decisions per hour matters because total exposure is the product of average bet size, decisions per hour, and edge per decision. A player who makes more spins per hour wagers more total units in the same time, and therefore faces a larger total expected loss even though the edge per spin is unchanged. Slowing down the game does not make it profitable, but it does reduce how much is put at risk in a given period.
FAQ
Can you calculate the exact probability of winning a roulette bet?
Yes. Count the pockets that win the bet and divide by the total number of pockets on the wheel. The result is exact because the wheel has a fixed, known set of outcomes and the ball must land in exactly one pocket.
Does the house edge change depending on which roulette bet you make?
For most bets the edge is the same because the payout is set relative to the wheel total. The main exceptions are wheel type, which changes the edge on every bet, and a small number of specific bets on the American layout that carry a worse edge than the rest of the table.
Can watching previous spins help you predict the next one?
No. Each spin is independent, and the pocket count does not change between spins. A run of one colour or number does not alter the probability of any outcome on the next spin, so no observation of past results improves your odds.
Play responsibly. Gambling should be a form of entertainment, not a way to make money. Set limits, take breaks, and never chase losses. If you or someone you know is struggling with gambling, help is available — contact the responsible gambling resources in your area.
Play responsibly
Gambling should be treated as paid entertainment, never as a way to earn income or recover losses. 1-800-MY-RESET (1-800-697-3738)
18+ or 21+ depending on where you are; follow the minimum age that applies to you.
This article is general information about how these mechanics work. It is not legal advice and not a recommendation to gamble or to use any particular operator. Availability and legality differ by jurisdiction — check the rules that apply where you are.