The wheel is a fixed instrument; the math follows from its layout
Before any spin, the roulette wheel is already a complete statement of probability. Nothing about the sequence of past results changes what the next spin can do. The wheel has a fixed number of pockets, each pocket has a fixed colour and number, and the ball has no memory. When you calculate anything about roulette, you are not predicting a spin. You are reading the geometry of the wheel and converting it into odds, payouts, and the long-run cost of each bet.
Two wheel layouts dominate real-money play. A European wheel has 37 pockets: the numbers 0 through 36, with a single zero. An American wheel has 38 pockets: the same 0 through 36 plus a double zero. That one extra pocket changes every calculation that follows, because the wheel's total number of pockets is the denominator in every probability.
Start with the probability of a single number
Probability is the number of ways a bet can win divided by the total number of possible outcomes. On a European wheel, a straight-up bet on one number has 1 winning pocket out of 37 total pockets, so the probability of winning on any single spin is 1 in 37. On an American wheel, the same bet has 1 winning pocket out of 38, so the probability is 1 in 38.
The same logic extends to every bet on the layout. A split bet covers 2 numbers, so its probability is 2 in 37 on a European wheel and 2 in 38 on an American wheel. A street bet covers 3 numbers, a corner covers 4, a six-line covers 6, and a dozen or column covers 12. Red or black, odd or even, and high or low each cover 18 numbers. On a European wheel, the zero is neither red nor black, neither odd nor even, and neither high nor low, so none of those even-money bets covers it.
Convert probability into odds and fair payout
Odds express the same relationship as probability but in a form that makes payout comparison easier. If a bet has 1 winning pocket and 36 losing pockets on a European wheel, the odds against winning are 36 to 1. A fair payout, meaning a payout with no house advantage, would be 36 units for every 1 unit wagered. The actual payout on a straight-up bet is 35 to 1. The gap between the fair payout and the actual payout is where the house edge lives.
Here is the arithmetic for a straight-up bet on a European wheel. Suppose you wager 1 unit on one number. In a perfect distribution of 37 spins, you would win once and lose 36 times. The win pays 35 units, and you also keep the winning wager, so the win returns 36 units in total. You wagered 37 units across those 37 spins. You are short 1 unit. That 1 unit, divided by the 37 units wagered, is the house edge expressed as a fraction of total wagers.
| Bet type | Numbers covered | Actual payout | Fair payout (European wheel) |
|---|---|---|---|
| Straight-up | 1 | 35 to 1 | 36 to 1 |
| Split | 2 | 17 to 1 | 17.5 to 1 |
| Street | 3 | 11 to 1 | 11.33 to 1 |
| Corner | 4 | 8 to 1 | 8.25 to 1 |
| Six-line | 6 | 5 to 1 | 5.17 to 1 |
| Dozen or column | 12 | 2 to 1 | 2.08 to 1 |
| Red, black, odd, even, high, low | 18 | 1 to 1 | 1.06 to 1 |
The fair payout column shows what the payout would need to be for the bet to have no house advantage. Every actual payout in roulette is slightly below the fair payout for that bet. The shortfall is the same fraction of each wager on a European wheel, regardless of which bet you choose, because the single zero is the only pocket that no standard bet covers.
The American wheel adds a second uncovered pocket
On an American wheel, the double zero joins the single zero as a pocket that red, black, odd, even, high and low do not cover. That changes the denominator in every probability from 37 to 38 but leaves the payouts unchanged. A straight-up bet still pays 35 to 1, but the odds against winning are now 37 to 1. The gap between the fair payout and the actual payout is wider on the American layout.
Run the same 1-unit straight-up bet on an American wheel. In a perfect distribution of 38 spins, you win once and lose 37 times. The win returns 36 units. You wagered 38 units. You are short 2 units, which is a larger fraction of total wagers than the 1-unit shortfall on the European wheel. The extra pocket increases the house edge on the American layout, as the published wheel comparison notes. The payouts are identical; the wheel's total pocket count is the only difference.
Expected value is the long-run arithmetic of any bet
Expected value is the average result per wager if the same bet were repeated an enormous number of times. It is not a prediction of the next spin. To calculate it, multiply each possible outcome by its probability and add the results. For a straight-up bet on a European wheel, the two outcomes are a win returning 36 units with probability 1 in 37, and a loss returning 0 units with probability 36 in 37. The expected return per 1-unit wager is the win's contribution minus the wager itself, which leaves the same shortfall identified above.
The same expected value applies to nearly every bet on a European wheel because the payout schedule is built to produce a uniform edge. The exception is the five-number bet on an American wheel, which covers 0, 00, 1, 2 and 3 and pays 6 to 1. Because it covers 5 numbers on a 38-pocket wheel but pays as if it covered only 5 of 36, its expected value is worse than the other bets on the same layout. Different bets on one table can carry materially different edges, and this is the roulette example of that principle.
House edge and RTP are complements
House edge and Return to Player (RTP) describe the same relationship from opposite sides. House edge is the share of each wager the operator keeps as a long-run average. RTP is the share returned to players. For the same bet under the same rule set, the two sum to 100%. If a bet's expected return is short by a certain fraction of each wager, that fraction is the house edge, and the remainder is the RTP.
You can calculate one from the other. Suppose a bet on a particular wheel layout has a house edge expressed as a fraction of each wager. The RTP is 100% minus that fraction. The calculation does not require knowing any operator's published figure; the wheel's pocket count and the payout schedule are enough. The result is a long-run statistical average, not a per-session guarantee, and it does not change based on how many spins you play or what has happened in the last hour.
What the player controls and what the wheel controls
The wheel controls the pocket count, the payout schedule, and the house edge. The player controls which bets to place, how much to wager on each, how many spins to play, and when to stop. None of those choices changes the built-in edge on a negative-expectation game. Choosing a European wheel over an American wheel changes the expected rate of loss, not the direction of it. Choosing a straight-up bet over a red or black bet changes the variance, not the edge.
Variance is how payouts are distributed over time. A straight-up bet pays rarely but pays a large multiple when it hits. An even-money bet pays often but returns only the wager as profit. Both have the same house edge on a European wheel. A session of straight-up bets will look very different from a session of red or black bets even though the long-run arithmetic is identical. Volatility does not change RTP, and it does not change the house edge.
Where the numbers come from and how to verify them
Every calculation in this article follows from two facts: the wheel's total pocket count and the payout printed on the table layout. The pocket count is fixed by the wheel type. The payout schedule is printed on the felt or shown in the game's information panel. If you are playing online, the game rules screen lists the wheel type and the payouts. If you are at a physical table, the layout itself is the source.
When a figure is not printed, the arithmetic still works. The house edge is not a number the operator needs to publish because it is implied by the wheel and the payouts. The RTP is the complement. You can calculate both from the table in front of you, and you can check your arithmetic against publicly published reference sources that describe roulette odds, with the usual caveat that such sources are snapshots and the wheel in front of you is the operative document.
FAQ
How do I calculate the odds of red or black?
Count the red pockets and the black pockets on the wheel you are using. On a European wheel, there are 18 red and 18 black pockets out of 37 total, so the probability of red on any spin is 18 in 37. On an American wheel, it is 18 in 38. The payout is 1 to 1 either way, which is why the American wheel's extra pocket increases the house edge on that bet.
Does the house edge change if I bet on multiple numbers?
On a European wheel, no. The payout for each bet is set so that the shortfall from the fair payout is the same fraction of each wager, whether you cover one number or eighteen. The exception is the American five-number bet, which covers five pockets but pays as if it covered fewer, making its edge worse than the other bets on that wheel.
Can I calculate the RTP of a roulette bet from the payout alone?
You need both the payout and the wheel's pocket count. The payout tells you what a win returns; the pocket count tells you how often a win occurs. Together they give the expected return per wager, and the RTP is that expected return expressed as a percentage of the wager. The house edge is the remainder, and the two sum to 100% for the same bet under the same wheel layout.
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.