The Wheel Decides the Math Before You Bet
A roulette spin looks like a pure guess, but the arithmetic is fixed the moment you know which wheel you are standing in front of. Every bet you can place pays out as if the wheel had no zero pocket. The wheel, of course, has one. That gap between the payout and the true odds of the event is the entire business model of the game, and solving roulette probability means measuring that gap precisely.
You do not need to memorise a chart for every bet. You need two numbers: how many pockets win the bet, and how many pockets lose it. The probability of winning is simply the winning pockets divided by the total pockets on the wheel. The payout the table offers is a separate number, and comparing the two tells you the expected loss on the wager.
European wheels have 37 pockets with a single zero. American wheels have 38 pockets with a single and double zero. That one extra pocket changes every calculation on the table, which is why the first question to ask at any roulette table is which layout you are playing.
Probability, Payouts, and the Missing Piece
A straight-up bet on a single number wins on exactly one pocket. On a European wheel, the chance of that number landing is 1 in 37. If the game had no zero and only 36 numbers, a fair payout for that bet would be 35 to 1, because you would win once and lose 35 times for every 36 spins on average. The table does pay 35 to 1. But the wheel has 37 pockets, not 36, so you lose 36 times for every one win on average. The payout is short by one unit per cycle.
That missing unit is the house edge. It is not a fee deducted from your balance. It is built into the mismatch between the payout odds and the true odds. The same structure applies to every bet on the layout, though the size of the gap differs between bet types.
The relationship between return to player (RTP) and house edge is a strict complement: for the same bet under the same rule set, RTP and house edge sum to 100%. If a bet returns a certain percentage to players over a very large number of spins, the house keeps the remainder. Neither number predicts your next spin or your next session.
| Bet type | Winning pockets (European wheel) | Payout | What the gap means |
|---|---|---|---|
| Straight-up (single number) | 1 of 37 | 35 to 1 | Payout is short by one unit per 37-spin cycle |
| Split (two numbers) | 2 of 37 | 17 to 1 | True odds are 17.5 to 1; the half-unit gap is the edge |
| Street (three numbers) | 3 of 37 | 11 to 1 | True odds are 11.33 to 1 |
| Red/black, odd/even, high/low | 18 of 37 | 1 to 1 | The zero pocket is the only reason the bet is not even |
| Dozens and columns | 12 of 37 | 2 to 1 | True odds are 2.08 to 1 |
On an American wheel, every winning-pocket count stays the same while the total rises to 38. The extra pocket increases the house edge on the American layout relative to the European one, but the arithmetic method does not change: divide winning pockets by total pockets, then compare that fraction to the payout.
Expected Value Is a Long-Run Average, Not a Session Forecast
Expected value measures what a bet returns per unit wagered over a very large number of repetitions. It is the cleanest way to see the edge in action. Suppose a player places a straight-up bet of 10 units on a European wheel. The probability of winning is 1 in 37, and a win pays 350 units. The probability of losing is 36 in 37, and a loss costs 10 units. The expected value per spin is (1/37 × 350) plus (36/37 × −10). That arithmetic works out to a small negative number per spin, and that negative number is the house edge expressed in units.
What expected value does not do is tell you what happens this evening. A session of 50 spins is far too small a sample for the long-run average to dominate. A player can win on the first spin, lose on the next ten, or hit two numbers in a row. None of those outcomes contradicts the edge; they are all part of the distribution the edge describes.
Variance is the reason short sessions look nothing like the average. Roulette is a high-variance game on inside bets: payouts are rare but large, so a small number of spins can produce results far from the expected value in either direction. The edge is always present on every bet, but variance determines what the session feels like.
Decisions Per Hour Multiply the Exposure
The house edge is a rate, not a one-time charge. Every spin applies it again. A player who bets 10 units per spin and sees 60 spins in an hour has put 600 units through the wheel. The expected loss for that hour is the edge multiplied by the total wagered, not the edge multiplied by the player's starting bankroll.
Game speed is therefore one of the few things a player genuinely controls. A slower table with fewer spins per hour reduces total exposure. A faster table or an online game with rapid auto-spin increases it. Bet size and bet selection also matter: different bets on the same table carry materially different edges, and the player chooses which one to place.
None of these choices removes the edge. They change the rate at which it applies and the variance the player experiences. A player who wants to minimise expected loss for a given amount of time can choose the bets with the smallest gap between true odds and payout, slow down the rate of play, and reduce the stake per spin. Those are real, mechanical controls, and they are the only ones the table offers.
Why Progression Systems Redistribute Outcomes
A progression system changes the size of the next bet based on the previous result. The most famous is the martingale: double the stake after every loss, then return to the base stake after a win. The logic feels sound because a win eventually recovers all prior losses plus one unit. The problem is not the arithmetic of the progression; it is that the progression does not change the probability of any spin.
Suppose a player starts with a base bet of 1 unit and doubles after each loss. After four consecutive losses, the next bet is 16 units. The probability of losing four even-money bets in a row on a European wheel is (19/37) raised to the fourth power, because 19 pockets lose an even-money bet and 18 win it. That probability is not zero, and when the losing streak arrives, the required bet has grown far beyond the original stake. Table limits and the player's own bankroll cap the progression, and the expected loss per unit wagered remains exactly what the edge says it is.
Progression systems redistribute outcomes rather than remove the edge. They turn many small losses into occasional large ones, or many small wins into occasional large losses, depending on the direction of the progression. The total expected loss across all bets is unchanged. No betting pattern changes the built-in edge of a negative-expectation game, and roulette is negative-expectation on every bet on the layout.
What the Player Can and Cannot Solve
Solving roulette probability means calculating the gap between payout and true odds for each bet, understanding that the zero pocket creates that gap, and accepting that the gap applies to every spin. It does not mean finding a bet that removes the gap, because no such bet exists on a standard wheel.
The player's real decisions are bet selection, stake size, and speed of play. Each one changes the rate of expected loss and the variance of the session. The wheel itself is fixed. The next spin is independent of the last one, and no amount of pattern-spotting changes the number of pockets on the wheel.
For the exact payout and rule set at any particular table, the table layout itself is the authoritative document. It prints the payouts for every bet and the wheel type in use. The game information panel on an online roulette title carries the same information, including the RTP for that specific game under its specific rules. Those figures are the ones to read before placing a bet, because they are the only ones that apply to the wheel in front of you.
FAQ
What is the formula for roulette probability?
Divide the number of pockets that win the bet by the total number of pockets on the wheel. On a European wheel the total is 37; on an American wheel it is 38. The result is the probability of winning that bet on a single spin. The payout printed on the table is a separate number, and the gap between the two is the house edge.
Does the house edge change depending on the bet I place?
Yes. Different bets on the same table carry materially different edges because the payout-to-true-odds gap differs by bet type. The even-money bets have one gap, the inside bets have another, and the five-number bet on an American layout has a larger gap still. The wheel type also changes the edge, because the extra zero pocket increases the house edge on the American layout.
Can a betting system overcome the roulette house edge?
No. A progression system changes the size and sequence of your bets, but it does not change the probability of any spin or the payout the table offers. It redistributes outcomes, turning many small results into occasional large ones, while the expected loss per unit wagered stays the same. No betting pattern removes the built-in edge of a negative-expectation game.
Gamble responsibly. If you or someone you know has a gambling problem, call the helpline printed in the footer below.
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.