Every casino bet has a measurable long-run outcome called expected value (EV), which calculates the average profit or loss per wager based on probability and payout structure rather than short-term results or streaks.
Every casino game contains a mathematical number that determines the long-run outcome of every wager placed on it. That number is expected value (EV). EV describes the average result of a bet if the same wager were repeated many times under identical conditions.
Casino outcomes feel random because each spin, card draw, or dice roll is independent. Short-term results can deviate significantly from long-run averages. A player may win several sessions in a row, or lose repeatedly despite playing a statistically favorable game. Expected value does not predict these short-term fluctuations. Instead, it measures the average profit or loss per bet when the number of trials approaches a large sample size.
Casinos structure games so that the player’s EV is negative. This mathematical advantage, called the house edge, ensures that the operator earns revenue across large volumes of wagers. The implication is mechanical rather than moral: the expected value of most casino bets is less than zero.
Understanding EV changes how you evaluate gambling decisions. Instead of relying on streaks, intuition, or anecdotal strategies, you can quantify the average outcome of a bet using probability and payout size.
Expected value is a statistical measure that calculates the average outcome of a probabilistic event when repeated many times. In gambling contexts, it represents the long-run average profit or loss per wager.
A positive expected value means the average result favors the bettor over repeated trials. A negative expected value means the average result favors the casino.
The key point is that EV describes long-run averages, not single outcomes. A wager with negative EV can still win on any individual attempt, and a wager with positive EV can still lose. The EV only emerges clearly when the number of trials increases.
If a bet has an EV of −€0.27 per €10 wagered, this does not mean you will lose €0.27 every time. It means that across thousands of identical wagers, the average loss per bet approaches €0.27.
Most casino games produce negative EV for the player because payouts are set slightly below the fair mathematical value of the probabilities involved. The difference between fair value and actual payout constitutes the house edge.
Expected value can be calculated using a weighted probability formula. For a simple bet with two possible outcomes—win or loss—the formula is:
$$EV = (P_{win} \times Amount_{won}) – (P_{loss} \times Amount_{lost})$$
Where:
Because probabilities must sum to 1:
$$P_{win} + P_{loss} = 1$$
In casino games with multiple outcomes (such as roulette bets that can return different payouts), the EV formula expands to:
$$EV = \sum_{i=1}^{n} (P_i \times Outcome_i)$$
Where $P_i$ represents the probability of each possible outcome and $Outcome_i$ represents the profit or loss associated with that outcome.
This structure is identical to the expected value formula used in probability theory and statistics.
European roulette contains 37 numbered pockets: numbers 1–36 plus a single zero.
A red/black bet pays even money. If you stake €10 and the ball lands on your color, the casino returns your €10 stake plus €10 profit.
There are 18 red numbers, 18 black numbers, and 1 green zero.
The probability of winning a red/black bet is therefore:
$$P_{win} = \frac{18}{37}$$
$$P_{win} = 0.4865$$
The probability of losing is:
$$P_{loss} = \frac{19}{37}$$
$$P_{loss} = 0.5135$$
Assume a €10 wager.
Winning returns €10 profit, while losing forfeits the €10 stake.
Substituting into the EV formula:
$$EV = (0.4865 \times 10) – (0.5135 \times 10)$$
Calculate each term:
$$0.4865 \times 10 = 4.865$$
$$0.5135 \times 10 = 5.135$$
Now subtract:
$$EV = 4.865 – 5.135$$
$$EV = -0.27 \quad$$
The expected value of a €10 red/black bet in European roulette is −€0.27 per spin.
If you repeat this wager 1,000 times, the expected loss becomes:
$$1000 \times 0.27 = 270$$
Expected loss after 1,000 spins:
$$-€270 \quad$$
This corresponds to a 2.7% house edge.
American roulette introduces an additional green pocket (00), increasing the total to 38 pockets and raising the house edge to 5.26%.
The change illustrates how small differences in probability structure can significantly alter EV over large samples.
Casino bonuses introduce a second layer of expected value because the bonus funds must usually be wagered multiple times before withdrawal. The wagering requirement determines how much total betting volume must occur.
Consider the following example:
First calculate the total wagering volume.
$$Total\ wagering = Bonus \times Wagering\ requirement$$
$$Total\ wagering = 100 \times 35$$
$$Total\ wagering = 3{,}500$$
Total wagering required: €3,500.
Next determine the house edge of the game.
If the RTP is 96%, the house edge is:
$$House\ edge = 1 – RTP$$
$$House\ edge = 1 – 0.96$$
$$House\ edge = 0.04$$
House edge = 4%.
Now estimate the expected loss across the required wagering.
$$Expected\ loss = Total\ wagering \times House\ edge$$
$$Expected\ loss = 3{,}500 \times 0.04$$
$$Expected\ loss = 140$$
Expected loss during wagering: €140.
Finally calculate the net EV of the bonus.
$$Bonus\ EV = Bonus\ value – Expected\ loss$$
$$Bonus\ EV = 100 – 140$$
$$Bonus\ EV = -40$$
Net expected value:
$$-€40$$
In this simplified example, the bonus produces a negative EV because the wagering requirement forces enough betting volume for the house edge to exceed the bonus value.
You can perform these calculations automatically using the Bonus EV Calculator.
A related tool for estimating rollover completion risk is the Wagering Calculator.
Casino games are designed with mathematical parameters that produce negative expected value for players. The mechanism is the house edge, which is the difference between fair probability payouts and the actual payouts offered.
If a roulette red/black bet were mathematically fair, the payout would reflect the true probability of winning:
$$Fair\ payout = \frac{1}{Probability}$$
Because the probability of red in European roulette is $18/37$, a fair payout would be slightly greater than 1:1. The casino instead pays exactly 1:1, creating a built-in deficit in the expected value calculation.
Slot machines implement the same concept through return-to-player (RTP) percentages. If a slot advertises a 96% RTP, the expected loss per €1 wagered is:
$$Expected\ loss = 1 – 0.96$$
$$Expected\ loss = 0.04$$
That corresponds to a 4% negative EV for the player.
The house edge functions as a statistical margin rather than a guarantee for any individual session. Over millions of wagers, however, the aggregate results converge toward the expected value predicted by probability theory.
Positive expected value in gambling contexts is uncommon and usually arises from external incentives or structural inefficiencies rather than the base game itself.
One example is matched betting, where promotional free bets are converted into withdrawable value by hedging the wager across multiple outcomes. The EV arises from the promotional credit rather than the underlying probability distribution.
Another scenario involves bonus promotions with unusually favorable conditions, such as low wagering requirements combined with high-RTP games. If the expected loss from wagering is smaller than the bonus value, the resulting EV may become positive.
A third example appears in blackjack card counting, where tracking the composition of remaining cards can shift the probability of favorable outcomes slightly in the player’s direction. When the deck contains a higher proportion of high cards, the expected value of certain bets increases.
Casinos generally monitor for behaviors associated with positive EV strategies and may limit or exclude players who consistently exploit them.
Expected value is the mathematical framework that determines the long-run outcome of gambling activities. Every casino wager can be evaluated using probability and payout size to determine whether the average result favors the player or the house.
Casino games are structured so that standard bets produce negative EV for the player. The magnitude of that negative EV corresponds directly to the house edge.
Tools that model EV, wagering requirements, and strategy simulations allow you to quantify these outcomes rather than relying on intuition. Examples include the Bonus EV Calculator, the Wagering Calculator, and simulation models such as the Martingale backtest available at this page.
No. RTP (return to player) represents the percentage of total wagers a game returns to players over time. Expected value measures the average profit or loss of a specific wager. RTP and EV are mathematically related because:
$$EV = Stake \times (RTP – 1)$$
Positive EV can occur when promotions or external incentives outweigh the house edge. Standard casino bets without promotions generally produce negative EV because payouts are structured below fair probability values.
No. Expected value is determined by the rules and probabilities of the game. Individual session results vary due to randomness, but the EV of the wager remains constant unless the payout structure changes.
Many participants treat gambling as entertainment rather than investment. The negative expected value functions as the price of that entertainment, similar to paying for a ticket to a live event or other recreational activity.
Betting systems alter the distribution of wins and losses but do not change the underlying probabilities of the game. Because EV depends on probability and payout structure, progression systems cannot transform a negative-EV wager into a positive-EV one.