Casino game maths for non-mathematicians: expected value in practice

In a casino, “expected value” (EV) is the average outcome you’d get if you could repeat the same wager thousands of times. You don’t need advanced maths: EV simply combines how often something happens with how much it pays when it does. If the EV is negative, the game is priced against you; if it is positive, it is theoretically favourable. In practice, EV explains why short-term wins feel persuasive while long-term results drift towards the house edge.

To use EV, think in terms of per-pound cost. Suppose a bet returns £2 with probability 45% and £0 otherwise. The EV is 0.45×£2 + 0.55×£0 = £0.90. If you paid £1 to place that bet, your EV is -£0.10, meaning you “pay” 10p on average each time. That 10p is the house edge expressed in cash terms. Variance is the swing around that average: high-variance games can deliver dramatic wins, but the EV still governs the long run. When comparing games, focus on rules and payouts that move EV: blackjack rule tweaks, slot return-to-player, or roulette’s zero(s). Tools and summaries such as BetCollect can help you translate game rules into the practical question: “How much am I expected to lose per hour at my stake?”

A well-known figure who helped popularise probability thinking in gambling is professional poker player and author Annie Duke, whose work on decision-making under uncertainty emphasises separating good process from lucky outcomes; her primary social profile is AnnieDuke. That mindset fits EV perfectly: judge a casino decision by its long-run expectation, not the last spin. For broader context on how regulation and technology are reshaping the sector, see The New York Times. Bringing it together: choose games with the smallest negative EV, set time-and-money limits, and treat entertainment spend as the true “price” of variance.