What Is Expected Value (EV) in iGaming? EV Explained

Expected value, EV for short, is the average result of a bet if you made the exact same bet over and over, forever. It is one number that tells you whether a bet is good for you or bad for you in the long run, before a single card is dealt or a single wheel spins. A fair coin flip for equal money has an EV of exactly $0. Almost every bet inside a casino has an EV below $0, and that gap is precisely what the casino is built to collect.

What does expected value mean?

Expected value is the average outcome of a bet, found by multiplying every possible result by how likely it is, then adding those numbers together. Say a friend offers you a simple bet: flip a fair coin, heads you win $10, tails you pay $10. Half the time you win $10, half the time you lose $10, so the math looks like this: (0.5 × $10) + (0.5 × -$10) = $0. That is a $0 EV bet, sometimes called a "fair bet." It does not mean nothing happens on any single flip, you could still win $10 or lose $10 the very first time. It means that if you repeated that exact bet thousands of times, your average result per flip would settle right around zero. Neither side has an edge over the other. What any one flip does, swinging up or down around that average, is a separate idea called slug:what-is-variance-in-igaming, EV is simply the number those swings are centered on.

Why does almost every casino bet have negative EV?

Because the payout on a casino game is always set a little worse than the true odds of winning, and that gap, the house edge, is exactly what turns EV negative. Take a $10 bet on red at a European roulette wheel. As slug:what-is-house-edge-in-igaming explains, that bet carries about a 2.7% house edge: the wheel has 37 pockets, 18 red, 18 black, and one green zero, so red wins 18 times out of 37 while the payout only covers those 18 numbers, never the zero. Run the actual math and you get EV = (18/37 × $10) + (19/37 × -$10), which comes out to about -$0.27. That is the expected value of one $10 spin. Scale it up and the pattern holds: over 1,000 spins at $10 each, that is $10,000 wagered for an expected loss of around $270, whether you win or lose any given spin along the way. That same figure, described as a percentage of what you wager rather than a dollar amount, is exactly what slug:what-is-rtp-in-igaming calls RTP, just subtracted from 100%.

Can any casino bet ever have positive EV?

Yes, in a handful of well-documented situations, though casinos and sportsbooks work hard to shut each one down the moment it shows up. Blackjack card counters are the classic example. By tracking which high and low cards have already been played, a counter can tell when the remaining deck favors the player and bet more heavily in those moments. According to Blackjack Apprenticeship, a well-known advantage-play resource, a skilled Hi-Lo counter typically swings the game to somewhere around a 0.5% to 1% player edge overall, a small but genuine flip away from the house's usual advantage. Casinos take this seriously enough to watch for exactly that pattern, a player suddenly raising their bet size right as the count turns favorable, and will ban a counter on sight even though counting itself breaks no law.

Sports betting has its own version of the same idea, usually called positive-EV or "+EV" betting. A sportsbook sets its odds based on where it expects the betting public to land, not necessarily on the true probability of the outcome. When a sharp bettor finds a line that undervalues one side, betting it is a positive-EV wager, even though the outcome of that one game is still unknown either way. It works exactly like the coin flip and the roulette wheel, just with harder-to-pin-down probabilities: find enough of these gaps and place enough bets, and the long-run average tips toward the bettor instead of the book.

Casinos and sportsbooks design their entire operation around keeping both of these as rare and short-lived as possible. Multiple decks, frequent shuffling, and floor surveillance make counting hard to sustain for long. Betting limits and fast line adjustments do the same job for sportsbooks: the moment a book suspects an account is consistently finding positive EV, it starts capping how much that account can wager, regardless of how any individual bet turns out. It is worth being clear about what does not create positive EV: betting systems like slug:what-is-the-martingale-betting-system or slug:what-is-the-dalembert-betting-system only change how wins and losses get distributed across a session. They never touch the underlying EV of the bet sitting inside them, which stays exactly as negative as it was before.

What is the Kelly Criterion, and how do pros size a positive-EV bet?

The Kelly Criterion is a formula for deciding how much of a bankroll to risk on a bet you hold a real edge on, based on the size of that edge and the odds being offered. It comes from a real, well-documented source: John Kelly, a researcher at Bell Labs, published it in 1956 in a paper titled "A New Interpretation of Information Rate," in the Bell System Technical Journal. Kelly was not writing about gambling specifically, he was working on information theory, but the math translated directly into a rule for sizing bets, and professional gamblers and investors have used it ever since.

In plain language, the Kelly Criterion says to bet a percentage of your bankroll that is roughly your edge divided by the payout odds. A bigger edge points to a bigger recommended bet. Better payout odds for the same edge also point to a bigger recommended bet. The formula is self-correcting too, because every bet is sized as a percentage of whatever the bankroll happens to be at that moment: a bad run automatically shrinks the next bet, and a good run grows it, which is what keeps a bettor with a genuine edge from ever risking total ruin on a single wager.

Full Kelly is aggressive. It maximizes long-run growth, but the bet sizes it recommends can swing a bankroll hard in the short term, so most professional bettors use only a fraction of it, often called half-Kelly or quarter-Kelly, trading away some long-run growth for a much smoother ride. None of it works without a real, positive edge to start from. Applied to a negative-EV bet, the formula simply says to bet nothing, which is the correct answer: there is no sizing trick that turns a losing bet into a winning one. Good bet sizing is really just one piece of slug:how-to-manage-your-casino-bankroll, the wider set of habits that keeps a session, or a career, from ending early.

How much does negative EV cost at your own bet size?

It depends on exactly three things: how much you bet each time, the house edge of that specific bet, and how many times you make it. Multiply those three together, with a minus sign, and that is your expected value, no matter what the payout odds happen to look like. A $5 bet on a single roulette number, paying 35 to 1 with the same 2.7% house edge as a bet on red, has the same expected cost as a $5 bet on red itself, only the size and frequency of the wins look completely different. Try your own numbers below.

So what is the bottom line on expected value?

EV is the one number that tells you whether a bet is good for you or bad for you, before a single result comes in. A fair coin flip nets to zero. Nearly every casino bet nets out below zero, by exactly the size of its house edge, and that gap is the whole business model, not bad luck on one unlucky night. The rare exceptions, a card counter's favorable count, a mispriced sports line, exist only because someone found a real gap between the true odds and the posted ones, and casinos spend real money closing those gaps the moment they show up. If you are building or running a casino platform, this same math sits behind every title in a page:casino-games library: the paytable is the product, and a carefully set house edge is what makes the business work at scale, spin after spin, across every player who ever plays.