Casino house edge explained: how much advantage does the casino have?
Every casino game is built so that the house wins slightly more often than the player over time. That built-in margin — the house edge — is the entire business model of the industry, and it is also the most useful number a player can know. It turns vague intuition (“the casino always wins”) into a precise figure you can compare across games, estimate hourly costs with, and use to steer real decisions about where and how to play. This guide walks through what the house edge actually is, how it is calculated, where it sits in every major casino game, and how it translates into money over a session.
What the house edge means in plain terms
The house edge is the average percentage of each bet that the casino expects to keep over the long run. If a game has a 2% house edge, the casino keeps, on average, 2 cents of every dollar wagered, and returns the remaining 98 cents to players as winnings. It is the mirror image of RTP: a slot with a 96% RTP carries a 4% house edge, a blackjack variant with a 99.5% return has an edge of just 0.5%.
Two features of the number deserve attention before anything else. First, it is an average across millions of bets, not a guarantee for any single one. A 2% edge game will happily let a player win ten bets in a row. Second, it applies to the total amount wagered, not the total deposited. A player who churns the same $100 through twenty $5 bets has wagered $1,000, and the edge applies to the full $1,000 — which is why bet speed and bet count matter as much as the percentage itself.
The edge comes from the rules of the game, not from manipulation. In roulette, the zero pockets are the mechanism: payouts are priced as if there were 36 outcomes, but 37 or 38 exist. In blackjack, the dealer acting last after player busts does the work. In slots, the paytable prices wins below their true probability. Every game has its own architecture of losing just enough, just often enough.
How the edge is calculated, with a worked example
The clearest way to understand the figure is to compute it for a real bet. Take a single number bet in European roulette. The wheel has 37 pockets, one of which is your number. A win pays 35 to 1.
The expected value of a $1 bet is: (1/37 × $35) − (36/37 × $1) = $0.946 − $0.973 = −$0.027. In other words, every dollar wagered loses, on average, 2.7 cents. The house edge is therefore 2.7% — and that single calculation explains more about roulette than any marketing copy ever will. The American version, with an extra zero, widens the edge to 5.26%, purely by adding one pocket and repricing nothing.
The same logic applies everywhere. Casino war, for instance, derives its edge from ties resolving in the house’s favour. Blackjack’s edge shrinks or grows depending on how many rules — dealer stands on soft 17, doubling after split, surrender — are offered in the player’s favour. The percentage is not a fixed property of “blackjack” as a concept; it is a property of the specific rule set in front of you.
House edge across the major games
Different games carry dramatically different margins, and the gap between the best and worst is enormous. The figures below cover the most common versions found in casinos, both online and land-based.
| Game | Typical house edge | Key factor |
|---|---|---|
| Blackjack (basic strategy) | 0.5% – 2% | Rule variations and player decisions |
| Video poker (full-pay variants) | 0.5% – 5% | Paytable accuracy and skill level |
| Baccarat (banker bet) | 1.06% | Fixed rule set with minimal decisions |
| Craps (pass line) | 1.41% | Fixed odds, no decisions after the come-out |
| French roulette | 1.35% – 2.7% | La partage rule returns half on zero |
| European roulette | 2.7% | Single zero pocket |
| Slots (online) | 2% – 12% | RTP set by the developer per title |
| American roulette | 5.26% | Second zero pocket |
| Keno | 20% – 35% | Extremely unfavourable paytable |
The spread tells its own story. A player at a blackjack table running basic strategy faces a margin roughly ten to twenty times smaller than one feeding keno tickets. Nothing in gambling offers a cheaper way to stretch a bankroll than simply choosing games from the top of this list rather than the bottom — and the choice costs nothing.
How the edge converts to real money per hour
Percentages stay abstract until they are attached to time and bet size. The conversion is straightforward: hourly theoretical cost equals edge multiplied by bets per hour multiplied by average bet. Working through a few realistic scenarios makes the economics tangible.
Consider roulette at 40 spins per hour, $5 per spin, with a 2.7% edge. Total wagered per hour is $200, and the theoretical cost is $5.40 per hour. Compare that with a slot machine at 600 spins per hour (a realistic pace for a fast online player), $1 per spin, 4% edge: $600 wagered per hour, $24 per hour in theoretical cost. The slot player is paying four times as much for entertainment — not because slots are “bad,” but because the volume of betting is far greater. Live blackjack at 70 hands per hour, $10 per hand, 0.5% edge works out to $700 wagered and $3.50 per hour, the cheapest seat in the building by a wide margin.
These figures are expectations, not bills. Individual hours can swing far above or below them — that is variance, not accounting. But over dozens of sessions, the average lands remarkably close to the computed cost, which is exactly why casinos can afford free drinks, lavish buildings and player rewards on margins of one or two percent.
Why the edge is not the same as losing every session
Many players misread the house edge as meaning the casino wins most individual sessions. It does not. Short sessions are decided by variance, and variance is wild. A player grinding a 2.7% edge game can walk away ahead on a given night with real probability — roughly speaking, close to half of all single visits end in profit for the player, depending on the game and session length. The edge is a slow, relentless pressure, not a per-session verdict.
What the edge does guarantee is that the total volume of betting is priced in the house’s favour, and that the longer someone plays, the more their results converge toward that expected margin. This is the classic “grind”: each individual bet is almost fair, but the cumulative effect is mathematically certain in the long run. Session length is therefore a genuine strategic variable. Playing fewer rounds of the same game stretches the same bankroll further against the edge than playing more rounds at higher speed.
Where players give back more than the edge takes
The stated house edge assumes perfect play. In practice, most of the excess money players lose comes not from the game’s margin but from mistakes on top of it. The most common leaks are worth naming plainly.
- Deviating from basic strategy in blackjack. Hitting a hard 17, refusing to split aces, or taking insurance adds percentage points back to the house edge, often doubling or tripling the stated figure.
- Playing slots with low RTP versions. The same title can run at 96% at one casino and 88% at another, and the difference is the edge — four times worse with one click.
- Chasing losses with bigger bets. Raising stakes to recover losses increases the volume of betting, which increases the theoretical cost at exactly the worst moment.
- Making side bets in table games. Insurance, perfect pairs and similar wagers carry edges far above the base game — often 7% to 15%.
- Choosing games at the bottom of the edge table. Keno and some novelty games cost multiples of what blackjack or baccarat cost per dollar wagered.
Every one of these leaks is a choice, not a rule of the game. A player who fixes them plays at or near the theoretical minimum cost of their chosen game — and nothing more is required to achieve that.
Reducing the cost of play within the rules
The edge itself cannot be beaten without changing the game — card counting, advantage play and similar techniques exist but demand extraordinary discipline and are largely neutralised by modern countermeasures. What can be influenced is the effective edge you face per dollar of entertainment. A few levers do almost all of the work.
- Pick games at the low edge end. Blackjack with good rules, baccarat banker bets and French roulette sit at the top of the value table; keno and side bets sit at the bottom.
- Play optimal strategy where decisions exist. Basic strategy charts for blackjack and full-pay strategy for video poker reduce the edge to its stated minimum.
- Slow the pace of betting. Fewer decisions per hour means fewer dollars at risk per hour, and the theoretical hourly cost falls in direct proportion.
- Compare rule sets, not just games. A blackjack table paying 6:5 on blackjack has a much higher edge than one paying 3:2; reading the layout pays better than most promotions.
- Use the same budget for fewer, longer sessions. Spreading a fixed entertainment budget across more sessions with smaller bets reduces total turnover over time.
None of these moves changes the mathematics of any single bet. What they change is the total volume of betting against you over a month, and volume is what the edge multiplies. Reducing volume at a fixed edge is the only lever that reliably works.
The honest bottom line
The house edge is not a secret, not a trick and not something the casino hides — it is published, calculable and remarkably consistent. It is also small in per-bet terms and enormous in aggregate terms, which is precisely the design. Understanding it changes nothing about any given spin, hand or roll, but it transforms how a player prices the whole activity. Blackjack at the right table, played correctly and slowly, costs a few dollars an hour as entertainment. Keno at a fast pace costs ten times that. Same player, same budget, wildly different bill — and the only variable that changed was knowledge. That is the real function of the house edge: not a warning against playing, but a price tag that lets you decide how much entertainment you are buying, and at what rate.
