House Edge: The Only Stat That Always Wins

Edge in one line

The house edge is the average share of each bet the casino keeps over time. A 1% edge means an expected loss of $1 per $100 wagered - not per bet, but across thousands. It is why individual players win constantly while the casino profits overall: the law of large numbers does the work.

Why it is unavoidable

Every payout is set slightly below true odds. European roulette pays 35:1 on a number that hits 1-in-37, not the fair 36:1. That gap, compounded over millions of bets, is the edge. The more you play, the closer your results track it.

Edge by game

Blackjack with basic strategy: ~0.5%. Baccarat banker: ~1.06%. European roulette: 2.7%. Most slots: 2-6% by RTP. American roulette: 5.26%. Some side bets and keno: 10%+. Game choice is the biggest lever you control - and the differences are enormous.

Playing against it

You cannot delete the edge, but you can shrink it: choose low-edge games, learn optimal strategy where skill counts, skip high-edge side bets, and claw back margin with rakeback. Then set a budget and treat the edge as the price of the entertainment.

Edge vs hold: what you actually pay

House edge is a per-bet figure, but what leaves your pocket depends on how many times you feed the edge - a quantity often called 'hold' or churn. Two players can face the same 1% edge and lose wildly different amounts because one wagers each dollar once and the other recycles the same bankroll dozens of times. Consider a $100 bankroll on a 2.7% single-zero roulette table. Bet it once on red and your expected loss is $2.70. But real play does not work that way: you win some, bet again, and the same money crosses the felt repeatedly. Wager a total of $1,000 across a session - entirely normal from a $100 start - and expected loss becomes $27, not $2.70. The edge did not change; your exposure to it multiplied. This is the single most important idea in casino math and the one most players miss: game choice sets the rate, but bet count and session length set the bill.

Turning any edge into a number

The formula is always the same: expected loss = total wagered x house edge. What changes is the inputs, and getting them right is where a tool earns its keep. To estimate a session cost you need three things - your average bet, how many bets per hour the game allows, and the edge. Slots can run 400-600 spins an hour; live blackjack maybe 60-80 hands; roulette perhaps 40-50 spins. Multiply bet x rounds-per-hour x edge and you have expected hourly cost. A $1 slot at 500 spins/hour and a 4% edge costs about $20 an hour in expectation; a $10 blackjack hand at 70 hands/hour and a 0.5% edge costs about $3.50, despite the far larger bet. The counterintuitive result - that a small-stakes slot can cost more per hour than a mid-stakes blackjack seat - falls straight out of the numbers, and it is exactly the kind of comparison a house-edge calculator is built to surface.

The edges people forget

The headline edge of a game is often not the edge you end up paying, because side bets and paytable variants hide costlier margins in plain sight. Blackjack's basic-strategy edge is around half a percent, but the 'insurance' and 'perfect pairs' side bets alongside it can carry edges of 6% or more - a different game bolted onto the same table. Caribbean and three-card poker's optional jackpot bets, keno's double-digit edges, and slot bonus-buy modes with reduced RTP builds all work the same way: they dangle a big payout to distract from a much larger cut. A disciplined player learns to price the whole bet, not the exciting part. The rule of thumb is blunt but reliable - the flashier the payout and the rarer the trigger, the larger the edge funding it.

Shrinking the edge, step by step

Since the edge is fixed per game but your total cost is not, every lever that matters is one you control. Start with game selection, the biggest single move: choosing blackjack over keno can swing your effective edge by a factor of twenty. Add correct strategy wherever skill counts, because the low blackjack figure assumes basic strategy - play by feel and you hand back a full point or more. Refuse the high-edge side bets no matter how tempting the payout. Then recover margin on the back end: rakeback and cashback return a slice of the edge you have already paid, effectively lowering your net cost per dollar wagered, which is why a value-focused player treats rakeback as part of the edge equation rather than a bonus. Finally, cap your exposure - shorter sessions and smaller bets mean fewer trips through the edge - and set a hard budget you treat as the price of the entertainment. You will never beat the house over the long run in a game of chance, but you can decide how much that entertainment costs, and how long it lasts. Only ever wager what you can afford to lose, and step away if it stops being fun. 18+ only.

Why individuals win while the house profits

One of the most misunderstood things about the house edge is how it coexists with the constant sight of players winning. The resolution is the law of large numbers. Over a handful of bets, results are dominated by variance - luck - and the edge is almost invisible, which is why any given player can walk away ahead on any given night. But the casino is not playing a handful of bets; it is aggregating millions of bets across thousands of players simultaneously. At that scale, the actual return converges tightly onto the theoretical edge, and a 1% or 3% margin on an enormous volume of wagers becomes a stable, predictable profit. The individual player rides the noisy short run; the casino banks the smooth long run. This is also why 'a hot streak' is real but unbankable: it is variance in your favour, and variance has no memory and no obligation to continue. Understanding this keeps you from two errors - believing a winning session proves a system works, and believing a losing session means a game is 'due' to pay.

Skill games versus pure chance

Not all house edges are equally fixed, and the difference between skill and chance games is worth understanding because it changes what you can control. In a pure chance game like roulette, keno or slots, the edge is baked into the payout structure and no decision you make during play alters it - your only levers are game choice and bet sizing. In a skill-influenced game like blackjack or video poker, the quoted low edge is a best-case figure that assumes you play the mathematically correct decision every time. Deviate from basic strategy and you are effectively playing a higher-edge game; the 0.5% blackjack figure can balloon toward 2% or worse in the hands of a player going on feel. This means skill games reward study in a way chance games cannot: the edge you actually face is partly a function of your own competence. For a numbers-minded player the implication is clear - if you are going to play a skill game, learn the optimal strategy first, because the low advertised edge is a reward for correct play, not a default you get for free.

A ranking you can actually use

Turning all of this into practice means keeping a rough mental ranking of games by edge and letting it guide where your money goes. At the low-cost end sit blackjack with basic strategy and full-pay video poker, both under or around 1%, followed by baccarat banker near 1.06% and single-zero roulette at 2.7%. In the middle sit most slots, spread across a 2-6% band depending on the deployed RTP build. At the expensive end sit American double-zero roulette at 5.26%, many game-show and wheel formats, and the truly costly options - keno and various side bets that can exceed 10%. This ranking is the single most powerful tool you have, because game choice dwarfs every betting system ever invented: moving from a 5% game to a 1% game cuts your expected cost fivefold for the same play. Pair a low-edge game with correct strategy, refuse the high-edge side bets, recover a slice through rakeback, cap your session, and set a budget you treat as the price of the fun. That is the whole of realistic edge management, and it is entirely within your control even though the edge itself never disappears. Only ever wager what you can afford to lose. 18+ only.

Variance versus the edge in a single session

A final idea ties the whole picture together: the house edge and variance are two different forces, and in any single session it is variance, not the edge, that you actually feel. The edge is a smooth, relentless average - the share of each wager the casino keeps over the long run. Variance is the noise around that average, the swings of luck that can leave you far ahead or far behind on any given night. Over a handful of bets the edge is almost invisible and variance rules, which is exactly why a player can win on a high-edge game or lose on a low-edge one in the short term. Over a huge number of bets the noise cancels out and the edge dominates, which is why the casino, aggregating millions of wagers, banks a predictable profit while individuals ride the swings.

Understanding this protects you from two expensive errors. The first is mistaking a winning session for proof that a system works - it is almost always variance in your favour, and variance has no memory and no obligation to continue. The second is believing a losing streak means a game is 'due' to pay, which is the same fallacy pointed the other way. The disciplined response is to accept that you cannot control variance and cannot beat the edge, and to focus entirely on the levers you do control: choosing low-edge games, playing correct strategy, refusing high-edge side bets, recovering a slice through rakeback, and capping your exposure with a budget and a hard stop. That is realistic edge management, and it is enough to make the entertainment affordable even though the edge never disappears. Only ever wager money you can afford to lose. 18+ only.