House Edge by Game Type — A Cheat Sheet

House edge is the percentage of every dollar wagered that the casino keeps over the long run. A 1% edge means $1 wagered returns $0.99 in expected winnings; a 5% edge means $0.95. The number is theoretical and applies across many sessions — any specific session will land above or below — but for choosing how to allocate bankroll, the edge tells you most of what you need to know.

The cheat sheet

Sorted by house edge, low to high. Lower edge = more value per dollar wagered.

• Blackjack (basic strategy) — 0.5% edge. The lowest-edge game in any casino if played perfectly. Wager contribution toward bonuses is heavily discounted (often 5-10%) which limits the bonus-clearing utility.
• Video poker (9/6 Jacks or Better, perfect strategy) — 0.46% edge. Even lower than blackjack at the full-pay tables. Same bonus-contribution penalty.
• Provably-fair originals (Crash, Plinko, Mines, Dice, Limbo) — 1.0% edge typical. Operator-published and verifiable per round. Excellent value when wager-clearing isn't the goal.
• Baccarat (Banker bet) — 1.06% edge. Player bet is 1.24%. Tie is 14% — avoid. Bonus contribution is usually heavily discounted.
• Live dealer Blackjack (basic strategy) — 0.5% edge. Same math as RNG, played at human pace.
• European Roulette — 2.7% edge. Single-zero. The 'European' qualifier matters; American double-zero is roughly double.
• Slots (industry average) — 4% edge. Varies from 1-2% on the highest-RTP titles up to 15%+ on novelty progressives. RTP is published on each slot's info panel.
• Sic Bo (Small/Big bets) — 2.8% edge. The good bets only. Single-number Sic Bo bets push 18-19% edge.
• American Roulette — 5.26% edge. The double-zero kills the math. Avoid where European is available.
• Craps (Pass Line + full odds) — 1.4% effective edge on the bet structure. Far worse on Big Red and other 'sucker bets' (16.7%+).
• Slot keno-style games — 5-10% edge typical. Wide variance; check the paytable.
• Lottery-style games and progressive jackpots — varies widely (3-30%+ on contribution to the pool). The jackpot itself is what justifies the structure.

Why the numbers matter

House edge compounds with volume. Across $10,000 of wagering, the difference between Crash (1% edge, $100 expected cost) and a typical slot (4% edge, $400 expected cost) is $300. Across a year of regular play that gap is real money. The catch is that low-edge games are also lower-entertainment — Crash and blackjack don't deliver the dopamine cadence of bonus-buy slots, and that's part of why people play slots despite the math.

Edge vs variance vs entertainment

Three distinct things get conflated. Edge is the long-run expected return. Variance is how widely individual sessions deviate from that return — high-variance games can win or lose huge in any session even at the same edge. Entertainment is what the game actually delivers per dollar of expected cost. Choosing among games means trading these off:

• Optimize for edge: blackjack with basic strategy, full-pay video poker, provably-fair Crash/Dice. Low edge, accept the lower entertainment density.
• Optimize for variance (volatility): Nolimit City extreme-variance slots (Tombstone, San Quentin), Money Train 4, Razor Shark. Brutal sessions on average but rare massive hits.
• Optimize for entertainment: themed slots with frequent bonus rounds, live game shows (Crazy Time, Lightning Roulette). Higher edge, but the experience is what you're buying.

When edge doesn't help you

A common mistake: clearing a slots-restricted bonus by playing blackjack to capture its lower edge. The game contribution rules block this — blackjack at 10% contribution means you'd need to wager 10x the slots target to clear the bonus, and at 0.5% edge instead of 4% edge that's roughly $5 of expected cost per $100 of wagering... but you'd need $30,000 of wagering instead of $3,000, totaling $1,500 of expected cost vs $120 on the slots route. Game contribution math beats house edge math when clearing bonuses. For non-bonus play, edge math beats everything.