The Price Tag on Every Bet
Every dollar you bet on American roulette costs you 5.3 cents. Every dollar on blackjack with basic strategy costs you half a penny. That’s a 10x difference in price between two games sitting twenty feet apart on the same casino floor.
The number that reveals these prices is called expected value, EV for short. It’s the weighted average outcome of a bet repeated many times, compressing every possible win and every possible loss into a single figure. That figure tells you whether the bet favours you or the house, and by exactly how much.
EV won’t predict your next spin or hand. What it will do is show you the long-run mathematical cost of every wager you place, whether you can see it on the felt or not.
Now that you can read these numbers, here’s what they look like across every major casino game. This table ranks every common bet from cheapest to most expensive per dollar wagered.
| Game | Bet Type | House Edge | EV per $1 | Hourly Cost ($10 bet) |
| Craps | Odds bet | 0.00% | $0.00 | $0.00 |
| Blackjack | Basic strategy, liberal rules | 0.28% | −$0.0028 | ~$1.68/hr (60 hands) |
| Video Poker | Deuces Wild, full pay, optimal | −0.76% | +$0.0076 | +$3.80/hr (500 hands) |
| Video Poker | Jacks or Better, full pay, optimal | 0.46% | −$0.0046 | ~$2.30/hr (500 hands) |
| Blackjack | Basic strategy, typical rules | ~0.50% | −$0.005 | ~$3.00/hr (60 hands) |
| Baccarat | Banker | 1.06% | −$0.0106 | ~$6.36/hr (60 hands) |
| Baccarat | Player | 1.24% | −$0.0124 | ~$7.44/hr (60 hands) |
| Eur. Roulette | La Partage, even-money | 1.35% | −$0.0135 | ~$4.05/hr (30 spins) |
| Craps | Don’t pass | 1.36% | −$0.0136 | ~$8.16/hr (60 rolls) |
| Craps | Pass line | 1.41% | −$0.0141 | ~$8.46/hr (60 rolls) |
| Eur. Roulette | All bets | 2.70% | −$0.027 | ~$8.10/hr (30 spins) |
| Slots | High-RTP (~98%) | ~2% | −$0.02 | ~$100/hr (500 spins) |
| Am. Roulette | All bets | 5.26% | −$0.0526 | ~$15.78/hr (30 spins) |
| Carib. Stud | Ante bet | 5.22% | −$0.0522 | ~$15.66/hr (30 hands) |
| Baccarat | Tie bet | 14.36% | −$0.1436 | ~$86.16/hr (60 hands) |
| Slots | Low-RTP (~85%) | ~15% | −$0.15 | ~$750/hr (500 spins) |
| Craps | Any 7 (proposition) | 16.67% | −$0.1667 | ~$100/hr (60 rolls) |
| Keno | Standard | Up to 25% | Up to −$0.25 | Varies by pace |
One number stands out: Deuces Wild video poker with optimal play actually favours the player. It’s the only common casino bet where the house has negative edge.
Reading the Table
The range here is staggering. A craps odds bet, the only true “fair bet” in a casino, with EV of exactly $0.00, costs you nothing. Keno can cost you a quarter of every dollar. That’s an infinite difference in price between the cheapest and most expensive wagers on the same casino floor.
A few things to keep in mind when using this table. Strategy-dependent games (blackjack, video poker) show optimal-play EV. Your actual cost is higher if you don’t play perfect strategy. Slots EV depends entirely on the specific machine, a 98% RTP slot and an 85% RTP slot are different products at different prices, and if the paytable isn’t posted, you can’t calculate EV at all. Video poker pace is estimated at 500 hands per hour, which reflects steady but not rushed play; faster players will see proportionally higher hourly figures.
This page covers the underlying concept that connects all of these numbers. For game-specific data, see how house edge works.
These three terms describe the identical mathematical relationship from different angles. Knowing one, you can derive the other two instantly.
EV per $1 is the dollar amount you gain or lose on average per dollar wagered.
House edge is that same number expressed as a percentage: house edge = −EV as a percentage of the wager.
RTP (Return to Player) flips the perspective: RTP = 100% house edge. It tells you what comes back.
| Game | EV per $1 | House Edge | RTP |
| Blackjack (basic, typical) | −$0.005 | 0.50% | 99.50% |
| Baccarat (banker) | −$0.0106 | 1.06% | 98.94% |
| European Roulette | −$0.027 | 2.70% | 97.30% |
| American Roulette | −$0.0526 | 5.26% | 94.74% |
| Slots (typical) | −$0.06 | 6.00% | 94.00% |
| Keno | −$0.25 | 25.00% | 75.00% |
A game with 97% RTP has a 3% house edge and an EV of −$0.03 per dollar wagered. Same math, three labels. If you encounter any of these numbers for a game, you already know the other two.
These exceptions are rare, require significant skill and discipline, and most players will never encounter them in practice. They’re worth understanding precisely because they sharpen the contrast with every other bet in the casino.
Full-pay Deuces Wild video poker with optimal strategy returns 100.76%, giving the player a +0.76% edge. Optimal strategy for Deuces Wild is complex, full-pay machines are increasingly scarce, and the edge is thin enough that variance dominates any reasonable session.
Card counting in blackjack can flip the edge to the player. It’s legal, but casinos will ask you to leave when they notice. It demands months of practice, a large bankroll to ride out variance, and unwavering discipline through long losing streaks.
Poker against weaker opponents is player-versus-player, not player-versus-house. The casino takes a rake from each pot, but your EV depends entirely on your skill relative to the table, the house has no mathematical edge on the bets themselves.
Casino bonuses under specific conditions can create positive EV when wagering requirements are low enough and the permitted games have high enough RTP. The next section breaks down exactly how to evaluate them.
Even when the math favours you, variance can wipe out a bankroll before the edge materialises. See responsible gambling resources before treating any of these as an income strategy.
The formula: Bonus EV = Bonus Amount − (Total Wagering Required × House Edge)
Negative EV Bonus: The Common Case
$100 bonus, 35x wagering requirement, played on a slot with 96% RTP (4% house edge).
Total wagering required: $100 × 35 = $3,500
Expected loss from wagering: $3,500 × 0.04 = $140
Net EV = $100 − $140 = −$40
The “free” $100 costs $40 to unlock on average. It’s not free, it’s a discounted session, and only a discount if you planned to wager $3,500 anyway.
Positive EV Bonus: The Rare Case
$100 bonus, 15x wagering requirement, played on a 99% RTP game (such as full-pay video poker or French roulette with La Partage, where permitted by the bonus terms).
Total wagering required: $100 × 15 = $1,500
Expected loss from wagering: $1,500 × 0.01 = $15
Net EV = $100 − $15 = +$85
Offers like this are rare. Operators aren’t in the business of giving away money, which is precisely why they’re worth learning to spot.
The Game Contribution Trap
Many bonuses restrict which games count toward clearing the wagering requirement. Blackjack might count at only 10%, meaning a $10 bet clears just $1 of the requirement.
A “10x wagering requirement” that only permits 10% blackjack contribution silently becomes 100x in effective terms. That transforms the entire EV calculation.
Same $100 bonus, 10x nominal wagering, blackjack at 0.5% house edge, but watch the contribution rate:
| Contribution Rate | Effective Wagering | Expected Loss | Net Bonus EV |
| 100% | $1,000 | $5 | +$95 |
| 10% | $10,000 | $50 | +$50 |
| 5% | $20,000 | $100 | $0 (breakeven) |
The game contribution percentage is the hidden variable that turns a positive-EV bonus into breakeven or worse. Always check which games count, and at what rate, before running any bonus EV calculation.
No operator recommendations in this section, just the math. VegasFreedom has no affiliate relationship with any operator referenced on this page.
You can walk away a winner from a negative-EV game. You can go broke playing a positive-EV one. Over any single session, luck dominates mathematics. The technical term for that noise is variance, the spread of actual outcomes around the expected average.
High-variance games like slots produce dramatic swings. Low-variance games like blackjack track closer to the expected outcome on any given night. Both converge to EV over enough bets. Variance shapes the journey; EV determines the destination.
How Quickly the Math Takes Over
A hypothetical player betting $1 per round on American roulette (red/black, 5.26% house edge). The “plausible range” covers approximately 95% of outcomes (±2 standard deviations).
| Bets | Theoretical EV | Plausible Range (±2 SD) | What This Means |
| 10 | −$0.53 | −$6.85 to +$5.79 | Anything can happen. Winning is completely plausible. |
| 100 | −$5.26 | −$25.23 to +$14.71 | You’re still more likely to be ahead than the EV suggests. |
| 1,000 | −$52.63 | −$115.78 to +$10.52 | Barely possible to break even. The range is closing in. |
| 10,000 | −$526.32 | −$726.04 to −$326.60 | The entire range is negative. The math wins. |
Standard deviation per bet for an even-money roulette wager ≈ $1.00. After n bets, the SD of the total = $1.00 × √n. The ±2 SD range covers roughly 95% of outcomes, derived from the binomial distribution.
The pattern is clear. Over 10 bets, variance drowns out EV entirely. Over 10,000 bets, EV drowns out variance. This is the law of large numbers and the reason casinos, which aggregate millions of bets across thousands of players, are always profitable even when individual players walk away winners.
For more on how variance shapes your sessions, see our guide on variance and volatility.
A $10 blackjack player using basic strategy can expect to spend about $6 across a two-hour session. A $10-per-spin slot player on a machine with a 6% house edge at 500 spins per hour should budget $600 for that same two hours. Same casino, same bet size, 100x difference in cost. Game selection is the single biggest factor in what gambling costs you.
Three things worth remembering:
Lower house edge means lower cost per hour.
You can’t control whether you win or lose on any given night. You can absolutely control which game you sit down at and that choice has more impact on your wallet than any betting system ever invented.
EV doesn’t predict tonight.
A single session is a small sample, and small samples are noisy. You might win $500 at a negative-EV game or lose $500 at a positive-EV one. EV tells you the average across many sessions, not the outcome of this one. That’s variance, and it works in both directions.
No betting system changes the math.
Martingale, Fibonacci, D’Alembert, they all produce the same long-run EV as flat betting. If the underlying game has negative EV, no pattern of bet sizing will overcome it. The only things that change EV are the game you choose and, in strategy-dependent games, the quality of your decisions.
Knowing the price before you sit down is the most practical thing EV can do for a recreational player. Set a loss limit informed by the EV of the game you’re playing, and treat it as what it is, the cost of entertainment.