Hourly Expected Loss on Stake: Calculating the True Cost of Each Game (2026)

House edge gets most of the attention when players talk about long-run cost on Stake, but on its own it does not tell you how fast money actually leaves your bankroll. A 1% edge on a Dice bot grinding 2,000 rounds per hour is a very different proposition from a 1% edge on a live blackjack table dealing 70 hands per hour. Hourly expected loss combines bet size, bet frequency and house edge into a single number that mirrors what your balance curve actually looks like over time. This guide walks through the math, applies it to the games you will encounter on Stake, and shows how to translate the result into session caps that hold up under normal variance.

The Formula: Three Variables, One Number

Hourly expected loss (EL/hr) is the product of three things you can measure or estimate:

• Bet size — the amount staked per round.
• Rounds per hour — how many bets you actually complete in an hour of play.
• House edge — 1 minus the RTP, expressed as a decimal (a 96% RTP game has a 0.04 edge).

EL/hr = Bet × Rounds per hour × House edge

A simple worked example: a $1 bet, 500 spins per hour, on a slot with 96% RTP yields $1 × 500 × 0.04 = $20 per hour of expected cost. The 4% edge sounds small in isolation; the hourly figure is what the bankroll actually sees on average.

Bet Cadence by Game on Stake

Games churn through bets at wildly different speeds. The frequency variable is usually the largest single driver of hourly cost, far more than house edge differences between most games.

Stake Originals

• Dice (manual): roughly 200 to 400 rounds per hour.
• Dice (auto-bet): commonly 1,500 to 3,000+ rounds per hour, capped by client throttle.
• Limbo (auto-bet): 1,000 to 2,000+ rounds per hour.
• Crash (auto-cashout): 80 to 150 rounds per hour because rounds run longer.
• Mines: 60 to 150 rounds per hour depending on tile selections.
• Plinko (auto-bet): 400 to 900 rounds per hour.
• HiLo: 40 to 120 rounds per hour, with cashout decisions stretching each round.

Slots

• Manual spinning at default speed: 400 to 600 spins per hour.
• Quick spin or turbo with auto-play: up to 800 to 1,200 spins per hour.
• Bonus rounds and long animation sequences pull the average back down on hit-frequent titles.

Live Dealer and RNG Tables

• Live roulette: 30 to 45 spins per hour.
• Live blackjack: 50 to 80 hands per hour.
• Live baccarat: 70 to 100 hands per hour.
• RNG table games (electronic versions): typically 5 to 10 times faster than their live counterparts.

Worked Examples

Example 1: Same Edge, Different Cadence

A $0.50 Dice bet at 2,000 rounds per hour with a 1% house edge: $0.50 × 2,000 × 0.01 = $10 per hour. The same bet and edge played manually at 300 rounds per hour: $0.50 × 300 × 0.01 = $1.50 per hour. The game and the math are identical; only the cadence changed, and the cost moved by a factor of more than six.

Example 2: Same Game, Different RTP Variant

Many slot providers ship multiple RTP variants of the same title to operators. A $0.40 spin on the 96% RTP variant at 600 spins per hour costs $0.40 × 600 × 0.04 = $9.60 per hour. The 94% variant of the same game, same bet, same speed, costs $0.40 × 600 × 0.06 = $14.40 per hour. Same screen, fifty percent more expected outflow. Checking the in-game info panel for the configured RTP is one of the cheapest edges available.

Example 3: Slots Versus Live Blackjack

A $5 bet on live blackjack at 70 hands per hour with basic strategy (about 0.5% edge) costs $5 × 70 × 0.005 = $1.75 per hour. The same $5 bet on a 96% RTP slot at 600 spins per hour costs $5 × 600 × 0.04 = $120 per hour. The slot extracts roughly seventy times the value at the same nominal bet size. Speed and edge compound; ignoring either one badly understates real cost.

Where Variance Distorts the Picture

Expected loss is a long-run average. Inside any single session, the actual outcome can be far above or below the expected figure. Two factors widen the spread between expected and observed:

• Volatility — the standard deviation of the per-round outcome. High-volatility slots can be 40 times the bet down or 30 times the bet up after 600 spins and still be inside one standard deviation of normal behavior.
• Sample size — short sessions are dominated by variance, long sessions converge toward the expected value. A two-hour bonus hunt is essentially a coin flip relative to the expected value; a 50,000-spin month settles much closer to the math.

Practical takeaway: use hourly expected loss to size the average session, and pair it with a variance-aware cap (often 5 to 10 times the expected loss) to absorb the downside swings without ending the bankroll.

Turning the Number Into Session Rules

Once you have an hourly cost, you have direct levers to control it. Each one is a knob you can dial before the session, when discipline is cheapest:

• Lower the bet size — linear impact on cost, no skill required.
• Slow the cadence — manual play, longer animations, mandatory pauses between bursts.
• Pick lower-edge games — live blackjack at 0.5% beats most slots at 4% on a per-round basis.
• Prefer higher-RTP variants when the operator offers them.
• Set a hard time limit — every minute is an hourly-cost multiplier.

Writing the rule down before play beats trying to enforce it mid-session. SSPilot lets you encode an hourly budget directly through max-bets-per-session caps, stop-loss triggers and time-based session logs, so the actual cost stays close to the planned figure regardless of how variance behaves in any given hour.

A Quick Reference for Common Stake Sessions

Rough order-of-magnitude figures at $1 average bet, useful for sanity-checking a planned session:

• Slots, 600 spins/hr, 96% RTP: about $24/hr expected loss.
• Dice auto-bet, 2,000 rounds/hr, 1% edge: about $20/hr.
• Limbo auto-bet, 1,500 rounds/hr, 1% edge: about $15/hr.
• Crash, 120 rounds/hr, 1% edge: about $1.20/hr.
• Live blackjack, 70 hands/hr, 0.5% edge: about $0.35/hr.
• Live roulette (European), 35 spins/hr, 2.7% edge: about $0.95/hr.

Multiply or divide by your actual average bet to scale these figures to your bankroll. The relative ranking is what matters: auto-bet originals and turbo slots are an order of magnitude more expensive per hour than slow live tables at the same nominal stake.

Conclusion

House edge alone systematically undersells the cost of play because it ignores how fast the game runs. Multiplying edge by bet size and rounds per hour produces a number that mirrors what the bankroll graph actually does over time. Use it to compare games honestly, to set loss caps that survive normal variance and to choose between speed and risk on purpose rather than by accident. Gambling on Stake is entertainment with a known unit cost — putting a clear hourly figure on that cost is how you make sure the session you planned is the session you actually play.