Three Numbers That Must Be Kept Separate
Most confusion comes from calling theoretical RTP, effective RTP and one player’s observed return the same thing.
| Question | Correct Answer | Reason |
|---|---|---|
| Does 100% RTP guarantee break-even? | No. | It is an expectation, not a session settlement promise. |
| Can observed return exceed 100%? | Yes. | A winning sample can return more than the theoretical average. |
| Can observed return be 0% on a fair game? | Yes. | A small sample can lose every wager. |
| Does provably fair prove 100% RTP? | No. | It checks outcome generation, not payout pricing. |
| Can a cap reduce effective RTP? | Yes. | A fair uncapped payout can be truncated at settlement. |
This is the foundation page for the zero-edge cluster. It defines RTP and expected value. Detailed implementations belong on separate pages:
- How Zero Edge Works — native pricing, distributions, account credits and payout caps.
- Zero-Edge Allowance Explained — tracker, reset, qualifying stake and post-cap state.
- Zero Edge vs Rakeback — instant return, theoretical-loss rakeback, lossback and VIP rewards.
- House Edge Cost — expected cost by turnover.
RTP, House Edge and Expected Value
RTP stands for Return to Player. For a game with outcomes i, probabilities pi and gross payout multipliers mi:
RTP = Σ pi × miHouse edge
House edge = 1 − RTPExpected net profit per unit wagered
EV = RTP − 1Casino multipliers normally include the returned stake. A fair 50% win-chance bet therefore pays 2.000x gross:
0.50 × 2.000 = 1.000, or 100% RTP
Half the rounds lose one unit and half return two units, including the original stake. The expected net profit is zero.
| Gross Multiplier at 50% Chance | RTP | House Edge | Expected Net Profit per Unit |
|---|---|---|---|
| 2.000x | 100% | 0% | 0.000 |
| 1.998x | 99.9% | 0.1% | −0.001 |
| 1.980x | 99% | 1% | −0.010 |
| 1.940x | 97% | 3% | −0.030 |
Theoretical RTP Is Not an Observed Session Return
Theoretical RTP is calculated from the game model. It does not need millions of rounds to exist. The large sample is needed for the observed return to approach the theoretical expectation.
Observed return is:
Observed return = actual gross payouts ÷ actual turnover
| Ten Fair Even-Money Bets | Wins | Gross Payouts | Turnover | Observed Return |
|---|---|---|---|---|
| Losing sample | 2 | 4 units | 10 units | 40% |
| Break-even sample | 5 | 10 units | 10 units | 100% |
| Winning sample | 8 | 16 units | 10 units | 160% |
All three samples can come from the same 100% RTP game. The game expectation is unchanged; only the finite result differs.
Why Turnover Matters More Than Deposit Size
Expected cost is applied to total wagering volume:
Expected cost = Turnover × House edge
| Total Turnover | 100% RTP | 99.9% RTP | 99% RTP | 97% RTP | 96% RTP |
|---|---|---|---|---|---|
| $1,000 | $0 | $1 | $10 | $30 | $40 |
| $10,000 | $0 | $10 | $100 | $300 | $400 |
| $50,000 | $0 | $50 | $500 | $1,500 | $2,000 |
A $100 deposit can generate thousands in turnover if the balance is repeatedly wagered. The deposit limits exposure; turnover determines the recurring mathematical cost.
Use the Edge Cost Calculator for custom volume and edge scenarios.
Why a Fair Game Can Still Lose
Removing expected cost does not remove the distribution of outcomes. A fair game can produce losing streaks, deep drawdowns and complete bankroll loss.
For example, ten consecutive losses in an independent 50/50 game have probability:
(1/2)10 = 1/1,024 ≈ 0.0977%
That is uncommon for one chosen block of ten rounds, but it is possible. Across many sessions and players, such streaks are normal.
Read Can You Lose with 100% RTP? for the full variance and bankroll analysis.
Theoretical RTP vs Effective RTP
A game can have a published native RTP while the player’s effective return changes because of eligibility, account credits or settlement limits.
| Mechanism | What Changes Effective RTP | Required Evidence |
|---|---|---|
| Native fair pricing | Probability and visible payout table | Complete paytable or formula |
| Allowance | Eligible volume and post-cap state | Tracker, reset, game list and crossing-bet rule |
| Instant turnover return | Automatic credit per eligible wager | Account ledger and cap rules |
| Rakeback or VIP rewards | Realized redeemable return | Calculation base, game weights and cash-equivalent value |
| Maximum-win cap | Truncation of high payouts | Stake, multiplier, cap and affected outcome probabilities |
| Strategy-dependent rules | Player decisions and table rules | Exact strategy, decks, payouts and allowed actions |
Blackjack is a clear example of strategy-dependent RTP: table rules and player decisions can materially change the expected return. Use the Blackjack RTP Calculator to compare payout, deck count, S17/H17, doubling, splits, surrender and strategy assumptions.
Five Common Meanings Behind a “100% RTP” Claim
| Claim Type | What It Can Mean | Primary Check |
|---|---|---|
| Native 100% RTP | The complete probability-weighted paytable sums to 1.0000. | Audit the formula or full paytable. |
| 100% inside an allowance | Eligible volume receives fair pricing or a full offset. | Check tracker, reset and post-cap state. |
| Reward-adjusted 100% | A lower native RTP plus automatic value may reach 100%. | Measure the realized return as a percentage of turnover. |
| 100% before a max-win cap | Uncapped math is fair, but large settlements may be truncated. | Recalculate RTP using capped payouts. |
| Best-strategy 100% | The return depends on playing an exact strategy under defined rules. | Audit the strategy-dependent state model. |
The headline may be technically correct for one state and misleading when applied to all bets, all stakes or all users.
Provably Fair Is a Separate Audit
Provably fair verification checks whether a completed result follows from committed seed data and the selected algorithm. It does not calculate the expected value of the paytable.
| Question | Provably Fair | RTP Audit |
|---|---|---|
| Was the revealed server seed committed earlier? | Yes | No |
| Can the completed result be reproduced? | Yes, with the correct algorithm | No |
| Does the paytable have a margin? | No | Yes |
| Did a cap reduce the settlement? | No | Yes, with payout and cap data |
| Will the platform process withdrawals? | No | No |
Use the Provably Fair Checker for supported completed outcomes and How to Verify Provably Fair Games for the process.
Current Examples of Different Models
The following examples illustrate mechanisms, not platform recommendations.
| Example | Model | What Must Be Checked |
|---|---|---|
| Duel | Allowance-based instant return on eligible play, with game-specific exceptions | Live tracker, qualifying stake, post-allowance state and displayed game edge |
| Gamdom | Around 99% base RTP plus rewards and instant returns on certain eligible volume | Actual account credits, reward value, eligible games and threshold behavior |
| MetaWin ZERO games | Selected native zero-house-edge games with maximum-win limits | Stake × multiplier exposure and cap-adjusted expected return |
For the full current comparison, use Zero-Edge Crypto Casinos Compared.
How to Verify a 100% RTP Claim
- Define the scope: identify the exact game, mode, stake, account state and time window.
- Obtain probabilities: use the published formula, paytable or complete outcome distribution.
- Calculate native RTP: sum probability × gross payout for every outcome.
- Check eligibility: record allowance, tracker, reset and qualifying stake.
- Add only realized credits: use automatic or redeemed value as a percentage of eligible turnover.
- Apply settlement limits: replace uncapped payouts with actual capped payouts where necessary.
- Verify outcomes separately: reproduce completed rounds through the documented fairness algorithm.
- Check platform risk separately: review KYC, withdrawals, restrictions and dispute options.
What Does Not Establish 100% RTP?
- A short winning sample: observed profit does not prove the theoretical paytable.
- A short losing sample: drawdown does not disprove a fair game.
- Provably fair alone: honest randomness can coexist with a house edge.
- “Up to” rewards: promotional maximums are not realized return.
- A single visible multiplier: one row may be fair while another is capped or mispriced.
- A platform-wide label: selected Originals do not define slots, live casino or sportsbook RTP.
Primary Operator Examples Checked
- Duel Zero Edge FAQ — allowance-based instant return and reset behavior.
- Duel Economics Changelog — game-specific and wager-dependent edge rules.
- Gamdom Originals Help Center — around 99% RTP, rewards and instant returns.
- MetaWin FAQ — selected games with zero house edge.
Frequently Asked Questions
Does 100% RTP mean the casino returns my exact wagers?
No. It means the mathematical expected gross return equals eligible turnover under the defined rules. Individual players and sessions can finish far above or below that value.
How is 100% RTP different from 0% house edge?
They describe the same theoretical pricing from opposite directions: house edge equals 100% minus RTP.
Why can observed RTP be higher than 100%?
Observed return is a finite sample. A winning sample can pay more than total turnover even when the theoretical RTP is 100% or lower.
Is effective RTP always the same as the published game RTP?
No. Allowances, automatic credits, rewards, strategy and payout caps can change the return that applies to a specific player or stake.
Can rakeback turn 99% RTP into 100%?
Only when the realized return equals the full 1% native edge on all relevant turnover without offsetting restrictions.
Does 100% RTP remove bankroll risk?
No. Variance and finite bankroll can still produce large drawdowns or complete loss.
Does a matching provably fair result prove zero edge?
No. It proves consistency with the tested outcome algorithm. The payout model must be audited separately.
Bottom Line
100% RTP is a mathematical expectation: expected gross payouts equal eligible turnover under the stated game and settlement rules. It is equivalent to 0% theoretical house edge.
The claim becomes meaningful only after scope is defined. Check native paytable math, account state, realized credits, strategy requirements and payout caps. Then keep outcome verification and platform reliability as separate audits.
