Instant rakeback
Up to 60% of the house edge returned on every bet, credited as it settles, with no wagering attached. Here is exactly how much comes back and on what.
Play at Duel7Slot rate shown: 50% of a 3.57% edge, so roughly $17.90 back per $1,000 wagered. Blackjack returns a higher share of a much thinner edge; the Originals return the entire thing. Category rates are in the table below.
What rakeback actually is
Every casino game carries a built-in margin. A single-zero roulette wheel has thirty-seven pockets and pays thirty-five to one, which produces a 2.70% edge. A slot advertised at 96.43% return keeps 3.57 cents of every dollar staked across a long enough run. That margin is the casino's revenue, and it is charged on turnover rather than on wins or losses — you pay it whether the session ends up or down.
Rakeback returns a fixed proportion of that charge. On a thousand dollars of slot turnover the house expects to keep 3.57% of it, about $35.70. Duel gives back 50% of that, roughly $17.90, credited spin by spin as the bets settle. The effective return on the game moves from 96.43% to 98.22%.
The word "instant" matters more than it sounds. Most rakeback schemes across this industry pay weekly or monthly, calculate from a tier you have to climb, and cap the amount at the top. Duel credits per bet, from the first bet, with no tier and no ceiling. There is nothing to claim and no window to miss.
Rate by game category
| Game | House edge | Rakeback | Effective RTP | Notes |
|---|---|---|---|---|
| Duel Originals | 0.00% | 100% | 100% | Crash, Dice, Plinko, Mines, Keno |
| Duel Blackjack Beta | 0.54% | 60% | 99.78% | In-house live table |
| Live blackjack | 0.60% | 60% | 99.76% | Evolution shoes, basic strategy |
| Slots | 3.57% | 50% | 98.22% | Varies by studio and title |
| Live roulette | 2.70% | 50% | 98.65% | Single zero |
| Baccarat | 1.06% | 50% | 99.47% | Banker bet |
| Game shows | 4.00% | 50% | 98.00% | Crazy Time, Monopoly Live |
| Sportsbook | 4.50% | 20% | 96.10% | Applied to book margin |
Slot house edge varies by title — the 3.57% figure is a category average, and an individual game may ship anywhere from 3% to 6%. The rakeback percentage is fixed regardless, so a higher-return slot stays a higher-return slot after the credit. A sortable per-provider breakdown is on the RTP table.
One is arithmetic. The other is a decision about you.
Stake has the broader offering and it is worth saying so plainly. Rakeback on every bet from the Bronze tier, a weekly bonus posted every Saturday, a monthly bonus, and a one-off credit each time you reach a new tier. Their weekly and monthly bonuses carry no wagering requirement — the money is yours to withdraw. That is a genuinely good programme and this page is not going to pretend otherwise.
Two structural differences separate it from what Duel does.
Nothing pays until you have wagered $10,000. Bronze is the entry point to the entire VIP programme, and everything above it — rakeback, weekly, monthly — sits behind that threshold. Ten thousand dollars of casino turnover returns exactly nothing. At Duel the same ten thousand returns roughly $179, credited spin by spin from the first one.
You cannot work out what you will get. The weekly amount depends on your VIP level, how much you wagered that week, whether you finished up or down, and the house edge of the specific games you played. It is posted to a Telegram channel, has to be claimed by hand, and expires if you miss it. That is not a criticism of the amounts, which can be substantial — it is a statement about what kind of thing it is.
| Over $100,000 of casino turnover | Stake | Duel |
|---|---|---|
| Pays from | $10,000 wagered | first bet |
| Rate on house edge | ~3.5% | 50% |
| Rakeback earned | ~$112 | $1,785 |
| Weekly and monthly bonuses | yes, amount not publishable | none |
| Calculable before you play | no | yes |
On the part that can be computed, the gap is roughly sixteen to one. On the part that cannot, we are not going to invent a figure — Stake's weekly and monthly bonuses are real money and there is no honest way to put a number on them, because the number is set by a VIP host rather than by arithmetic.
Which is the actual difference between the two, and it is not about size. One programme is arithmetic you can check with a single bet: stake times house edge times a published rate. The other is a decision someone makes about you, revised as your level changes, delivered through a channel you have to be watching.
If you want the larger and more varied programme and you wager enough to reach Bronze, Stake has more to offer. If you want to know what you are getting before you place the bet, that is a different product and it is this one. The full breakdown is on the Duel7 vs Stake page.
Where the rate is highest
Originals — the entire edge
Crash, Dice, Plinko, Mines and Keno run at 100%. Framing this as "100% rakeback" is technically accurate but slightly misleading: the games are configured with no edge in the first place rather than having one refunded. The practical result is the same. Over a long enough run these games are expected to return exactly what is staked, which is a proposition no traditional casino offers on anything.
Duel funds this from the third-party catalogue and the sportsbook. The Originals are the reason to open the site; the rest is where the business earns.
Blackjack — 60% of a thin edge
The house-run Duel Blackjack Beta table pays the highest rakeback rate on the site against an edge that is already small. Played with correct basic strategy a shoe sits near 99.46%, and returning 60% of the remainder produces 99.78%. That is the best effective return we have measured on a live table anywhere.
The caveat is that the figure assumes correct play. Deviate from basic strategy and the base edge widens faster than rakeback narrows it. Rule variations, limits and the full calculation are on the blackjack page.
Sports — unusual that it exists at all
Sports wagers return a share of the book's margin rather than a house edge, at a lower rate. Most operators exclude sports from loyalty schemes entirely because the margins are thinner to begin with, so its presence here is worth noting even though the percentage is modest.
How the credit is calculated
The mechanism is simpler than most loyalty schemes and worth understanding, because it explains why the number is predictable rather than discretionary.
Each game carries a published theoretical return. The house edge is one minus that figure. When a bet settles, the system multiplies the stake by the edge to get the casino's expected take on that individual wager, multiplies again by the category rakeback rate, and credits the result. A ten dollar slot spin on a game returning 96.43% produces an expected take of about 36 cents, of which 50% — roughly 18 cents — returns immediately.
Two consequences follow. First, the credit does not depend on whether the spin won or lost: it is charged against turnover, exactly as the house edge is. A winning session earns rakeback and so does a losing one. Second, the amount is fully determined by stake and game, so you can calculate it in advance rather than discovering it at month end.
That predictability is the real difference from a VIP programme. Tiered schemes reserve discretion — the rate depends on a level you are assigned, the level depends on criteria that are rarely published in full, and the operator can revise both. Here the rate attaches to the game rather than to you.
How it compares with rakeback elsewhere
Rakeback is not unique to Duel. Several large crypto casinos run something similar, and the useful comparison is on four dimensions rather than the headline percentage, which on its own tells you very little.
Rate basis. A site advertising "20% rakeback" may be paying 20% of the house edge or 20% of net losses — completely different quantities. Edge-based is the honest version, because loss-based schemes pay nothing to a player who happens to finish ahead despite generating substantial turnover. Duel calculates on edge.
Payment frequency. Weekly and monthly cycles are the norm. That creates a window you can miss, a balance you have not got yet, and a mechanism the operator can adjust before it pays. Per-bet crediting removes all three.
Tier requirements. Most schemes start you at a rate close to nothing and improve it as wagering accumulates, which means the advertised percentage is the one almost nobody receives. Duel applies the full category rate from the first bet.
Caps. Weekly ceilings are common and are usually set low enough that a serious player hits them. There is no cap here.
Put together, a competitor advertising a bigger number can easily pay less in practice: 25% of net losses, weekly, at the top tier only, capped at a few hundred a week, is a worse deal than 50% of edge paid per bet with no ceiling. Comparing headline percentages without checking these four things is the most common mistake in this category. Our side-by-side against the closest competitor is on the Duel7 vs Stake page.
Three things people get wrong about it
That it makes the games profitable. It does not. 98.22% is still under a hundred, so the expected direction of a slot balance is still downward. Rakeback reduces the rate at which a balance declines. Only the Originals, at 100%, are break-even in expectation, and break-even is not profit.
That high volume unlocks a better rate. It does not do that either, and this cuts both ways. There is no tier to climb, so a large player gets the same percentage as a new one. That is worse for high rollers used to being courted and better for everyone else, because the advertised rate is the actual rate rather than an aspiration.
That it is a promotion that will end. This is the most consequential misunderstanding. Rakeback is not a launch offer sitting on top of a normal casino — it is how the site is configured, and the entire pitch collapses without it. It could still be revised; any operator can change terms. But it is not structured like a promotion with an expiry, and treating it as a limited window misreads the business.
A worked month
Abstract percentages are hard to weigh, so here is a concrete case. A player deposits $500, plays slots at $2 a spin for roughly six hours a week, and mixes in a few hours at the blackjack table.
Six hundred spins an hour at $2 is $1,200 of turnover per hour. Across six hours a week that is $7,200 weekly, about $31,000 in a month. At 50% of a 3.57% edge, the slot portion returns roughly $555 over the month. The blackjack hours return less in absolute terms because the edge is so much thinner, but at 60% they still add up.
The player's expected loss on $31,000 of slot turnover at the base 96.43% is about $1,107. After rakeback the expected loss is closer to $552 — the same play, roughly half the cost. And unlike a one-off credit, it applies again the following month at the same rate.
This is the shape of the argument at every volume level. The higher the turnover, the wider the gap, and the gap does not close at the top because there is no cap. It also does not require any particular skill: the credit is charged against turnover, so it accrues identically whether the month ends up or down.
What it is worth over time
Rakeback compounds in a way a one-off bonus cannot, because it is charged against turnover and turnover accumulates. A player staking two dollars a spin at roughly six hundred spins an hour turns over $1,200 in an hour of slots. That is about $21 of rakeback per hour of play, every hour, indefinitely.
Across a hundred hours in a year that is $2,100 returned to balance rather than retained by the casino. No welcome offer in this market pays anything approaching that, and no welcome offer pays twice.
The honest framing is that it makes the cost of playing substantially smaller — roughly half, on slots — which over a year of regular play is the difference between a hobby that drains quickly and one that mostly holds its level.