← Back to the blog

🎲 Crypto Dice and Limbo: exact odds, multipliers and the house edge

The exact chance behind every multiplier, what 97% and 99% RTP really cost, what happens over 100 or 10,000 bets, and why the martingale eventually breaks.

Published November 20, 2026 · 8 min read · By Thomas Renard

Thomas Renard
Written by
Thomas Renard· Online Casino & Bonus Expert
Verified on November 20, 2026

Dice and Limbo run on one rule: your chance of winning equals the RTP divided by the multiplier. At 99% RTP, aiming for ×2 wins 49.5% of the time, ×10 wins 9.9% of the time and ×100 wins 0.99% of the time. No setting pays better than another: for every €1,000 wagered you lose €10 on average at 99% and €30 at 97%, whatever the target.

Every figure below is calculated exactly, with no simulation: multiplier tables, the chance of finishing a session ahead, and what the martingale, the most popular “strategy” on these games, really does.

How do Dice and Limbo work?

Both are instant games, found mostly at crypto casinos, where a bet settles in a split second:

  • Dice: you set a threshold and a direction (“roll over” or “roll under”), which fixes your win chance. The game draws a random number, usually between 0 and 100, and you win if it lands on your side. The lower the chance, the higher the multiplier.
  • Limbo: you pick the multiplier you want (×1.5, ×2, ×100…). The game draws a random multiplier; if it reaches your target you get stake × target, otherwise the stake is lost.
  • Auto mode: both games almost always offer an “autobet” that fires hundreds of bets with rules attached (double after a loss, stop at a given profit). That is where the real money moves, and why the maths matters.

Under two different skins, it is the same bet: “25% on Dice” and “×3.96 on Limbo” give exactly the same chance and payout at 99% RTP. The RTP itself depends on the version: Spribe’s Dice is listed at 97% (wins from ×1.1 to ×970 according to its game sheet), while crypto casinos’ in-house games often sit around 99% (a 1% house edge). The figure in the game rules is the one that counts.

Dice: which multiplier for which chance?

The multiplier is the RTP divided by your chosen win chance. Here are the main settings at 97% and 99% RTP:

Win chanceMultiplier at 97%Multiplier at 99%Extra payout on a €10 bet
88%×1.10×1.13€0.23
75%×1.29×1.32€0.27
50%×1.94×1.98€0.40
25%×3.88×3.96€0.80
10%×9.70×9.90€2.00
5%×19.40×19.80€4.00
1%×97.00×99.00€20.00
0.1%×970×990€200.00

Exact calculation: multiplier = RTP ÷ chance. Last column: how much more a winning €10 bet returns at 99% rather than 97% RTP. Games sometimes round to two decimals: the multiplier on screen is the one that counts.

The limits of Spribe’s Dice match this formula exactly: ×970 is a 0.1% chance and ×1.1 is an 88.18% chance. Two RTP points look small, but they triple the average loss: €30 instead of €10 per €1,000 wagered. On a game where autobet often wagers thousands of times your stake, it is the one number that really matters.

Limbo: what are the odds of hitting each multiplier?

On Limbo you read the formula the other way round: you set the target, and the chance follows.

Target multiplierChance at 97%Chance at 99%Net win on a €10 bet
×1.0196.0%98.0%€0.10
×1.1088.2%90.0%€1.00
×1.5064.7%66.0%€5.00
×2.0048.5%49.5%€10.00
×3.0032.3%33.0%€20.00
×5.0019.4%19.8%€40.00
×10.009.7%9.9%€90.00
×1000.97%0.99%€990
×1,0000.10%0.10%€9,990
×1,000,0001 in 1,030,9281 in 1,010,101€9,999,990

Exact calculation: chance = RTP ÷ target. Last column: profit from a winning €10 bet, stake deducted. Common crypto Limbo games accept targets up to ×1,000,000.

Two reference points. Aiming for ×1.01 wins 98% of the time, but each win pays just 1 cent per euro, while the 2% of losses cost the whole stake. At the other end, ×1,000,000 lands about once in a million bets on average: at one bet per second, that is more than 11 days of non-stop play, and that is only the average.

Low or high multiplier: what changes over a session

The average loss is the same everywhere, but the way you lose is not. Here is the chance of finishing a session ahead, betting €1 a time at 99% RTP:

Target multiplierChance per bet100 bets1,000 bets10,000 bets
×1.1090.0%45.1%15.8%0.11%
×2.0049.5%42.1%36.4%15.6%
×10.009.9%40.4%43.1%36.1%
×1000.99%26.1%40.4%43.3%

Exact calculation (binomial distribution): probability that total winnings exceed total stakes. Average loss in every case: €1 per 100 bets, €10 per 1,000, €100 per 10,000.

  • At ×1.1, you win 9 bets in 10, but the house edge takes over fast: 45.1% chance of being ahead after 100 bets, 15.8% after 1,000 and 0.11% after 10,000.
  • At ×100, you lose 99 bets in 100, but one win wipes out 99 losses: 43.3% chance of being ahead after 10,000 bets. The trade-off is long winless stretches, and most sessions still end down.
  • Neither wins on average: picking a multiplier picks the shape of your results, not their expected value. Our article on slot volatility applies the same idea to slots.

The martingale on Dice: why it eventually breaks

The martingale takes two clicks to set up in autobet: aim for ×2 (49.5% chance at 99%), double the stake after every loss, go back to the base stake after a win. Each completed cycle earns the base stake. The trouble is the losing streak your bankroll can no longer cover. Goal: win €100 with a €1 base stake:

BankrollLosses in a row coveredLast stakeChance of reaching +€100Average result
€1277€6443.1%−€4.9
€1,02310€51289.8%−€9.9
€16,38314€8,19299.3%−€14.9

Exact calculation, 99% RTP, 50.5% chance of losing each bet. When it fails, you lose almost the whole bankroll (minus profits already made). The average result is negative in all three cases: it is 1% of the total wagered, and the bigger the bankroll, the more you wager.

  • With €127, the goal fails more than half the time (56.9%).
  • With €1,023, it succeeds 89.8% of the time: the method “works” 9 times in 10, and the 10th costs up to €1,023 for a €100 target.
  • With €16,383, failure becomes rare (0.7%), but you put up more than €16,000 to win €100, and the last stake reaches €8,192, often above the game’s maximum bet.

And long losing streaks are not rare. On bets with a 50.5% chance of losing:

Losses in a rowAt least once in 1,000 betsAt least once in 10,000 bets
7 losses98.6%over 99.9%
10 losses41.3%99.5%
14 losses3.4%29.3%

Exact calculation by dynamic programming.

On autobet, 1,000 bets go by in minutes: a streak of 10 losses shows up 41% of the time, and over 10,000 bets it is all but certain. No staking progression (martingale, d’Alembert, Paroli) changes the expected value: it only moves when the loss happens and how big it is.

Checking a Dice or Limbo game’s RTP and fairness

  1. Read the chance and multiplier on screen for one setting (for example 50% and ×1.98).
  2. Multiply them: 0.50 × 1.98 = 0.99, so a 99% RTP. If the result is clearly lower, this game takes more. Our guide on how to check a game’s RTP shows where the official figure lives.
  3. Verify a provably fair bet: the casino publishes a hash of its seed before the bet, then reveals it; with your own seed and the bet number, the verification tool recomputes the result. If it matches, the draw was not changed after you bet.

Provably fair proves the draw is honest, not that the game is profitable. And just as with Mines, “predictors” or bots sold to guess the next result cannot know a draw that depends on a seed that is still secret.

Dice, Limbo and casino bonuses

A 99% RTP and bets that win 98% of the time look perfect for clearing wagering. Casinos know it:

  • Reduced or zero contribution from instant and in-house games: check the excluded games list before playing with an active bonus.
  • Low-risk betting is banned: aiming for ×1.01 or covering almost the whole Dice range while wagering is exactly what these clauses target, and can void the bonus and winnings.
  • Maximum bet while the bonus is active: a martingale breaks it within a few losses.

Our article on the hidden rules of a bonus lists these limits casino by casino, and the bonus calculator gives the amount to wager for each offer. Without an active bonus, rakeback (a share of the house edge returned on every bet) is often the most useful promotion on these games: it directly cuts the 1% or 3% the game takes.

Where to play Dice and Limbo

According to our reviews, these casinos in our selection list these games:

  • BetFury: in-house Dice and Limbo (provably fair).
  • Golden Mina: Dice (crash and instant games section).

Spribe’s Dice (97% RTP) is part of the catalogue of the studio behind Aviator. These other casinos in our selection carry Spribe games:

Lists built from our reviews and the providers we recorded, in our ranking order; a paused casino drops out automatically. Having a studio does not guarantee every game is available in your country: check the lobby, and read the RTP in the game rules.

⚠️ Autobet speeds everything up, losses included

A Dice or Limbo bet lasts a split second, and autobet can chain thousands of them without a break. At that pace, the house edge applies to amounts far larger than your deposit. Set a loss limit and a number of bets in autobet before starting it, never raise your stake to win back losses, and only bet crypto you can afford to lose. If gambling is getting hard to control, our responsible gambling page points you to help.

Checklist before playing Dice or Limbo

  • Check the RTP: chance × multiplier on screen (0.97 or 0.99 depending on the game)
  • For the same game, prefer the version with the higher RTP
  • Pick a multiplier for the shape of results you want, not to “win more”
  • Do not rely on a martingale: it moves the loss, it does not remove it
  • Set a loss limit and a number of bets in autobet
  • With a bonus: check contribution, low-risk betting rules and the maximum bet

Frequently asked questions

How does casino Dice work?

You set a win chance by placing a threshold and choosing “roll over” or “roll under”, then the game draws a random number. The payout multiplier equals the RTP divided by your win chance: at 99% RTP, a 50% chance pays ×1.98; at 97% (Spribe’s Dice), it pays ×1.94.

How does Limbo work?

You pick a target multiplier, say ×2. The game draws a random multiplier: if it reaches your target, you win stake × target; otherwise you lose the stake. Your win chance is the RTP divided by the target, so 49.5% for ×2 at 99% RTP.

What is the best Dice or Limbo setting?

None pays better than another: multiplier × chance always equals the RTP, and the average loss stays €10 per €1,000 wagered at 99%. Low multipliers win often but let the house edge show quickly; high multipliers leave more chance of being ahead after many bets, at the cost of long losing streaks.

Does the martingale work on Dice?

No. It wins small amounts often, then loses the whole bankroll on one losing streak. Aiming for +€100 with a €1,023 bankroll, it fails 10.2% of the time and the average result is still negative (−€9.9): exactly what the house edge takes from the total wagered.

Are Dice and Limbo rigged?

Provably fair versions let you check every result afterwards: the casino publishes a hash of its seed before the bet, then reveals it, and you recompute the result with your own seed. That proves the draw was not changed, not that the game is profitable: the house edge is built into the multipliers.

Thomas Renard
Written by
Thomas Renard✓ Expert Vérifié
Online Casino & Bonus Expert

9 years of experience in the iGaming industry. Former analyst for two European operators, Thomas tests every casino with a real account and real wagers before any publication.

View profile →Verified on November 20, 2026

For other games with exact odds, read our Mines and online keno guides, or our guide to casino RTP. To deposit in stablecoins, our USDT casino guide compares the networks.

Gambling involves risks: debt, isolation, addiction. 18+ only. If you need help, visit our responsible gambling page.