What a 99% RTP Dice Game Actually Costs
This page published a breakdown of the “real” house edge behind a 99% RTP dice game: +0.42% from payout rounding, +0.65% from minimum bet increments, +0.81% from auto-bet rounding and a penalty for reusing a client seed. Those numbers were never measured, and as an accusation aimed at an entire category of operators they should not have been published. The honest version is less dramatic and more useful, because the real cost is not hidden at all.
Where the money actually goes
| What the old page claimed | What is actually the case |
|---|---|
| Payout rounding adds +0.42% | Balances are kept to eight decimals. On a $1 bet that is a rounding step of about $0.00000001 — six orders of magnitude below the claim. |
| Minimum bet increments add +0.65% | A minimum bet limits how small you can go; it does not skim a percentage from each settled bet. |
| Auto-bet rounding compounds +0.81% | Auto-bet repeats the same settlement logic faster. Speed multiplies turnover, and turnover is the real cost — see the line below. |
| Reusing a client seed carries a penalty | It carries no mathematical penalty at all. The outcome depends on the server seed, which you cannot predict — see the provably fair mistakes piece. |
The one number that does cost you
A 99% RTP means the house keeps 1% of every amount staked, every time. Ten thousand rolls of $1 put $10,000 through the game and cost about $100 on average — and on auto-bet ten thousand rolls take under three hours. Nothing is concealed in that sentence: the edge is published in the game’s own info panel, and the multiplier you pick sets the win chance around it.
The second real cost is bonus turnover. A match bonus with a wagering requirement of, say, 40× on $500 asks for $20,000 of staking before withdrawal, which at 1% carries an expected cost of about $200 — which is why the bonuses page lists the wagering figure as published and refuses to invent it. Third come network fees and exchange spread on crypto deposits and withdrawals, which are outside the game entirely. What doubling does to turnover is in the Martingale study, the impossible version in the “$200 into $9,400” story, and pacing in bankroll management.
Is 99% RTP a marketing trick?
No, it is the published figure and it is accurate per bet. The trap is turnover: 1% of a lot of bets is a lot of money.
Does rounding really cost anything?
At eight decimals it is negligible — far below any figure worth putting in a table.
How do I estimate my own cost?
Multiply your average bet by the number of bets, then by the house edge. That is the expected cost of the session before variance.
