Cashier clauses, read 26 August 2026
What these documents say a transfer costs, and where the cost actually hides
Not one of the twelve rows compared here publishes a fee schedule, and no fee figure exists anywhere in the library of a hundred operators. What is published is a set of ceilings, and a ceiling quietly decides how many separate withdrawals a balance has to become.
Every column on this site has a denominator. This one has no numerator.
Nought of twelve, and nought of a hundred
No fee schedule was found for any of the twelve operators compared here, and none exists anywhere in the wider library of a hundred. Not a percentage, not a flat amount, not a free-withdrawals-per-month allowance, and not a statement that transfers are free.
That is a blank on this site, and a blank has one meaning: the document behind it was not opened, or was opened and did not contain the figure. It is not a zero, and reading it as one would hand a hundred businesses a favour manufactured out of our own reading list.
It is also, by a distance, the widest gap on any page here. Territory lists were read at 38 operators, verification clauses at 49, payout ceilings at 14. Fees, nowhere.
Where the money actually goes on the way out
Three things are published often enough to be worth reading, and none of them is a fee. Together they describe the cost of a withdrawal better than a fee schedule would on its own.
The first is the ceiling. Seven of the twelve rows attach one to a numbered clause: 4,000 EUR a day at Oshi under clause 12, 7,500 EUR a week at PlayAmo under clause 12, 10,000 EUR a week at Mirax under clause 8, 10,000 USD per transaction at Bets.io under clause 10, 15,000 EUR or USD a week at Slotum under clause 13, 20,000 EUR a month at BitDreams under Section 11, and 50,000 USDT a month at CryptoLeo under clause 7.9. The BitStarz cell is blank.
The second is the processing period, published by one row of the twelve: up to three working days, clause 8.1.
The third is the instalment rule, published by the same row: anything above 50,000 USDT is paid in parts, clause 8.8, with the size of a part unstated.
Each of the three converts a balance into a number of events. That conversion is where an unpublished fee would be applied.
A ceiling is a multiplier
Take a balance of 20,000 units and run it through the ceilings above.
| Operator | Ceiling | Clause | The balance becomes |
|---|---|---|---|
| Bets.io | 10,000 USD per transaction | 10 | 2 transfers |
| Oshi | 4,000 EUR a day | 12 | 5 transfers, on 5 days |
| PlayAmo | 7,500 EUR a week | 12 | 3 transfers, over 3 weeks |
| Mirax | 10,000 EUR a week | 8 | 2 transfers, over 2 weeks |
| Slotum | 15,000 EUR or USD a week | 13 | 2 transfers, over 2 weeks |
| BitDreams | 20,000 EUR a month | Section 11 | 1 transfer |
| CryptoLeo | 50,000 USDT a month | 7.9 | 1 transfer |
Two ends of the same table, one balance, and the number of separate transfers runs from one to five. Take the balance to 100,000 and the Oshi row becomes 25 transfers across 25 days.
If a transfer costs something — an operator's fee, a chain's fee, or both — the cost is paid once per row of that right-hand column. So the ceilings are the closest thing these documents contain to a fee schedule, and the operator that publishes the tightest ceiling is the one whose customers pay whatever the unpublished figure is most often.
The Bets.io shape is worth separating from the rest. Its cap is per transaction rather than per period, which is the only one of the seven that does not slow a payout down at all: two transfers on the same day satisfy it. It multiplies the count of transfers without stretching the calendar, and it is the clearest example of a ceiling doing something other than what a reader assumes ceilings do.
Where those seven figures sit against each other, and why four different windows resist being flattened into a ranking, is on the withdrawal limits.
The cost nobody in this pool controls
A crypto transfer carries a network fee, set by the chain and paid out of the transfer, and it is invisible in every contract read for this site.
No entry in the library names a network at all. Not the choice between two versions of the same stablecoin, not a minimum number of confirmations, not a statement of who bears the cost. That absence is a reading gap and also a habit of these documents, which describe money in units of account and leave the plumbing to pages that can be edited quietly.
For a reader it is the practical part. The chain fee is charged per transfer, so it lands on the same multiplier as everything above, and it is the one cost that does not care which brand is being used.
Why this column was attempted at all
Because an empty column is a finding when it is empty everywhere.
A field that half the pool publishes is a good discriminator. A field that nobody publishes says something about the trade: the price of moving money out is the least documented part of these arrangements, and it is documented less thoroughly than the price of a bonus, the size of a bet, or the countries an operator refuses.
That is the honest end of this article. There is no fee comparison here, there is no fee comparison available from documents, and a page elsewhere presenting one is presenting something other than a reading of the operators' own terms.
The person doing the reading, and what these pages refuse to claim, is described on the comparisons editor.