Cloudbet review: 40 coins, two refused countries, 8.5

This Cloudbet review reports the highest rating our formula has produced for any record, jointly with one other: 8.5 out of 10, earned across six scored components of seven. Four numbers explain most of it – a licence row that survives a register lookup, a coin list of forty settlement currencies where the library median is nine, a restricted-country list of exactly two entries, and a verification clause with a number on it but no figure inside it. One caveat belongs in the first paragraph rather than a footnote: this record was read from a stored copy of the operator’s pages rather than live, on 26 August 2026, and a Cloudbet review built on a stored copy inherits whatever was true of those pages when they were captured.

What this Cloudbet review is built on, and its one weakness

The Cloudbet record carries retrieved: archived – meaning the same pages were read, but from a stored copy rather than fetched live on the day. Forty-one of our hundred records are in that condition, thirty-eight were read live, and twenty-one could not be read at all.

It matters here more than usual because this is a high-scoring record. A live reading and a stored reading produce the same kind of statement about a clause, but only the live one is a statement about the clause as served today. So every figure in this Cloudbet review should be read as: this is what those pages said when they were captured, checked 26 August 2026, with the location-masking position checked a day later on 27 August 2026.

The second source is the register lookup, made the same day and keyed by domain, which returned a full certificate for cloudbet.com. The third is the score sheet, regenerated 17 September 2026. The record also carries an established year of 2013, one of only eleven records in the library that has one at all – a field, not a claim we can check further, and nothing in our arithmetic depends on it.

The licence row, confirmed twice on the same day

Licence number OGL/2024/328/0599, holder Halcyon Super Holdings B.V., status Active. The register lookup adds two fields the operator record leaves empty: company number 148526 and a grant date of 23 May 2025.

That is the useful case of reading two files: they agree on the number and the status, and one fills in what the other omits. Compare it with the rows where they do not agree – a domain found in neither register, a brand named under two different companies, a row that reads valid past its own end date – and the value of asking twice becomes obvious. The full audit of those disagreements is at where the two registers contradict each other.

The licence component takes the full 2.5 of 2.5 here, the largest single weight in the formula, with the note that the number is current in the register as read. There is no Anjouan row for this domain, so the licence evidence is one number from one source, cross-checked against a second lookup of that same source. This Cloudbet review does not treat that as stronger than it is: one current certificate, read once, on a stated date.

Forty coins – the widest list in a hundred records

Cloudbet’s coin list is the most extreme single field in the library. Forty settlement currencies are recorded: BTC, BCH, ETH, USDT, USDC, USDP, PAXG, DAI, DOGE, LTC, DASH, LINK, DOT, ADA, UNI, SOL, ALGO, AVAX, POL, SHIB, BNB, XLM, XRP, STETH, BSV, EOS, TRON, ZEC, TON, ENA, USDE, SUSDE, BRETT, DEGEN, PONKE, TOSHI, DOGS, HMSTR, HBAR and TRUMP.

For scale, and it is the one place where this Cloudbet review reports an outlier rather than a middling figure: the next widest list in the file has nineteen entries, then fifteen, and only thirty records carry a coin list at all. The component is worth 1.5 points and forty takes all of it, but the interesting part is not the score. It is the composition. Eight of the forty are dollar-referenced tokens of one kind or another and several are small-cap tokens, which tells you the operator has chosen breadth of deposit rails as a selling point. Nothing in our files says what the withdrawal path is for each of the forty, and this Cloudbet review does not assume that deposit breadth and withdrawal breadth are the same list.

Nothing in the record states a withdrawal ceiling either: the limits block is entirely null – no figure, no period, no clause. Fourteen records in the library do publish one, and the fact that this one does not is a gap rather than an absence of limits. Those fourteen, with their clause numbers, are at the written withdrawal ceilings, set side by side.

Clause 18.1.1: a right reserved, not a threshold

Cloudbet’s verification summary is the longest in our file and it is worth reading as written. The policy requires customers to provide reliable independent identity documents – identity document, passport, bank statement, utility bill are the examples given – and the company reserves the right to verify any customer it deems risky or suspicious, and may request updated documents at any time, even after a customer has already been verified. The clause reference is 18.1.1, and the page it was read from is the operator’s own German-language help terms.

Three things follow. First, there is no figure: no balance triggers the check and none exempts it, so the verification-clarity component scores 1.5 of 2 – full marks for a clause reference, short of full for having no number in it. Only five records in the entire library state a numeric trigger. Second, the check is explicitly reserved from the start, so the no-verification-at-sign-up component scores 0 of 1. Third, the phrase even after prior verification means being verified once is not a completed state; it can be reopened.

A reader deciding between records should weigh that against the alternative, which is silence: fifty-one records say nothing readable about verification at all. A named clause with no figure is better evidence than no clause, and worse evidence than a figure. What each of these clauses commits an operator to, and what it does not, is worked through at the guide to reading a verification clause before depositing.

Two refused countries, and a confiscation note

The Cloudbet restricted-country list holds two entries: AU and BE. GB is not among them. The list was read from a page on the operator’s own site under a beta path, which is worth printing because a clause read from a beta path is a slightly thinner source than one read from the published terms.

Two entries is the second-shortest list in the library that is longer than a single code, and it earns both country components in full: 1.5 for publishing a list at all and 1.5 for reach, since refusing two countries is the opposite of the seven records that list more than 215. In this Cloudbet review those three points are a third of the total score, and they come from a field two codes long.

The record also carries a location-masking position, one of only three in the library: the practice is recorded as prohibited, with balance confiscation noted as the consequence for play from a prohibited country, checked 27 August 2026. No clause number is attached to that note, which makes it the weakest-sourced fact on this page, and it is printed because a confiscation clause is a payout fact whatever else it is. This site publishes no methods for disguising a location and holds none in any file. The other two records with such a clause, and the full country-list arithmetic, are at the country lists, sorted by length.

How 8.5 is put together

The full arithmetic, so that the number can be recomputed or rejected:

Six scored components is the most any record in the library manages, and it is the real reason this Cloudbet review ends at the top of the table: not that it is remarkable in any single field, but that six of its seven cells were readable. Fifty-three records of a hundred have too few readable cells to be rated at all, and the minimum before a rating is printed is three.

Two corollaries worth stating plainly. The record loses the full point for demanding verification from the start, so a perfect score is not reachable for any operator that reserves a document check – by design. And the missing 1.5 points come from a clause with no number in it, which means the fastest way for this record to improve is not a better offer but a more specific sentence in the terms. One other record shares the 8.5, and the whole ranking is on the front page: the crypto casino UK table this record heads.

The other 8.5, and what happened to its term

One other record shares this rating. DuckDice also scores 8.5 on six components: licence 2.5 for ALSI-202509013-FI1 held by Zentari Limitada and current in the register, 1.5 for publishing a sixteen-entry country list, 1.5 for reach, 1.5 for a ten-coin list, 1.5 for a verification clause given as Section 2 with no figure, and 0 for reserving a check from the start.

The difference between the two records is what their licence rows do as they age. That one comes from a register that prints a term: issued 16 September 2025, ending 15 September 2026. Our reading was made on 26 August 2026, three weeks before that end date, and nobody has re-read it since. The certificate quoted on this page prints a grant date and a current status and no end date at all, so it does not expire on paper – it simply stops being current whenever it stops, without announcing it in advance.

Neither arrangement is better for a reader; they fail differently. A term gives you a date to check and goes stale loudly. A status word with no term never goes stale on its own and has to be re-read to mean anything. The joint top of our table happens to hold one of each, which is a neat accident and a useful warning: two identical ratings can rest on evidence with completely different shelf lives.

Six records in the library have passed their printed end date since our reading, and that is the column to watch when comparing crypto casinos a month after any table was published.

What the record does not cover

No payout processing window: that cell is empty here, as it is for ninety-nine of a hundred records. No withdrawal ceiling. No provider list. No figure anywhere in the verification clause. And the whole reading comes from a stored copy rather than a live fetch, with a location-masking note that carries no clause reference.

Set against that, six readable components, a licence number confirmed by two lookups on the same day, forty settlement currencies, a two-entry country list and a clause number for the document check. A top score in this library means the reading was unusually complete, not that the operator was unusually good.

The comparison that shows what the same formula does with a sparser record is at the record that scores nothing on two readable components, and the shortlist these readings feed is at the bitcoin shortlist drawn from these readings.