Methodology
The arithmetic behind every number on this site
Five checks, read from each operator's own terms and licence register, each worth one point — nothing is estimated and nothing is guessed.
Every number on this site traces back to a document you could read yourself. That's the whole point of publishing the method separately: nothing here is a house opinion dressed up as a score.
The five checks
Each of the 12 operators in the main comparison is checked against five things, each worth one point out of a possible 5, all read from that operator's own published terms of service and its entry in whichever licensing authority's register it names.
- Does an identity-verification clause exist at all in the operator's terms of service.
- Does that clause state when checks can start, before a first deposit, only before a withdrawal, or left open-ended.
- Does the clause tie checks to a specific amount, such as a deposit or withdrawal threshold, rather than leaving the trigger entirely at the operator's discretion.
- Is a deadline set for supplying documents once a check is triggered, giving a player a defined window rather than an open-ended wait.
- Does the licence number the operator quotes resolve under that operator's own domain in the issuing authority's own public register, confirming the licence is actually held by the site you'd be using rather than a related entity.
Bitcasino.io, Rocketpot, and Wild Fortune each score 4 out of 5 on this set; TrustDice scores 0, meaning none of the five items were found published in what we read. A score is not a ranking of trustworthiness in some general sense; it's a narrow measure of how much a given operator has published about its own verification practices and how well its stated licence checks out against the register.
Why a blank field scores nothing, not a penalty
If a clause isn't found in the terms we read, that field is left blank rather than marked as a failure. Those are two different claims. A blank means the document didn't contain it, or we didn't find it there; it does not mean the operator has no such policy, only that nothing on the point has been published where we looked. Treating a blank as an active strike would conflate an operator that has simply not published something with one that is actively obscuring it, and those are different problems that deserve different treatment. This is also why a low score should be read as “less has been published,” not as an accusation.
Vave's placement is separate from the score
Vave's own terms are still read the same way as any other operator's, clause 8.7 covering identity verification, and that detail lives on the Vave review rather than in the scored table.
What's read, and what's actually tested
The great majority of what's on this site comes from reading documents: terms of service, promotions pages, and licence registers, each with the date it was checked. That's true of all 12 operators in the scored comparison and of the 18 further names listed without scores. Two operators are the exception. Vave was tested with the owner's own money in August 2026: a deposit and withdrawal of roughly 5,000 USDT that went through without an ID request, described in full on the Vave review. TrustDice was separately tested in May 2026, also with the owner's own funds: withdrawals cleared quickly for several weeks until cumulative withdrawals reached roughly 5,000 USDT, at which point a withdrawal went to review, a compliance email cited Veriff verification, a driving licence was rejected, and a passport was accepted on the fifth attempt, followed by a second document request sent directly by email. Those two tests are real, first-hand, and limited to those two operators; everything else on this site is a document read, and we say so plainly rather than letting the distinction blur.