A game page may mention a random number generator (RNG), return to player (RTP) or an independent test. Those terms are not interchangeable. An RNG concerns how random outcomes are generated, RTP describes a theoretical return for a particular game configuration, and a test report records work performed within a stated scope.
None of them, by itself, proves that an operator is licensed, that withdrawals will be handled well, that the game version currently served is the version tested or that gambling is financially safe.
1. Separate the three questions
| Term | Question it can help answer | What it does not establish alone |
|---|---|---|
| RNG | Does the tested system generate or select outcomes using the specified random process? | The game’s theoretical return, the operator’s identity or the behaviour of the live deployment. |
| RTP | What theoretical proportion of stakes is the specified game configuration designed to return over its modelled play? | What one person will receive in a session, or whether a win or loss is “due”. |
| Test report or certificate | What product, version, configuration, criteria and result were covered by a named test body at a stated time? | Every game, every later software version, every operator control or the legality of play in the reader’s location. |
Start by identifying which question a public claim actually addresses. “RNG tested” is not an RTP figure. A displayed RTP percentage is not proof of an independent test. A certificate logo without a traceable scope is not enough to match evidence to the game in front of you.
2. Understand what theoretical RTP means
RTP is normally a theoretical property of a specified game and configuration. It is an expectation across the game’s mathematical model, not a promise to an individual player and not a forecast for a short session.
For example, a published 96% theoretical RTP should not be read as “every £100 produces £96 back”. Actual session results can differ sharply because outcomes vary. It also does not mean the player will recover earlier losses by continuing to play.
For relevant products licensed in Great Britain, the UK Gambling Commission’s Remote Technical Standard 3 requires accessible rules and information that can include RTP, house edge or the probability of winning before gambling begins. That is a rule for the regulator’s licensees and applicable products; it should not be presented as a universal requirement for every website worldwide.
Check the value belongs to the game you can access
A useful RTP check records:
- The exact game title and provider shown on the page.
- The software or game version, where disclosed.
- The RTP percentage and whether it is labelled theoretical.
- Any selectable stake, feature, jackpot or configuration that affects the figure.
- The jurisdiction, channel or platform covered by the source.
- The date on which the information was checked.
Do not transfer a percentage from a similarly named game, another version or an unrelated operator page. ORC’s games directory and provider pages such as Pragmatic Play and Evolution can help identify public labels, but a provider category is not evidence of a particular game’s RTP or certification.
3. Ask what the RNG evidence covers
“Uses an RNG” is a description, not a complete test result. The useful questions are which component was assessed, under what technical standard, in which software version and with what outcome.
The UK Gambling Commission’s Remote Technical Standard 7 sets requirements for random outcome generation within its regulatory scope. Its separate testing procedure describes information expected in test reports, including the game, software number or version, digital signature where applicable, RTP, platform, test scope and result.
That level of detail matters because an RNG component test and a complete game test may have different scopes. A report for one release should not automatically be applied to a later release unless the evidence says it remains covered.
4. Read a certificate as a scoped record
A certificate is most useful when a reader can trace it to a genuine issuer and match its identifiers to the product being assessed. A logo or screenshot can be copied; verifiable details are harder to substitute.
Look for the following fields:
- Issuer: the full name of the laboratory or test house.
- Client: the developer, supplier or licensee whose product was tested.
- Reference: a report or certificate number that can be checked.
- Product identity: exact game title, software number, build, version or digital signature.
- Configuration: RTP, platform, channel and any relevant variation.
- Test basis: the jurisdiction, technical standard and tests applied.
- Result and scope: what passed, what was excluded and whether conditions apply.
- Date and status: when the work was completed and whether a later record supersedes it.
If those details are missing, record the public claim as unverified. Do not fill the gap with an assumption based on the site’s design, HTTPS certificate or reputation.
5. Verify the test house separately
Check the test body through the regulator or accreditation source relevant to the claim—not only through a link or badge supplied by the operator.
The UK Gambling Commission maintains an approved test-house list for the testing work in its regime. Other jurisdictions use their own approval processes, so appearing on one list does not automatically establish authority everywhere.
Some laboratories also refer to ISO/IEC 17025, a standard covering the competence, impartiality and consistent operation of testing and calibration laboratories. Accreditation can be relevant, but its schedule and scope still matter. It does not mean that every service associated with the laboratory—or every game using its logo—has been tested.
6. Match the evidence to the live game
Even an authentic report can be irrelevant if it covers a different game build. Compare the record with information available in the live game’s rules, information screen or version display.
| Match point | Question to record | Stop condition |
|---|---|---|
| Game identity | Do the title and provider match exactly? | The report names a different or ambiguous product. |
| Version | Does the software number, build or signature match? | No live version can be found or the values differ. |
| RTP | Does the tested percentage match the displayed configuration? | The game shows another value or the configuration is unknown. |
| Platform and channel | Does the report cover the web, mobile or other channel used? | The tested platform is different and equivalence is not documented. |
| Status | Is the record current and not superseded? | The source is expired, withdrawn or replaced. |
If a match cannot be made from public evidence, the accurate conclusion is “not verified from the reviewed material”. It is not evidence that the game is unfair, but it is also not a reason to repeat the claim as fact.
7. Testing does not replace live monitoring
Pre-release testing is a point-in-time assessment. Configuration mistakes, integration changes or issues that were not detected during testing can still affect a live product.
The UK Gambling Commission’s live RTP monitoring guidance explains why licensees within its scope must monitor live performance and investigate significant deviations. The Commission’s testing-strategy summary also separates test-house work from the licensee’s continuing responsibilities.
Users generally cannot inspect private monitoring systems. Public reviewers should therefore distinguish a regulator’s requirement from evidence that a particular operator actually complied with it during a particular period.
What fairness evidence cannot prove
RNG, RTP and testing evidence should not be used as shortcuts for unrelated questions. They do not independently establish:
- The legal identity of the operator.
- A valid licence for the reader’s location.
- Fair bonus terms or reliable withdrawal handling.
- Good customer-support performance.
- Secure account, payment or personal-data systems.
- Responsible-gambling safeguards.
- That a game is affordable or suitable for the individual.
- Future results or a way to recover losses.
Those topics require separate evidence. ORC’s How We Review methodology explains why public observation, operator claims and independently verifiable records are labelled separately.
A practical fairness-evidence workflow
- Record the exact game, provider, operator domain and date checked.
- Open the game rules or information screen before playing.
- Record the displayed theoretical RTP and configuration.
- Find the report or certificate reference—not just its logo.
- Verify the issuer through the relevant regulator or accreditation source.
- Match title, version, signature, RTP, platform and jurisdiction.
- Check whether the record is current or superseded.
- Keep unverified items labelled as unresolved.
- Assess licensing, operator identity, payments and responsible-gambling controls separately.
Final takeaway
Good fairness evidence is specific and traceable. It identifies the game, configuration, version, test body, standard, result and date, and it can be matched to the live product. RNG, RTP and certificates each add a different piece of that record; none is a universal “safe casino” badge.
Sources
- Online Review Casino — How We Review
- Online Review Casino — Games Directory
- Online Review Casino — Pragmatic Play Games
- Online Review Casino — Evolution Games
- UK Gambling Commission — Remote Gambling and Software Technical Standards
- UK Gambling Commission — RTS 3: Rules, Game Descriptions and the Likelihood of Winning
- UK Gambling Commission — RTS 7: Generation of Random Outcomes
- UK Gambling Commission — Testing Strategy Summary
- UK Gambling Commission — Procedure for Testing
- UK Gambling Commission — Live RTP Monitoring
- UK Gambling Commission — Approved Test Houses
- ISO — ISO/IEC 17025:2017