Each week, we may have 20 to 30 new online slots to consider for our casino news report. We do not expect every member of our team to play every game. Instead, the releases are divided across our nine experienced slot testers. We divide the number of slots by our team members who them play the demo version, which uses the same RNG as the real money version of the game.
Every tester follows the same testing routine and scores the game independently. We then combine the human scores with a smaller AI-assisted testing score using the fixed formula explained below. The ten highest-scoring eligible games form that week’s Top 10.
Our rankings represent the collective opinions of our testing team. They are not official industry ratings, guarantees of quality or predictions of how much anyone might win. Other players may reasonably give the same games different scores.
Important: Our Casinoplusbonus ranking process is an informed editorial assessment, not an official industry rating or a prediction of how much anyone will win. Different players will naturally value different themes, mechanics and bonus features.
How Our Weekly Slot Testing Works
1. We Identify the Eligible New Releases
We begin by compiling the slots released during the period covered by that week’s report. A game must be available through a working demo or approved testing environment and provide enough information for us to assess its rules, features and published game data.
Our weekly Top 10 is therefore selected from the eligible new games we were able to test for that report. It does not mean we tested every online slot available worldwide, nor does it claim that these are the ten best slots ever released.
2. We Distribute the Games Across the Team
Every eligible game is assigned to at least three members of our nine-person testing team. Assignments are rotated to reduce the influence of individual preferences for particular themes, providers, volatility levels or feature types.
For example, if there are 30 eligible releases, three testers per game creates 90 individual assignments. This works out at approximately ten games for each team member. Nobody needs to play all 30 releases, but every game still receives the same minimum level of human testing.
Reviewers record their scores before seeing the ratings submitted by other team members. This helps prevent one person’s opinion from influencing everybody else’s assessment.
3. We Complete a Standard Number of Spins
Each assigned human tester completes at least 200 base-game spins. With three testers assigned to a slot, that produces a minimum of 600 human-controlled base-game spins per release.
Our normal testing target for each compatible slot consists of:
Testing is normally completed in provider demo mode or an approved testing environment. If regular spins do not provide enough access to a bonus round, we may use a demo bonus buy, feature purchase or provider testing tool where it is available and permitted.
Purchased or directly activated features are logged separately. We use them to examine how a bonus works, not as evidence of how frequently it occurs during ordinary play or how profitable the game may be.
4. We Investigate Major Differences of Opinion
Three people will not always agree about the same game. If the highest and lowest human scores differ by more than 2.0 points out of 10, we assign the slot to a fourth human tester.
We also review incomplete sessions, technical failures and scores that do not appear consistent with the recorded testing notes. The additional review is completed before the final ranking is calculated.
The Five Online Slot Rating Categories
Every tester scores the same five categories from 1 to 10, using half-point increments. A score of 5 represents an ordinary but functional release. A score of 7 indicates a strong game, while 9 is reserved for an exceptional release. Scores of 10 should be extremely rare.
This category is based on published game information and the transparency of the design. It is not based on whether our testers won or lost during a short session.
Our Online Slot Scoring Algorithm
We calculate each human reviewer’s weighted score using the following formula:
Reviewer score = (Gameplay × 0.25) + (Bonus Features × 0.25) + (Presentation × 0.20) + (Originality × 0.15) + (Player Value × 0.15)
We then calculate the average of all valid human scores:
Human score = Total of all valid human reviewer scores ÷ Number of human reviewers
The AI-assisted assessment applies the same five-category scoring structure but has a deliberately limited influence on the result:
Final score = (Human score × 0.90) + (AI score × 0.10)
For example, suppose a game receives an average human score of 8.16 and an AI-assisted score of 7.70:
(8.16 × 0.90) + (7.70 × 0.10) = 8.114
The game’s published rating would be **8.11 out of 10**. We retain the complete figures throughout the calculation and only round the final result to two decimal places.
How Our AI-Assisted Slot Testing Works
For compatible game demos, an AI agent uses Playwright browser automation to open the slot, interact with its controls and complete a standardized testing sequence. Playwright provides the browser-control layer, while the AI agent interprets the information collected during the session and applies our scoring criteria.
The automated test records:
- Successful loading and startup behaviour
- Control behaviour and interface consistency
- Feature events encountered during the session
- Error messages or technical problems
- Available game information used in the five-category assessment
The AI score contributes 10% of the final result. It acts as an additional consistency check and cannot overrule the combined judgment of our human team.
Some online slots use interfaces that cannot be controlled reliably through our automated routine. We never invent or estimate an AI score when a test cannot be completed properly. We either rerun the automated test successfully or assign the game to a fourth human tester and calculate a human-only rating.
Our AI-assisted session does not prove that a slot is fair, verify its random number generator or calculate its true RTP. Those claims cannot be supported by a short automated test.
How We Select the Weekly Top 10
To qualify for the final ranking, a game must:
- Fall within the release period covered by that week’s report
- Be available through a working demo or approved testing environment
- Complete our minimum human-testing requirement
- Provide enough game information for a fair assessment
- Pass our final check for scoring, data-entry and technical errors
We order the eligible games by their final calculated scores. If two games have the same published score, the unrounded result determines their positions.
If the unrounded scores are also identical, we use the higher human score as the first tie-breaker. The next tie-breakers are the Gameplay and Mechanics score, followed by the Bonus Features and Feature Design score. If the games still cannot be separated fairly, the editorial team completes an additional review.
Why Short-Term Wins Do Not Determine Our Ratings
Six hundred human-controlled spins and a 500-spin automated session provide enough time to inspect a game’s operation, features and overall experience. They do not provide a statistically meaningful test of its long-term RTP or randomness.
For that reason, a tester’s wins and losses do not increase or reduce the rating. We use the RTP displayed within the tested version of the game or supplied through official game documentation. Players should also be aware that operators may offer different RTP configurations of the same slot.
Editorial Independence and Corrections
Commercial relationships do not increase a game’s score or change its calculated ranking. Any sponsored placement must be clearly labelled and kept separate from our independently calculated Top 10.
Where the information is available, we record the game provider, testing date, tested version and displayed RTP. If a studio changes a game, an operator supplies a different RTP configuration or we discover a material error, we may retest the slot and correct the relevant weekly report.
Our Ratings Are Editorial Opinions
Our weekly rankings represent the combined opinions of experienced slot players applying the same structured testing process. They are not official ratings issued by a regulator, testing laboratory, game provider or casino.
Different players may prefer different themes, pacing, volatility levels and bonus features. It is entirely reasonable for another player to give one of our selected games a different score.
A high rating does not mean that a slot is more likely to produce a profit. Online slot outcomes are random and highly variable, and no ranking can predict the result of a playing session. RTP settings and available features may also differ between casinos and jurisdictions, so players should always check the information displayed in the version they are considering.
Online slots are intended for adults in jurisdictions where online gambling is legal. If you choose to play, set firm time and spending limits and never treat gambling as a way to make money.
Our Most Recent New Casino Game Reports
On the Relax Gaming news pages, the message is clear. The software provider aims to further strengthen its[...]
Playtech kicked off 2026 with three major deals. We will cover the software provider's custom slots for Novibet,[...]
BGaming has launched Surrender Multihand Blackjack, bringing back the surrender option to online blackjack tables with a competitive[...]
We reveal 3 New Slots from EGT DigitalPlay that are coming to online casinos near you in[...]
Stakelogic releases 5 mini games, offering quick rounds, high RTPs and decision-driven gameplay. Demo links included. [...]
FAQs
Does a high slot rating mean the game is more likely to pay out?
No. A high rating reflects gameplay, features and presentation, not profitability — slot outcomes are random. The rankings can’t predict the result of any playing session.
Does this testing verify that a slot is fair or check its RNG?
No. The human spins and AI session inspect how a game operates, not its fairness, randomness or true RTP. Those claims can’t be supported by short-term testing.
Are these slot rankings official industry ratings?
No. They’re editorial opinions from the testing team, not ratings issued by a regulator, testing laboratory, game provider or casino. Other players may reasonably score the same games differently.
Does the weekly Top 10 mean these are the best slots available?
No. It’s selected only from that week’s new releases that could be tested — not every slot worldwide, and not the ten best ever released.
Is the RTP shown here the same at every casino?
Not necessarily. Operators may offer different RTP configurations of the same slot, so always check the RTP displayed in the specific version you’re playing.
How many people actually test each slot before it’s ranked?
At least three independent human testers per slot, each completing 200+ base-game spins — a minimum of 600 human spins — plus one 500-spin AI session on compatible games.
Does AI decide the slot rankings?
No. The AI-assisted score contributes only 10% and can’t overrule the human team. If automation fails, the game is reassigned to a fourth human tester and scored on humans alone.
Do sponsorships or paid placements affect a game’s ranking?
No. Commercial relationships don’t change a game’s score or position. Any sponsored placement must be clearly labelled and kept separate from the independently calculated Top 10.
Does testing a bonus with a feature buy show how often it triggers normally?
No. Purchased or directly activated features are logged separately and used only to examine how a bonus works — not how frequently it occurs or how profitable the game is.