Review Methodology

Each Runewager page here that carries a score builds it from six criteria. Each is scored out of 10 and then combined into the final score with the weights shown.

A score below 5 on any single criterion caps the overall score at 7, even if the other criteria are strong — because a cashier that fails or terms that are unclear outweigh an impressive lobby.

There are two things our scoring deliberately does not capture. First, it does not weigh the odds a specific player will win — that depends on the games played, the strategy applied and the mathematics of each game, and no operator review can tell an individual reader what their expected outcome will be. Second, it does not score the ethics of the gambling industry as a whole — our scores are relative, operator against operator, not operator against a world without operators.

If you want a score that captures those two things, read the Responsible Gaming page instead of this one.

Each criterion is scored on a 0-10 scale. The six scores are combined using the weights in the table above into a weighted average, rounded to one decimal place. The overall score is also presented as a star rating out of 5 — the weighted average is simply divided by 2 and rounded to the nearest 0.1.

Scores do not have inflation. A 7 is a solid operator with no major issues; an 8 is above average on at least four criteria; a 9 is reserved for operators that are the market leader on at least two criteria at the time of writing. We have not awarded a 10 and do not plan to — a perfect score would suggest no further work for the operator to do, and that is never true.

The three fastest ways to lower a score on this site are: a cashier failure that cannot be resolved by support, a bonus clause that we think is unenforceable in the United States, and a change to the terms that materially tightens them without a notice period for existing players.

Slower, cumulative drags on a score are: a licence line that has moved to a less credible regulator, a support response time that has lengthened by more than a factor of two since the last cycle, and a lobby that has lost a studio without replacing it. Each of these is noted on the page it appears on, and the operator is given space in the Updates section to show that the change has been reversed before the next cycle.

Testing is live, not paper. The reviewer registers an account at Runewager, deposits the minimum amount with a card and a second rail, runs the welcome bonus end to end, places withdrawals through at least two rails and times each stage. Support is tested at three different times of day on each test cycle, from the three time zones that cover the US daytime hours. The app is installed on an Android and an iOS device and run alongside the desktop site for feature parity.

Each test cycle lasts two to three weeks. The resulting notes are longer than the published page by a factor of three — most of the detail stays in the notes, and only the signal from them is lifted into the review. If a figure cannot be reproduced on a second account, it does not go into the review.