A new casino is an unproven casino. We are not going to pretend otherwise.

What "new" is actually telling you

Every other page on this site lists operators that have passed a test. This one does not. It lists operators that have not yet taken one.

"New casino" is among the highest-intent searches in this entire vertical, and it is also the one with the least evidence behind it. The pages competing with this one will tell you a new operator is exciting, fresh, and offering a generous welcome package. What they will not tell you is the thing that actually matters: a new operator has no track record of paying people, because nobody has needed to be paid yet.

That is not an accusation. It is a statement about the information available, and the honest response to missing information is to say it is missing.

The cold-start problem, and what we do about it

Every rating system built on player reviews has a structural flaw at the start: with very few reviews, the score is noise. Three enthusiastic reviews on launch week can produce a perfect ten, and a perfect ten is exactly what a new operator's marketing budget is trying to buy.

We damp it deliberately. The JuryScore is a weighted Bayesian average, which in plain English means every operator's score is pulled toward a neutral middle until real testimony accumulates. The engine holds 5 phantom reviews at 5.5 out of 10 against every operator, and they only wash out as genuine testimony arrives.

Here is what that does in practice:

Three perfect scores do not produce a perfect score. They produce a 7.19, flagged as an early verdict, and they still lose to forty merely-good ones. Reviews are also weighted by verification — a juror who has proved they held an account counts for 1.6 times an anonymous one — and decayed on an 18-month half-life, so praise ages out.

The problem this used to have, and what is left of it

Now the part a competitor would leave out.

The JuryScore is damped for new operators. The TrustAudit Index could not be, because it measures documented findings and a new operator has almost no documents. Work through what that meant. The score starts at 10.0 and subtracts; an operator with a decent licence, terms nobody had read yet, no cap on record and no 24-month complaint history at all — because it had not existed for 24 months — had almost nothing to deduct. It computed a high number for the worst possible reason: we had not had time to find anything.

That was a real defect and it is largely fixed. The audit now runs on an evidence file with six channels — the licence checked against the regulator's own register, the terms read for unfair clauses, the withdrawal cap, the payout time, the KYC turnaround, and the wider reputational record searched across published complaint databases. Below three examined channels no number is published at all; at three or more examined channels a score can be published, with research gaps or unresolved findings marked provisional. An empty file no longer produces a ten. It produces "Not yet rated", which is the honest output.

What is left is smaller but real, and worth saying plainly. A young operator has genuinely had fewer opportunities to be caught, so even a properly examined file has less in it. The reputational channel narrows that too — a complaint record exists from the first month, where a 24-month black-point history does not — but it cannot close it entirely. Time on the market is evidence, and a new brand has not served any.

A smaller version of the same distortion runs through the ordering of the register. It used to be severe: an operator with no published score contributed its absent audit as a top mark and sorted straight to the front of the directory, which is the opposite of what a missing audit should do. An unrated operator now contributes a neutral 5.0 to the blend instead, so a thin file is read as no signal rather than a good one. What remains is milder and structural. Data freshness carries 5% and a new operator's data is fresh by definition; open cases past deadline are penalised and a new operator has none. We would rather tell you that on the page than let you discover it.

The one page on this site where the difference is visible is the clean sheet list, which requires an operator to have been publicly accused and to have answered every time. No operator on this page can qualify for that one. That is not a coincidence; it is the whole point.

The most important check: is it actually new?

Many "new" casinos are not new companies. They are a new brand — a new skin, a new domain, a new logo — over an existing operator, an existing licence and an existing set of terms. Sometimes this is entirely ordinary corporate practice. Sometimes it is an operator with a bad record starting again under a name nobody recognises.

The brand is the one thing that is genuinely new. The company behind it is where the record lives.

We check this, and the engine acts on it: where we can establish that an operator belongs to a group, it inherits a share of that group's complaint record, weighted by the strength of the link, and it can move its TrustAudit Index by up to 1.5 points in either direction. Groups share reputation both ways — a new brand from a group that pays people gets the benefit of that too.

You can check it yourself in about five minutes:

1. 01Find the licence number in the footer or the terms, then look it up on the regulator's own public register — navigate there yourself rather than following the operator's link, which is the one link the operator controls.
2. 02Read the company name on the licence, not the brand. Then search that company name. It will usually be attached to other brands.
3. 03Check the trading names listed on that licence. A real licence number attached to a different brand is the oldest trick on the internet.
4. 04Compare the terms and conditions to those of the sister brands. Predatory clauses travel between brands in a group, because the legal team does not get rewritten when the logo does.

What to do before you deposit at a new operator

1. 01Find the withdrawal limit before you find the welcome bonus. A monthly cap is the clause that decides whether a win is a win. The arithmetic is here, and it is brutal.
2. 02Read the maximum-win cap in the bonus terms. New operators lean hard on welcome offers, and those offers carry the tightest caps. How to price a bonus.
3. 03Complete identity verification on day one, while your balance is zero and nobody has an incentive to be difficult.
4. 04Deposit an amount you have fully accepted losing — which is the only amount you should ever deposit anywhere, and doubly so with an operator whose payment behaviour is entirely undemonstrated.

Why we publish this page at all

We could omit it. Some sites would argue the responsible thing is not to list unproven operators.

We disagree, for a simple reason: people search for new casinos in very large numbers, and they will find a list either way. Every other list they find will be written by someone paid to be enthusiastic about it. If the choice is between a page that says "new and exciting" and a page that says "new, therefore unproven, and here is the arithmetic you should run before you deposit", the second one is safer — and it is the only one either of us would want a friend to read.

None of which changes the game itself. A new operator's games carry the same built-in house edge as an old one's, which means that over any meaningful number of bets the expected outcome of play is a loss. Novelty is not an edge, a welcome bonus is not an edge, and nothing on this page makes gambling a way to make money. What the arithmetic above can do is stop an unproven operator turning an ordinary loss into an unrecoverable one.

The full method, including what it gets wrong, is on the methodology page. Operators we have delisted are on the blacklist, with reasoning. And if what brought you here was a bad experience rather than curiosity, the Tribunal is where you file it.