How We Price Promos: The Formula, the Sources, and the Limits

Every figure on this site comes out of the same short calculation, applied to terms taken from named sources. Both are printed here so you can check any number yourself rather than trust it.

The formula

realisable    = min(bonus granted, maximum cash-out)
turnover      = wager base x wagering multiple
effective     = turnover / game contribution
clearing cost = effective x house edge
Real Value    = realisable - clearing cost
Terms Tax     = 1 - (Real Value / advertised headline)

Read it in words: work out what the terms will actually let you take away, subtract what it costs in expected house edge to satisfy the wagering requirement, and compare the result with the number on the advertisement. The wager base is whichever the operator specifies: the bonus alone, or the deposit plus the bonus. That single choice moves the answer more than any other term.

It is a closed form on purpose. You can check any figure on this site with a calculator, which would not be true of a number produced by a simulation. Real Value is a price, not a prediction: it says what a promotion is worth on its terms, not what will happen to you.

Why the chance of clearing is a separate number

Whether a playthrough can realistically be finished is a different question from what the offer is worth, and folding the two together would be wrong twice over. The clearing cost above already assumes the full turnover was played; a player who runs out of money never pays it. And expected value and the spread of outcomes are simply different properties. A staking pattern reshapes the second and cannot touch the first.

So the chance of finishing is published beside the value, never inside it, with the seed and the number of simulated attempts printed next to it so the figure can be reproduced exactly. No arrangement of these offers produces an edge over the operator, and nothing here should be read as suggesting one does.

Sweepstakes offers, priced differently

Sweepstakes sites have no deposit match and no house-edge grind against your own money. They sell coin packages. Gold Coins carry no redemption value; Sweeps Coins redeem at roughly a dollar each after a playthrough, above a minimum balance the operator sets. So Real Value there is the redeemable leg minus what was paid, with the Gold Coin headline valued at nothing, because that is what it is worth.

Most welcome grants land below the operator's own redemption minimum, and the obvious move is to price them at nothing on the grounds that they cannot be cashed out as they stand. The register does not do that, for the same reason it refuses to fold clearing odds into a deposit match: a balance played toward a threshold keeps its expected worth, less the edge charged on the way. Zeroing the row would charge that edge a second time and call the result a measurement.

What the threshold changes is the chance of ever converting the grant, and that is published beside the value as its own figure. On a game with no house edge, climbing from a balance to a threshold succeeds with a probability equal to the ratio of the two. Two coins against a fifty-coin minimum is four percent, and every real game has an edge, so the ratio is a ceiling. Two numbers, two questions: what it is worth, and how likely you are to see it.

Where the terms come from

1. The operator's own terms, rules or promotions page. Always wins.
2. The operator's own marketing or explainer pages. What is usually reachable in practice.
3. Two or more independent trackers, plus a dated check of the signup flow.

A tracker never overrides the operator. That rule is not ceremonial: trackers currently disagree with BetMGM about whether its wagering requirement runs on the bonus alone or on the deposit as well, a difference of three times in required turnover. Following the more popular source would have understated the cost of that promotion badly.

Sources are recorded per field, not per entry, because a single row routinely takes its wagering multiple from an operator page and its game contribution from a tracker. Each operator page lists which field came from where.

The three precision states

Operators frequently do not publish a maximum cash-out. Because that term is a cap, a figure calculated without it is an honest upper bound rather than a guess, so we publish the ceiling and label it instead of leaving the row empty:

• = $X — every field confirmed.
• at most $X — no cash-out cap published; the figure is a ceiling.
• not calculable — the wagering multiple or the bonus amount itself is missing. No number is shown.

We would rather print a bound than a guess, and rather print nothing than a bound we cannot justify. A figure invented to fill a column would defeat the point of the site.

What happens when an entry goes stale

Bonus terms change without notice and this site is a set of static pages, so an entry left alone becomes wrong. Every entry carries the date it was last checked, and its age is recalculated against the build date each time the site is rebuilt:

• Under 30 days — shown normally.
• 30 to 180 days — the date is marked, more prominently past 90 days.
• Over 180 days — the figure is withdrawn and the row drops out of any ranking.

That last rule costs us traffic and revenue whenever a refresh slips, which is the point. The alternative is not that the pages keep earning; it is that they keep earning while quietly publishing figures we no longer stand behind. The register was last refreshed on Aug 18, 2026.

How we make money, and what it does not buy

Some links on this site earn a commission when a reader opens an account. Those links are marked wherever they appear. Commission is not an input to the formula above, so it cannot move a row: tables are sorted by Real Value, and you can check the ordering against the marks yourself.

We also cover operators we earn nothing from, and we publish unfavourable findings about operators we do earn from, including the negative figures and the public-record material on the verification pages. If those two things were not true, the sorting rule would be a slogan.

What we do not do

We do not rate casinos. There is no score, no star, no verdict on game libraries or apps, because none of that can be evidenced from a document and a number attached to a taste is just an opinion wearing a uniform. We verify and we price. Where the evidence runs out, the page says so.

Questions about the method