Why do we refuse to value 18 casino reward programmes?
We refuse to compute a cashback value for 18 of the 28 crypto casino programmes we examined, and we publish every refusal with its reason (latest programme read 2026-08-31). The three commonest reasons: the programme is rakeback mislabelled as cashback, the operator publishes no rate or base to compute from, or the reward exists only as discretionary VIP credit. A refusal is not an exclusion: the casino stays fully tracked, ranked and scored, and the ledger entry costs it exactly one point on the Record axis of RAW Score v0. Publishing a computable rate and base removes the entry at the next read. A number we cannot derive from published terms is a number we will not print.
What a review site refuses to publish tells you more than what it publishes. This is our refusals ledger.
The evidence
| Casino | Why we refuse |
|---|---|
| BitStarz | Its 10% weekly net-loss cashback belongs to the invite-only Starz Club VIP. An ordinary account does not receive it, so it is not a rate a typical player can plan around. |
| Metaspins | Not a cashback programme, and the coin flip is the smaller half of why. Metaspins publishes a fixed 3% to 15% RAKEBACK ladder on its Level Up page, which is a share of the theoretical house edge rather than of your losses, so it fails the denominator test before any mechanic is considered. The "up to 120%" headline is then that top 15% put through an optional x8 coin flip, so it is the outcome of a gamble rather than a rate. Reason widened 2026-08-14: attacking only the coin flip left us exposed to Metaspins simply dropping it, after which the listing would still not qualify. |
| Gamdom | Not a cashback programme: Gamdom pays a share of its margin on every bet, win or lose, which is rakeback. Researched 2026-08-03 and the numbers confirm the headline is not reachable: instant rakeback is 6% of margin at Level 1 and 10% at Level 50, the Royalty Up tiers run 11% to 15%, and the advertised "up to 60%" is invite-only. |
| BC.Game | Not a cashback programme: BC.Game pays rakeback as wagered x 1% x an unlock rate of 20% to 24% by VIP tier. Because that 1% house edge is notional rather than the real game edge, it works out at a flat 0.20% of turnover at entry, and all 69 VIP levels only lift it to 0.22%. Re-checked 2026-08-14, and this is the entry most likely to change: a weekly cashback of up to 25% on net losses paid on Mondays now circulates, which would qualify on basis, but it appears on third-party pages only, with no entry rate, no ladder, and availability said to depend on country and tier. Both bc.game and its help subdomain returned HTTP 403 to us, so we could not read the promotions page. Until we can, there is no rate here to compute with. This also supersedes the older "1% to 5% loyalty cashback" figure we used to cite, which we can no longer confirm anywhere. |
| Roobet | Not a standing cashback programme. Re-read 2026-08-14: Roobet’s own help centre now states that all four of its reward types, the instant rakeback credited every 30 minutes and the daily, weekly and monthly bonuses, are computed as a percentage of the amount you WAGER. That is rakeback on turnover, not a share of your losses, so none of it belongs on this axis. This corrects our earlier wording, which said Roobet never states what its rewards are calculated from; that was accurate when written and is now out of date. Roobet does run a genuine 20% net-loss cashback, but only as a welcome offer covering your first seven days, which is a one-off rather than a rate a player can plan around, and we have seen it stated on affiliate pages rather than on Roobet’s own. Separately, the dead news subdomain that once described a reward computed from deposits minus withdrawals still does not resolve, so that basis remains unverifiable. |
| Mybookie | MyBookie does rebate losses, but it publishes no percentage and no cap, so there is nothing to calculate. Read on 2026-08-13, its loyalty page describes a WEEKLY rebate on both sportsbook and casino losses over a Monday 00:00 to Sunday 23:59 ET cycle, gated behind Elite Level 3, with sportsbook losses returned as Free Play at 1x and casino losses as cash at 1x, and the rate itself replaced by a note saying to stay tuned. This corrects us as well as MyBookie: until 2026-08-13 we described it as the one monthly operator in this dataset, on terms it published with a dollar cap. The monthly cycle and the cap are both off the page now, and we do not restate dead terms here in case they get quoted back as live ones. |
| Rainbet | We hold firsthand-tested payout data for Rainbet, but not a cashback rate we can compute with. Sharpened 2026-08-14 now that we know more than "contradictory or undisclosed": Rainbet’s headline reward is rakeback, computed as your wager times the house edge times a percentage, which is the wrong denominator for this tool. It does separately run a weekly lossback, which IS loss-based, but Rainbet publishes no ladder for it, the upper tiers are individually negotiated per player, and the only entry figure we have seen, around 4%, comes from third parties rather than Rainbet. Both rainbet.com and its help subdomain returned HTTP 403 to us on 2026-08-14, so nothing here could be confirmed in the operator’s own words. |
| Shuffle | Not a cashback programme: Shuffle pays a share of the house edge on every bet, win or lose, which is rakeback and does not belong on a net-loss axis. Corrected 2026-08-14: Shuffle now DOES publish the percentage, stating on its own help pages that you get "5% of the house edge back as cash". Until today this entry said it declined to publish a percentage at all, which was true when written and is not true now. The exclusion is unchanged, because the basis is the house edge rather than your losses, but the reason it is excluded is the denominator, not secrecy. |
| Chips.gg | Not a cashback programme: its own help centre states that rakeback "is calculated based on the house edge of the game and your bet amount", which is turnover, not losses. Chips.gg does separately pay a "Lossback, which includes a portion of your net losses from the past 28 days" inside its VIP weekly bonus, and that would qualify on basis, but the same article says "The exact calculation formula is not publicly shared", so there is no percentage to compute with. |
| Trustdice | Not a cashback programme despite the name. Its published formula is your bet amount times a per-game percentage times a cashback percentage, and the per-game term tracks the game margin: 1.5% on dice, 3% on slots, 0.5% on live casino. That is rakeback wearing the word cashback, the same shape as the BC.Game exclusion. The "up to 20%" headline is 20% of that margin term and belongs to the invite-level Satoshi tier, so on slots it works out at roughly 0.6% of turnover rather than 20% of anything you lost. Every trustdice.win page returned HTTP 403 on 2026-08-14 and trustdice.io now redirects to a parked host, so the formula above reaches us through search extracts rather than a page we retrieved. |
| Fairspin | Fairspin does run a loss-based weekly cashback, separate from its TFS token rewards, so unlike most entries here it fails on evidence rather than on basis. Every fairspin.io path we tried returned empty content on 2026-08-14, leaving us zero operator wording, and the third-party figures contradict each other flatly: 15% wager-free, 50% at 60x, 50% at 45x, and a separate 5% to 20% claim all circulate. Only the 5% wager-free entry row is consistent across sources, and picking a top rate out of that spread would be exactly the interpolation this dataset refuses. Its TFS programme is a token emission paid per bet (0.2% on slots, 0.075% on live roulette, 0.03% on blackjack), which is rakeback on turnover and out of scope regardless. |
| Vave | There is no programme to list. Asked directly about monthly, weekly and daily bonuses, Vave’s own help centre answers verbatim that "Vave does not currently offer those". The 3% to 7% weekly net-loss ladder attached to Vave exists on aggregator pages only. What Vave does operate is a VIP points mechanic it states itself, one point per 20 USDT wagered with 100 points converting to 1 USDT, which works out at roughly 0.05% of turnover and is rakeback, not cashback. |
| Jackbit | Not a cashback programme, and the brand has moved: jackbit.com now redirects to jack.com, observed 2026-08-14. Its "30% rakeback" headline is a POINTS contribution on wagers rather than money. At the Rookie level that contribution is 20% and 100 points convert to 1 USD, so a 100 USD slots wager returns about 0.20 USD, which is 0.2% of turnover. Same arithmetic as the BC.Game exclusion, and none of it is computed from your losses. |
| Betplay | Probably a real net-loss cashback, but not one we can stand behind yet. Betplay.io returned HTTP 403 to every request on 2026-08-14, so we hold not one word from the operator: the 10% weekly figure, the Bronze I entry gate and the Thursday 13:00 UTC credit all come from third parties. Two of those then contradict each other on what counts toward the base, one saying sportsbook and live table games are excluded and the other saying all sportsbook, casino and esports play contributes. Betplay also runs a SEPARATE daily rakeback paid on wagers "regardless of wins or losses", which is precisely the pair of programmes that gets conflated into one wrong number, so publishing an unverified rate here would risk putting a rakeback figure in the cashback column. |
| Cryptoleo | There is no operator page left to quote. Cryptoleo.com publishes no A, AAAA or CNAME record, verified two ways on 2026-08-14 (a fetch that failed to resolve, plus a DNS lookup returning SOA only) even though the zone itself is live on Cloudflare nameservers. The reward also fails the label test independently: two affiliates describe a 10% to 20% table gated on the previous day’s DEPOSIT SIZE rather than on losses, AskGamblers files it as "Up to 25% Cashback Bonus (Rakeback)", and that same page carries a player report that the feature was removed. |
| Coinsgame | Unreadable, and from what circulates it is not a clean loss measure anyway. Every coins.game path we tried returned HTTP 403 on 2026-08-14, which we read as a bot block rather than a dead site, and we would rather say that than quote an aggregator in the operator’s place. The third-party ceilings are mutually incompatible (a 10% maximum across 20 levels, a flat "up to 20%", and a promo-code 15%), no source gives a per-level rate, and the programme is described as combining "Profit and Loss" with lossback while activating it also triggers a rake boost. A loss measure bundled with a rake measure is not a usable denominator. The widely quoted 10% daily figure is a welcome offer limited to the first seven days, not an ongoing rate. |
| Spin Samurai | Spin Samurai publishes rates (15%, 20% and 30% at Hatamoto, Daimyo and Shogun) and wagering (x3, x3 and x10) but never says what the percentage is taken FROM, so there is nothing to compute. The one threshold it does state reads verbatim "Minimum deposit to get the cashback reward shall be 200 EUR/USD ... per day". Affiliate pages restate that same sentence with the word "loss" swapped in for "deposit", which is their edit, not the operator’s. A 30% rate would also sit far above the roughly 20% ceiling every genuine net-loss programme we track tops out at, which is itself a signal the base is not net loss. Levels 1 to 4 of its ladder pay no cashback at all, so an ordinary account gets nothing either way. Disclosure: Spin Samurai is one of our affiliate partners, and it is excluded here anyway. |
| Wild Fortune | There is no cashback programme. The word does not appear in Wild Fortune’s own bonus terms; its bonuses page lists a welcome offer plus six reload and free-spin promotions with no cashback among them; and its loyalty page lists all seven levels, Barefoot Wanderer through Gold Reaper, paying free spins only. Three operator pages, three misses. Treat the figures circulating for it with suspicion: affiliate claims of 7% to 15% and of 5/7/15% disagree with each other, and the "up to 20%" attached to Wild Fortune is almost certainly bleed from Wild.io, a different operator that IS in this dataset. Its Comppoints exchange is rakeback-shaped rather than loss-based. Disclosure: Wild Fortune is one of our affiliate partners. |
How we know
Each programme was read the same way as the computable ten. The refusal reasons quote or paraphrase the operator's own published wording where it exists, and say plainly when nothing verifiable is published.