What the record shows, market by market

The United States block takes catalogue depth: how much is actually in each lobby, and how much of it repeats Roobet's. The coin questions are taken further down, by Canada, Sweden and Russia, so that no two blocks measure the same thing.

Roobet publishes sixty-four studios. Eight of the twenty-two alternatives whose lists we have read publish more, and the top three publish very substantially more. Bitcasino.io names a hundred and twenty studios, Fairspin a hundred and fourteen, Sportsbet.io a hundred and ten. At the other end, CryptoLeo names thirteen and BetPlays and DuckDice name nineteen each.

Depth is not the same as difference, though. Bitcasino.io's hundred and twenty include forty-nine of Roobet's own sixty-four, and Rainbet's much shorter list of thirty-three includes twenty-six of them. A reader who wants more of the same should look at the third column below; a reader who wants something else should look at the fourth.

For Canada the measure taken is stablecoin support: which cashiers take which stablecoin.

Roobet lists two, USDT and USDC. Across the twenty-two alternative cashiers we have read, USDT appears on eighteen and USDC on eleven. They are not interchangeable at a cashier even though they are treated as interchangeable in conversation, and eleven of the twenty-two would take one and not the other.

Cloudbet is the outlier with six stablecoin entries, USDT and USDC among them alongside USDP, DAI, USDE and SUSDE, plus the gold-backed PAXG. Rocketpot lists four, and is the only cashier in the set that separates tether by network, printing USDTE and USDTT as two entries. Three of the twenty-two read cashiers list no stablecoin at all: Bitdreams, KatsuBet and Playbet.io.

For Germany the measure is the published verification figure, one of the few questions in this comparison where an actual number exists for some brands and not others.

Five of the forty-one publish one. Rocketpot states 2,500 US dollars at clause 11.4. Bitcasino.io, Sportsbet.io and Livecasino.io each state 2,500 euros at clause 6.6. Empire.io states 2,000 USDT at clause 5.4. Roobet publishes none: clause 4.9 reserves documents at any time and clause 10.5 refuses withdrawals until identity is verified, with no amount attached to either.

All five clauses are triggers for additional checks rather than guarantees about anything below them, and all five reserve the right to check smaller withdrawals as well. The verification page sets out the clause numbers for the other thirty-six.

For Spain the slice taken is a documentary one: given a brand's licence number, can you get from that number to a register entry, or only to the brand's own page saying it has one?

Thirty-nine of the forty-one name a licensing authority and carry a licence number. Where those numbers point differs sharply. Sixteen brands' records link to the Anjouan Gaming Board's public register. Seventeen link to a certificate identifier on the Curaçao Gaming Authority's certificate service. Six link only to a page on the operator's own site, which is a claim rather than a confirmation. Two, Nitrobetting and BetPlays, carry no licence authority, no number and no register link at all.

One caveat applies to the Curaçao column and it is important. On 2026-08-28 the Curaçao certificate register returned no result for any domain queried, including brands that certainly hold Curaçao licences, so a domain search there settles nothing in either direction. Certificate identifiers still resolve; domain searches do not.

For Brazil the measure is the published withdrawal cap: the way out rather than the way in, and the number that decides how a large balance is released.

Eight of the forty-one state one, and the spread is more than two hundredfold. Bitcasino.io sets a million USDT a week at clause 6.10. Wild.io sets a hundred thousand US dollars a week at 9.6. CryptoLeo sets fifty thousand USDT a month at 7.9, Bitdreams twenty thousand euros a month, Wild Fortune fifteen thousand euros a month at 11.8, Mirax ten thousand euros a week, BetPlays five thousand euros a week at 6.11 and Rocketpot five thousand dollars a month at 11.5.

Roobet states no cap in anything we have read. Three further brands, Metaspins, Rakebit and Nitrobetting, have a withdrawal clause recorded with no figure in it. The remaining thirty publish nothing on the point, which is not the same as having no limit.

For the United Kingdom the slice taken is the least glamorous one in the set and the hardest to fake: how long the operator says it has been going.

Ten of the forty-one publish a founding year. Cloudbet and Nitrobetting both give 2013, FortuneJack 2014, Bitsler 2015, and Rocketpot, Bspin, CoinPoker and Sportbet.one all give 2018. Wild.io gives 2022 and Rakebit gives 2024. That is the whole of it: thirty-one brands publish nothing we have read on the question.

Roobet publishes no founding year either. It is one of the more conspicuous absences in its own record, given how much of the brand's public identity rests on having been early, and it is why no page on this site states one.

For France the measure is live dealer supply, which concentrates far harder than slot supply: who actually runs the tables.

Evolution appears on twenty of the twenty-two alternative studio lists we have read, and on Roobet's. Ezugi, which Roobet does not list, appears on eleven. Below those two the field fragments immediately: Live Solutions, Live88 and Bombay Live on five lists each, CreedRoomz and Vivo Gaming on four.

The library separately tags thirty-six of the forty-one as carrying live dealer games. That tag is a description of the product rather than a supplier list, so it tells you the tables exist and not who is running them. Where both are known, the studio list is the stronger evidence.

For Sweden the measure is cashier breadth: how many tickers a cashier prints at all, and how far the set stretches.

It stretches a very long way. Cloudbet prints forty tickers and KatsuBet prints one. Between them sit Sportbet.one on nineteen, Rakebit on fifteen, Wolf.bet and Rocketpot on thirteen, and a long middle of cashiers listing between five and ten. Roobet prints eleven, which puts it almost exactly in the middle of the twenty-two lists we have read.

Breadth and overlap are different things. LTC Casino prints ten tickers and every one of them is on Roobet's list; Cloudbet prints forty and misses two of Roobet's eleven, because it lists neither SUI nor TRX.

For Croatia the honest slice is the one nobody else prints: which brands we know least about, and why that should count against them rather than for them.

Nineteen of the forty-one publish no cashier list we have read and nineteen publish no studio list. Five publish no verification wording at all: BC.Game, Flush, BetFury, TrustDice and Betcoin.ag. Two publish no licence record of any kind: Nitrobetting and BetPlays.

Those absences are the reason the scoring routine on this site divides by a fixed maximum rather than by the number of fields we happen to hold. A brand with two published facts and both of them good is not a better-documented brand than one with nine, and the methodology page sets out exactly how that is enforced in the code.

For Portugal the question taken is not which brand is best documented but which is the same kind of product, so the slice here is the library's own product tags.

Thirty-six of the forty-one are tagged as carrying live dealer games and thirty-three as carrying a sportsbook. Twenty-seven carry a VIP programme tag, nineteen a provably fair tag, eighteen an original games tag and twelve a crypto-only tag. Eleven carry an instant-withdrawal tag.

These are the library's descriptions of the product, not quotations from anybody's terms, and they are the weakest evidence on this page by some distance. They are useful for narrowing a list of forty-one to a list of ten and useless for settling anything after that. Roobet carries no tags in the library at all.

For Russia the measure is the tail of the coin lists rather than the head: the chains beyond bitcoin and ether.

Tron appears on twelve of the twenty-two cashiers we have read, and Roobet lists it too. TON appears on four. Beyond that the tail thins fast: monero on DuckDice and Rakebit, dash on DuckDice and Cloudbet, zcash on Rakebit and Cloudbet, EOS on three cashiers.

Cloudbet's forty-ticker list is where nearly all the remaining variety sits, taking in HBAR, ALGO, XLM, AVAX, STETH, BSV and a set of memecoins. Three cashiers list a token of the operator's own instead: SHFL at Shuffle, TXT at TrustDice, and SBET alongside VAULTA at Sportbet.one. A house token widens a coin count without widening the set of ways money gets in.