What the licensed studios actually run

The British live-dealer baseline is set by a small number of large studios whose tables appear across most Commission-licensed lobbies, and the category structure they run is remarkably uniform, which makes it a useful reference point. Roulette leads the roster: European single-zero wheels dominate, streamed from dedicated studio wheels with real croupiers, and the house margin sits at 2.7 percent on standard outside and inside bets. Blackjack follows in popularity, dealt to British conventions with the dealer standing on soft seventeen at most tables and paying three to two on a natural, which puts the basic-strategy margin at roughly half a percent, the lowest on the floor. Baccarat and its speed variants fill the high-limit corner, and game-show formats, the wheel-and-bonus hybrids that grew out of the last few years, occupy the entertainment end with margins that run materially higher than the classic tables.

The reason the studio structure matters is that it sets expectations a reader can carry anywhere. A licensed studio table publishes its rules on screen: deck count, dealer actions on soft totals, side-bet payouts and the theoretical return of each wager type. Nothing about the deal is opaque, the shoe is shuffled on camera or drawn from a continuous shuffler with the audit documented, and every result is logged to the account history. That transparency is a product of licensing rather than of the studio’s generosity, because the Commission requires operators to make game rules and return information available, and the studios build to that requirement across their whole estate.

The offshore lobbies imitate this structure closely, often carrying the same table categories with the same game names, and on the surface the difference can be hard to spot. Look closer and the differences are in the frame rather than the picture: which regulator audited the wheel, what complaint route exists if a hand is misdealt, and whether the published return figures were verified by anyone other than the operator. The classic tables themselves are hard to rig in any sustained way, since the physics are on camera, but the surrounding terms, cashout handling and dispute resolution are where the licence class actually shows, and those are covered later on this page.

One more structural point is worth settling early: live tables are shared, not private. Dozens or hundreds of accounts bet on the same wheel or shoe simultaneously, results are identical for everyone at the table, and the pace is set by the dealer and the studio rather than by any player. That makes live play slower per round than slots but larger per round than almost anything else in the lobby, and the two facts together shape the whole cost model this page builds: fewer rounds, bigger stakes, thinner margins, and a very different relationship with bonus terms than the reel games the terms were written around.

Table limits and what they do to a session

Table limits are the practical gate on who plays what, and they vary more between studios and licence classes than almost any other visible feature. At the licensed end of the market the category-typical floor for live roulette sits around 50p on straight-up and outside positions, blackjack tables commonly open at £5 a hand, and baccarat rooms tend to start at £10, with high-limit variants of each running five to twenty times those figures. The minimum is not a formality: it sets the unit of risk, the speed at which a bankroll turns over, and the size of swing a player experiences per hour at any given edge.

The effect compounds with pace. A roulette table completes roughly forty to sixty rounds an hour once betting windows and spins are counted, so a single £1 outside bet player turns over £50 to £60 in that hour against a 2.7 percent margin, an expected cost of about £1.50. The same player at a £10 table turns over ten times as much and pays ten times the expected cost for identical entertainment time. Blackjack runs faster, sixty to eighty hands an hour at a full table and considerably more heads-up, so the per-hand minimum matters even more there. These are model figures, not measurements of anyone’s session, but they show how the minimum converts directly into hourly expected cost.

Offshore lobbies present a different limit profile. Stake floors at the low end are frequently lower than the licensed studios, with roulette positions from pennies and blackjack hands under a pound, because smaller deposits are the commercial focus of that market. The ceilings, however, are usually higher too, which suits the VIP-heavy segment, and between those poles sits a practical problem: at the offshore end the limits are set by the operator’s own terms rather than by any published standard, and they can change without the notice conventions a British licence would impose. A table minimum that doubles overnight is an annoyance; a maximum withdrawal that scales with the table limits is a structural term worth reading before any deposit.

For anyone planning sessions rather than single bets, the useful discipline is to fix a per-round stake as a fraction of the bankroll and then check that the table minimum permits it. A £200 bankroll at a £5 blackjack floor allows forty units, which is a workable margin for the swings of a low-margin card game; the same bankroll at a £10 baccarat floor allows twenty units, which tightens the odds of surviving a normal run of variance considerably. The calculator later on this page is preloaded with exactly that shape of session, and the assumptions behind its figures are stated beside it.

How bonus wagering treats live-table stakes

This is the section that saves the most money per word. Bonus playthrough terms are written around slots, and live-dealer play is almost always either excluded from the wagering count or weighted at a small fraction of the stake. The common pattern across the market is a contribution table: slots and instant games count at 100 percent, live roulette and blackjack contribute 10 percent or less, and some terms list live tables at zero outright. We take ten percent as the working assumption where a figure is needed on this page, because that is the middle of the range this part of the market usually publishes, but the actual figure for any specific offer lives in that operator’s own contribution table and must be read there.

The arithmetic is unforgiving once the weighting is applied. Consider the standard modelled case used across this site: a £25 deposit matched at 200 percent produces £50 of credit, and at forty times the playthrough asks £2,000 of qualifying stake. If the plan was to clear that on blackjack, a ten percent weighting means every £10 actually wagered counts as £1, so the real volume required becomes £20,000. At a £5 table playing seventy hands an hour, that is around fifty-seven hours of play, and at blackjack’s roughly half-percent house margin the expected cost of that volume is about £100, comfortably more than the £50 credit the match delivered. If the contribution is zero, no amount of table play moves the counter at all, and any bonus balance simply sits until its expiry date takes it.

The weighting rules exist for a plain commercial reason: low-margin games with optimal strategy are the cheapest way to grind a playthrough, so operators price them out of the bonus economy. The consequence for the reader is that a welcome package and a live-dealer evening are almost always incompatible products. The honest ways to combine them are narrow: play the table with real money only and skip the bonus, accept that the bonus must be cleared on slots first before the tables, or find the rare terms that weight live play generously and read their other clauses twice, because a generous contribution column usually travels with a maximum-bet clause and a shorter expiry window.

That maximum-bet clause deserves its own warning at the tables specifically. Most bonus terms cap the stake while wagering, commonly at £5 or thereabouts, and a single hand above the cap can void the bonus and the winnings built on it. Live tables make accidental breaches easier than slots do, because chip denominations are physical on screen, side bets are tempting, and a doubling decision in blackjack can push a hand past the cap without any deliberate act. A player clearing any wagering at all should treat live tables as off-limits until the playthrough is done, and the cost model below prices exactly that separation.

Streaming quality and stake floors offshore

The visible product at the offshore end of the live market ranges from indistinguishable-from-licensed to clearly second-tier, and the differences cluster in three places: stream engineering, table staffing and the small print around the stake. The better offshore operators license feeds from the same major studios that serve the licensed market, and on those tables the picture, latency and dealing standards are identical, because the studio is identical. The second tier runs tables from smaller regional studios with thinner staffing, longer waits between rounds, occasional rebalances of table limits, and streams that degrade under network load at peak hours in ways the major studios have engineered out.

Stake floors offshore tend to sit lower than the licensed minimums, as noted earlier, and that is a genuine difference with genuine appeal at small bankrolls: penny roulette positions and sub-pound blackjack hands exist at offshore lobbies that the licensed studios do not currently offer at scale. The trade is that the protections around those tables are thinner, and the practical risks concentrate at cashout rather than at the table itself. A misdealt hand at a major studio feed is the same event whichever lobby carries it; the difference is who adjudicates the complaint, and that difference is the subject of the next section but one.

Latency and device behaviour are worth a sentence of expectation-setting. Live streams run at several megabits per second for the top-tier feeds, and on a congested connection the video falls back to a lower resolution before it drops entirely, which is tolerable for roulette and much less so for blackjack where card values need to be read quickly. Mobile play at the licensed studios is generally well handled in both orientations; at smaller offshore studios the mobile presentation is sometimes a scaled desktop layout, and betting windows that assume a larger screen lead to mis-taps at precisely the moments a stake is live. None of this is a licensing matter, it is an engineering one, but it shapes the experience enough to check before committing a deposit.

The fair summary is that the offshore live product is not a worse simulation of the same game; on the major-studio feeds it is the same game. What it is, is the same game wrapped in different terms, different limits governance and a different dispute environment, and those wrappers are where the cost of the cheaper stake floor actually sits. A reader comparing a penny table offshore against a 50p table at a licensed lobby is comparing stake floors, and should also compare everything in the columns that follow.

The margin question: which tables cost least

Live-dealer games differ more in house margin than any other lobby category, and choosing by margin is the cheapest optimisation available. European roulette at 2.7 percent is the reference point; American double-zero wheels run at 5.26 percent and should be passed over wherever a single-zero alternative exists, which is nearly everywhere. Blackjack under British conventions with the dealer standing on soft seventeen and three-to-two naturals costs roughly 0.5 percent at basic strategy, the cheapest gambling in the building, while tables paying six-to-five on a natural nearly double that figure and should be treated as a different, worse game wearing the same name. Baccarat’s banker bet sits near 1.06 percent after commission, with the player bet slightly behind and the tie bet far behind at over fourteen percent, a wager that exists to be declined.

The game-show formats invert this hierarchy. Their headline appeal is a bonus round with a large multiplier, and the published return on the base game is correspondingly lower, with effective margins commonly in the three-to-five percent band or wider depending on the title. They are entertainment products rather than cheap table games, and pricing them as such is the honest framing: the same hour at a game show costs several times the expected cost of the same hour at a blackjack table, in exchange for a different shape of variance.

Side bets deserve the same treatment. Every classic table carries them, the payouts print attractively on the felt, and the margins behind them typically run from four percent into double digits. A player who declines every side bet and every insurance offer, and who plays single-zero roulette or properly-paid blackjack, is playing the cheapest games the live market sells. None of this requires any strategy study beyond the basic blackjack chart, which is legal to consult at the table in every jurisdiction this page covers, and the margin savings from doing so compound with every hour of play.

Stated as figures rather than judgments, at £5 a hand over seventy hands an hour: blackjack at half a percent expects to cost about £1.75 an hour, European roulette at the same stake expects about £9.45, and a game-show title at four percent expects about £14. The variance in any single hour will swamp all three numbers, sometimes violently, but the long-run ordering never changes, and a reader who knows the ordering can decide deliberately which product they are buying.

The licence gap at the dealer table

A Commission licence obliges an operator to participate in GAMSTOP, the national self-exclusion scheme, so that one registration blocks accounts across every licensed British site. An operator outside that licence is not required to participate, which means a GAMSTOP registration does not by itself stop accounts being opened at offshore lobbies, however familiar their live-dealer feeds look. Anyone who has self-excluded should hold that fact clearly before reading any further offer, because the gap is structural, not a matter of operator goodwill.

The second structural loss is the complaint route. A dispute at a licensed table, a misdeal, a voided round, a stalled withdrawal, runs first through the operator’s own complaints procedure and then through an independent Alternative Dispute Resolution body approved by the Commission. A player of an offshore operator has no access to a Commission-approved ADR for those transactions, so the operator’s support desk is the final word in most disagreements, and where a term is ambiguous the ambiguity is resolved by whoever holds the funds.

The third layer is the softer statutory machinery: customer funds held separately or protected to a stated level, reality checks on session length, account-level limit tools and recognised safer-gambling information, all of which Commission conditions require and offshore terms do not. Fund protection matters at the live tables more than anywhere else in the lobby, because the stakes per round are larger and the temptation to chase within a session is sharper when the dealer is waiting on your decision.

None of this changes the arithmetic of the wheel or the shoe, which is identical wherever the feed is carried. What changes is the machinery behind the balance: who holds the money, who adjudicates a disagreement, and whether the tools that exist in Britain to stop or slow play are present at all. A costed decision at the live tables includes those three facts alongside the margin figures, and this page exists to keep all of them in the same frame.

A session cost model you can check

The table below runs one representative live-dealer evening through the model with every intermediate figure visible. The assumptions are stated in the caption and repeated here: a £200 bankroll, £5 stakes, a two percent blended house margin reflecting mixed play across roulette and blackjack at category-typical rules, six sessions of about an hour each, and stakes that contribute at the full rate because no bonus is active. Change any assumption and every downstream figure moves with it.

Reading the table honestly means noticing how small the expected cost looks against the variance. Six sessions at £300 of turnover each cost £36 in expectation, but the swings at a £5 table routinely exceed £100 in an evening in either direction, and the bankroll’s survival depends on that variance far more than on the margin. The two percent figure is a blend: a pure blackjack evening at half a percent expects £1.50 a session, a pure roulette evening at 2.7 percent expects £8.10, and the blend sits between them depending on how the hours are split.

The model also shows why the bonus interaction from the third section matters so much at these stakes. If the same six sessions were played with an active bonus weighting live stakes at ten percent, none of the £1,800 of volume would count in full, and the playthrough would barely move while the expected cost of playing it ran exactly as above. The player would be paying the full price of the tables and earning none of the credit the bonus promised, which is the single most expensive misunderstanding in this category.

Finally, the model takes no account of comp schemes, cashback on losses or any promotional attachment, all of which operators use to soften table economics retroactively. Those attachments carry their own terms and their own weighting rules, and on the figures used here none of them changes the sign of the model, though a generous live-cashback term can narrow the gap on the operators that publish one. Verify any such term on the operator’s own page before relying on it.

How the market compares, row by row

The comparison below sets the British licensed baseline against offshore operators of the kind this page discusses. The columns that matter at the live tables are the licence label, the shape of the welcome range, the typical multiplier, the deposit minimum, the payout window and the rails the cashier carries, because the table product itself is shared and the wrapper is where the classes differ.

Figures in this table are category-typical ranges compiled from operator-published terms at the time of writing, not quotes of any single promotion. The UKGC rows are the British baseline shown for comparison; none of them is claimable at the offshore offers this page sits beside.

The pattern reads as a trade rather than a ranking. The licensed rows carry smaller printed numbers but contractual payout windows, published table rules audited to a recognised standard, and the full Commission machinery behind every hand. The offshore rows print larger matches at larger multipliers with lower stake floors, on cashiers that clear crypto fastest and everything else slowly, with the protections described above absent by design. The live feed may be the same studio in both columns; everything wrapped around it is not.