Studio rules, not reel rules: the edge by game

A live table looks like television and settles like arithmetic. Cards are dealt by a person in a studio, the wheel is spun on camera, and the rules printed on the table interface, the deck count, the payout ladder and the side-bet menu, decide what the game costs to play. That cost is the house edge, and it moves more between live games than it does between studios running the same game. Reading the studio rules is therefore not decoration; it is the step that decides which column of the price list an evening lands in.

The model on this page carries three category-typical values for the main live tables. Blackjack sits at 0.5%, baccarat at 1.06% and roulette at 2.7%. These are working figures rather than quotes from any single studio, and they do the job a budget needs: they order the games by cost and put a number on the gap between them.

At a £5 stake the spread is easiest to feel in pence. A hand of blackjack costs 2.5p in expectation, a baccarat hand about 5.3p, and a spin of the wheel about 13.5p. Over a hundred hands at the same stake, the expected cost runs from £2.50 at the blackjack table to £13.50 at the roulette one, a five-fold difference for the same hour and the same banknote. The percentage looks small in every column; the cash column is where the gap shows.

Two studio-side details bend the numbers further. Side bets and bonus spots sit above the main game in modelled cost, so a menu of extras is a menu of prices. And the pace of a table sets how much turnover an evening produces: a slow table with a thin edge can cost less per hour than a fast one at the same edge, because the clock multiplies the per-hand figure. Neither detail changes the ordering of the three main games; both change the bill.

Watching costs nothing either. Most studios let an account open a table and observe without staking, and a few minutes of observation settles the questions that matter for the model: the actual minimum, the rhythm of the dealing and how long the betting window stays open. None of that appears in a payout percentage, and all of it appears in the evening.

None of this argues that one game is right and another wrong. It makes the price visible before the first hand. The rest of the page takes a blended 1.5% between the thin end and the thick end and builds the session arithmetic on that blend, and the figure above is the reason the blend can be defended: the modelled middle sits between 0.5% and 2.7% with room on both sides.

Table minimums and the shape of a session

Minimums do the quiet work in live play. The minimum a table prints sets the smallest unit a bankroll can be cut into, and the number of units, not the number of pounds, decides how long a float lasts and how wide the swings feel. Two players can arrive with the same £200 and have completely different evenings, purely because one sat where the units were plentiful and the other sat where they were scarce.

This model takes a £5 working minimum, because £5 is the stake the calculator in the next section runs with. At that level a £200 float is forty units. At a £10 minimum the same float is twenty units, and every swing of one bad hand costs twice as much of it. Where a studio serves lower-stake tables the float stretches the other way, and the unit count is what the session actually spends. The interface shows the minimum before you join, which makes it the cheapest reading of the evening.

Minimums also anchor the expected-cost arithmetic from the first section. The per-hand figures there, 2.5p, 5.3p and 13.5p, scale in a straight line with the stake: a £1 roulette spin costs 2.7p in expectation, a £25 spin costs 67.5p, and the edge percentage never moves while the cash does. Fix the stake first, check the minimum second, and let the session run third. Variance then does what variance does on top of the model, in units the bankroll can either absorb or not.

Pace belongs in the same conversation. A live round takes longer than a reel spin: dealing, betting windows and settling all eat seconds, so the number of hands an hour is lower at a table than at a machine. Fewer hands an hour means less turnover for the same clock time, which trims the expected cost of an evening even when the stake and the edge stay put. Forty hands is a sitting of reasonable length rather than a full evening of continuous play, and stretching a session across more hands is how the same float ends up turning over more money than planned.

The session model, worked and rerunnable

The calculator on this page ships prefilled with the model this site uses for a month of live play: a £200 bankroll set aside at the start, £5 a hand, a blended house edge of 1.5% and four sessions. The blend is the deliberate middle of the first section, above the blackjack line and below the roulette one, and it fits a month in which most hands land at the cheaper tables and a minority land at the expensive ones. Every setting is an assumption, and every setting can be changed.

Worked at those settings the arithmetic is short. A session of forty hands turns over £200, because forty hands at £5 is £200 of stakes. At 1.5%, the expected cost of one session is £3. Four sessions turn over £800 in total and carry an expected cost of £12, which is 6% of the float. Expressed per hand, the model prices a £5 hand at 7.5p of expected cost no matter how the evening lands or which table it lands on.

Two caveats keep the model honest. The expected cost is an average across evenings, not a forecast of any single one, and a live session can finish a long way from its expectation in either direction, because swings at a table are measured in units while the expectation trickles in pence. And the £200 works as a cap as much as a fund: when the float is spent, the model says the month is over, not that the next deposit is owed to the ledger. The calculator exists so the second set of numbers, the ones built on your own stake and your own session count, are on the table before the first hand is.

Why four sittings rather than one long one? Because the model treats live play as a planned expense with a shape, not a mood. Splitting a month into sittings gives each one the same stake discipline, the same stopping point and the same trip through the cashier, and it stops a good evening from quietly becoming a longer one at a worse table. Changing the session count moves the turnover and the expected cost in a straight line, which is exactly the sensitivity a budget wants to see before the month starts.

Run it with your own settings:

Live play inside bonus wagering: weightings and exclusions

Bonus offers price their playthrough on the reels, and live tables are the awkward guest at that party. In the terms of this part of the market, live dealer games are commonly excluded from wagering altogether or weighted at zero to ten per cent, so £100 staked at a table with a 10% weighting counts £10 towards the playthrough, and the same £100 at a 0% weighting counts nothing at all. These are category-typical ranges, not quotes of any single promotion, and the exact weighting for an account sits in its own bonus terms.

The arithmetic of a live-first bonus plan is bleak enough to run once. Take a £50 bonus carrying a 40× playthrough: that is £2,000 of wagering to clear. On products weighted at 100% it is £2,000 of turnover. On live blackjack weighted at 10% it becomes £20,000 of table turnover, which at this page’s blended 1.5% edge carries an expected cost of £300 to release a £50 bonus. At a 0% weighting the playthrough never moves, however long the session runs. The multipliers printed in the table in the next section run from the low thirties to fifty times, and they sit on top of whichever weighting applies, so the two numbers multiply rather than add.

The practical reading is unglamorous. A welcome bonus is a slots-shaped instrument, and live play is what the stake money does outside it. If a live session is the plan, the cleaner route is usually to decline the bonus, keep the stake money unencumbered and let the withdrawal request follow the session free of wagering conditions. Where a bonus is taken anyway, the weighting line in the terms decides how much of the evening counts, and reading it before the first hand is the whole trick.

The cashier after the session: windows on table wins

A win at a live table leaves through the same cashier as any other online win and passes the same stages: the request, the operator’s pending review, processing, and then the transit time of the payment rail itself. Table winnings carry no separate queue and no special window. What separates accounts in this market is the speed of each stage, and that is where the shopping happens.

The payout windows in the table in the next section run from one to two banking days at the fast end, through one to three in the middle, out to one to five at the slower end. Those are category-typical ranges compiled from operator-published terms, not quotes of any single cashier. The rail moves differently inside the range: card payouts wait on the banking system, while the crypto routes some accounts offer can settle in minutes to hours.

Verification is the stage a first withdrawal meets head on. A UK-licensed operator must verify a player’s age and identity, and document checks are a condition of licensing rather than an operator’s choice, so the first payout request after a table win is commonly where the paperwork lands. Offshore accounts run their own version of these checks under their own rules, with no British register to escalate to if a review drags.

Two habits keep the stage short. Clear the account checks before the session rather than after it, and request the payout on the same rail that funded the deposit, because a mismatch between the two gives a review a reason to reopen. Neither habit changes the window printed in the terms; both reduce the chance of adding days to it.

Licences, offers and payout windows compared

The single comparison table on this page sits in this section, assembled from operator-published terms at the time of writing. Nine rows carry a UK Gambling Commission licence and three sit offshore without one, and the licence column is the one that decides which protections travel with the account. The welcome bands, the wagering multipliers, the minimum deposits, the payout windows and the payment rails are category-typical ranges for shopping rather than quotes of any single promotion. No row is a recommendation; the columns are the raw material the rest of the page models.

Figures are category-typical ranges compiled from operator-published terms at the time of writing, not quotes of any single promotion. UKGC-licensed rows are the British baseline shown for comparison; none of them are claimable at the offshore offers modelled on this page. Verify every term on the operator’s own page before you take it.

The licence behind the stream

The licence behind a live casino is the part that never appears on camera. Providing remote gambling facilities to people in Great Britain requires an operating licence from the Gambling Commission, and that licence binds the operator to the Licence Conditions and Codes of Practice, including the social responsibility requirements on customer interaction and product design, with the Commission able to fine, suspend or revoke when the conditions are breached. The minimum age for casino products in Great Britain is 18, and age and identity verification is a condition of the licence rather than a courtesy the operator extends. An account without a UK licence sits outside all of it: no entry on the public register, no GamStop coverage, no free dispute route and none of the fund-protection levels. That is the trade-off the licence note beneath this article keeps beside every figure on the page, and it is the one column of the comparison table that no welcome band can compensate.