Live casino UK: what the lobby is and how its economics differ from reels

A live casino lobby is a set of real tables, streamed. There is a dealer, a wheel or a shoe or a wheel of prizes, cameras, a betting interface and a chat window, and the result comes from the same physics that governs a casino floor rather than from a random number generator behind a screen. That matters for costing, because the mathematics of a real wheel is public property: a single-zero wheel takes 2.7 percent of everything staked over the long run, and no skin, theme or presenter can change it. The lobby can be dressed however the operator likes; the arithmetic underneath is fixed.

Reels behave differently. An online slot publishes a return to player figure, but the experience is built on volatility: long droughts, sudden spikes, and a distribution whose average only becomes visible across enormous numbers of spins. Live tables are the opposite temperament. The edge is smaller, the results sit closer to the average hour by hour, and the cost of play is nearly linear in time. That is why the honest question at a live table is not what might happen but what the hour costs, and the answer is multiplication rather than mystery.

The economics of the studio explain the limits. Dealers, cameras, croupier training and studio floor space are expensive, and they cost the same whether a table is busy or empty, so every live table carries a minimum stake that covers its keep. Reels have no such floor, because a server slot costs almost nothing to run, which is why a spin can be priced at 10p while a live hand cannot. The minimum at a live table is a rent figure as much as a gambling one, and it is the first number to read, ahead of any bonus the lobby advertises on its front page.

There is also a legal difference that shapes limits, and it belongs up front. At operators licensed by the Gambling Commission, online slots are capped at £5 per spin for adults and £2 per spin for 18 to 24 year olds, limits in force since September 2024. Those caps are written for slots; live tables sit outside them, so table minimums and maximums are set by the operator rather than by the statute. Offshore lobbies hold no UK licence at all, so even the slot caps lapse there, and the ceiling on a round is whatever the game allows. The licence section below unpacks what that difference costs a reader in practice.

The method on this page is costing, and it takes three inputs: the house edge of the game, the stake per round, and the pace of the table in rounds per hour. Multiplied together they give the expected hourly cost of play, the share of turnover the edge quietly takes back while the session runs. Each input is stated as an assumption where it has to be one, each published figure is labelled as one, and nothing is rounded in anyone’s favour. Where a multiplier has to be chosen, the choice is printed next to the number it produces, so the reader can argue with it and re-run it.

One limit on the method belongs in the open. Minimums move, studios retire tables, and a game costed today may run under different terms next month. The page therefore costs structures rather than named tables: the standard games at their published edges, at typical paces, at stakes a reader chooses. That is slower than a verdict, and it survives contact with next month’s lobby, which is the only test a costing has to pass.

The table-costing model: edge, pace and the hourly bill

The model is one line of arithmetic: expected hourly cost equals rounds per hour, times stake per round, times house edge. Nothing else enters it. The edge is the published share of turnover the game takes over the long run; the stake is what goes on the felt per round; the pace is how many rounds the table resolves in an hour. Change any input and the bill moves in exact proportion, which is what makes live tables so much easier to price than reels, where the same hourly figure would need thousands of spins to become visible.

The edges used here are the published rates for the standard versions of each game. Blackjack played with basic strategy costs about 0.5 percent. Baccarat, on the banker bet, costs about 1.06 percent. European roulette, the single-zero wheel, costs 2.7 percent. The game shows, the money-wheel and prize-round formats that dominate the front of many lobbies, run around 4.5 percent on the common bet patterns. Each figure is a long-run rate, not a promise about any hour, and the blackjack figure assumes basic strategy throughout, because playing hunches instead of strategy widens that edge several times over.

Pace is the input most readers skip, and it does as much work as the edge. The costing runs at forty rounds an hour for the wheel games, reflecting the spin cycle of a standard live studio table, where betting windows, ball drop and payouts set a rhythm no player can hurry. Card games deal faster, and the model takes fifty to sixty rounds an hour for blackjack and baccarat. The game shows sit back at forty, held down by their bonus rounds and their presenters. Every pace figure is an assumption, printed here, and every row of the table states the pace it uses.

The table below runs the line of arithmetic across the lobby at £5 stakes. Typical minimums are category-typical figures of the kind printed on live tables, not quotes of any single table. The rows are ordered from the cheapest hour to the dearest, and the pace column is the one that separates the pairs.

Read the spread, because the spread is the finding. At the same £5 stake, an hour of basic-strategy blackjack costs £1.50 in expectation and an hour of the game shows costs £9.00, a multiple of six for two formats that sit a few tiles apart in the same lobby. Moving from a standard roulette table to a speed variant lifts the hourly bill from £5.40 to £8.10 without moving the edge by a penny. At fixed stakes, the choice of game and the pace of the table move the cost of an evening further than any welcome offer ever will.

Single sessions price from the same line of arithmetic, and the calculator below runs them directly. It takes a bankroll, a stake per round, a house edge and a number of sessions, and it returns the expected cost of that volume of play. It is a costing tool, not a forecast. It does not know a lucky shoe from a cold one, it does not care how the last session went, and it does not schedule a recovery; it prices the arithmetic of volume, which is the only part of gambling that behaves predictably.

The note behind the numbers: the prefill sets a £100 bankroll, £5 stakes, the 2.7 percent European roulette edge and ten sessions. At that edge, every £5 round carries an expected cost of 13.5 pence, the per-round figure the table above is built from, so a forty-round hour at the wheel prices out at about £5.40 and ten such evenings at about £54 against the bankroll in the prefill. Change the edge to 0.5 percent and the same stake prices at £1.50 an hour; the tool is honest about whatever it is fed. Real weeks are lumpy and the model flattens them on purpose, so the cost of the volume shows in one line rather than scattered across moods.

What the model refuses to do matters as much as what it does. It does not project recoveries, endorse staking systems, or imply that a losing hour is owed a winning one; nothing in the arithmetic schedules a turnaround, because nothing in volume play produces one. The expected cost is a floor on the average price of the entertainment, not a ceiling, and individual sessions land above and below it with no memory of each other.

The edges across the lobby, charted

Percentages hide their consequences behind decimals, so the chart below puts the four edges in the model on one axis. Blackjack’s 0.5 percent and baccarat’s 1.06 percent hug the floor; European roulette’s 2.7 percent sits in the middle; the game shows’ 4.5 percent towers over the rest, highlighted as the expensive corner of the lobby. The bars are shares of turnover, so a bar twice as tall charges twice as much on every pound staked.

Read the chart as a price list rather than a verdict on any game. Every format is priced fairly for what it is; the game shows trade a higher edge for bonus rounds, presenters and production value, and plenty of players decide that trade is worth making. The costing exists so the decision is made with the price visible, not so the entertainment is refused.

Two notes travel with the picture. The blackjack bar assumes basic strategy, so it is a floor a player builds up to rather than a default. And the bars measure edges, not hourly costs, because pace multiplies them: the fuller bill is the one the table in the model section prices, where the same edges meet stated paces and stated stakes.

Table limits, pace and the rounds-per-bankroll arithmetic

Minimums first, because they are the gate to everything else. Live minimums exist to cover the studio’s costs, and they vary by format: card tables commonly open higher than the wheel, the game shows open lower than either, and the speed variants undercut the standard tables to keep their volume up. A typical minimum is a category figure, not a quote of any table, and the number that matters to a reader is the gap between the minimum and the stake a bankroll can sustain. For a £5 player, a £1 roulette minimum and a £5 blackjack minimum are different doors into the same building, and the table in the model section shows what each door costs per hour.

Maximums matter less to most readers and enormously to a few. Every live table runs between its minimum and its published maximum, and the maximum is where a lobby says plainly whom the table is for. One caution belongs here: the statutory stake caps do not reach live tables, so at both tiers the table’s range is set by the operator’s policy rather than by law. What the licence changes is everything around the table, the recourse and the protections, and the licence section below takes that up in full.

Pace is the quiet multiplier. The costing runs at sixty hands an hour for blackjack, reflecting how quickly a dealer clears a shoe once the table is flowing, and at forty rounds an hour for the wheel. A player who moves from a standard roulette table to a speed variant has not changed the game’s edge by a penny; the hourly bill has risen by half regardless, because the edge is charged on turnover and pace decides how much turnover an hour contains. Time at a table is the unit the bill is metered on.

The rounds-per-bankroll arithmetic follows directly. A £100 bankroll at £5 a round buys twenty rounds at full exposure, which is half an hour at the standard wheel pace. Expected survival runs far longer, because the edge only takes its small share per round: at the 2.7 percent roulette rate the expected cost of a round is 13.5 pence, so £100 prices out at roughly 740 rounds in expectation, near nineteen table hours. At the blackjack edge the same bankroll prices at roughly 4,000 rounds. These are expectation figures, not promises: variance ignores averages on any single evening, and the same bankroll can be gone in twenty rounds or last for months.

What the arithmetic is for. The fewer rounds a bankroll buys, the more the session is decided by variance and the less by the numbers; the more rounds it buys, the closer the session sits to its expected cost. Neither outcome can be scheduled, which is precisely why the numbers are worth knowing in advance. A session that names its stake, its pace and its length before the first bet is a session with a price tag; a session that decides these things at the table is priced by the table instead.

Session length completes the frame. Forty rounds at the wheel is an hour; twenty rounds is half of one. The calculator in the model section prices exactly that shape of evening, and the tools listed in the care section below are what hold the shape in place at a licensed operator. The limits section of the terms, the cashier’s minimum and the table’s own posted range do the rest, and all of it is readable before a pound is deposited.

The comparison table: licences, offers and payment rails

The table below gathers a slice of the market onto one page so the licence label, the offer ranges and the payment rails can be read side by side. Bands are category-typical and vary row by row, and the licence column is the one that outranks the rest, because it decides whether everything printed beside it is enforceable or merely published. The descriptor beside each name says what kind of operation runs the till, which is the shortest summary of what the cashier will look like.

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

Two cautions travel with any such table. Ranges describe categories, not promises, and an individual cashier can sit outside its band in either direction at any time. And the licence label outranks the row: a generous offer at a lobby with no enforceable complaints route costs more than it looks, while a plainer offer behind a regulator costs exactly what it prints. Read the rows in that order and the table does its job.

The rails column also feeds the live lobby question directly. Offshore-facing tills lean on crypto and e-wallets such as MuchBetter, which settle on their own rhythms, while the UKGC rows cluster around Visa, Mastercard, PayPal and plain bank transfer. Which rail a cashier leads with tells you something about the operator long before the terms do, and the deposit minimum in the table should be checked against whatever the welcome terms demand before the first funding.