A £0.20 spin on a 96 percent return-to-player game expects to cost 0.8p, which is £2.40 an hour at 500 spins, and the difference between a 93 percent and a 97 percent game is more than a third of that cost.
Slots are priced in pence per spin and paid back over thousands of rounds, so the honest way to read any of them is through the return-to-player figure, the volatility of the game and the stake you set, all of which sit in the game’s own information page before you wager a penny. The board below carries an editorial score out of 100 for each entry, a set of term chips under each row and its own comparison, and the guide underneath works out what the published numbers actually cost per hour, per session and per bonus wagered.
The rack below is a paid placement: this site is paid a commission when a reader opens an account through one of these rows, and that payment is what sets their order. The costing underneath is not for sale, and every figure carries the date it was read.
Five accounts, costed on 8 November 2026
Every figure here was read from the operator’s own terms on 8 November 2026. Terms change often, and the operator’s page is the only binding record.
What a spin actually costs
Every slot spin has a price and an expected return, and both are knowable before the reels move. The price is the stake you set, which on most modern five-reel games runs from a penny or two up to several pounds per spin. The expected return is the stake multiplied by the game’s return-to-player percentage, usually printed on the game’s information screen. A 96 percent game returns 96p of every £1 wagered over the long run, which means it expects to keep 4p. That 4p is the cost of playing, and once a reader sees the cost stated that plainly, every other decision on this page follows from it: how long a bankroll lasts, which game family to pick, how big a stake to set, and whether a bonus with wagering attached is worth the schedule it demands.
The reason the cost figure matters more than any headline is that slots are the highest-margin game in most lobbies. A European roulette table expects to keep 2.7 percent of everything wagered, and blackjack played to basic strategy expects to keep around half a percent. A typical online slot expects to keep 4 percent, and jackpot variants expect to keep 7 percent or more. Playing slots is not a mistake, and the entertainment per pound can be excellent, but the margin difference is large enough that a reader who plays both should at least know they are paying a different rate. A £500 evening of blackjack at basic strategy expects to cost £2.50; a £500 evening of slots at a 4 percent house edge expects to cost £20. Same money wagered, eight times the expected cost, and the difference is entirely in the game’s mathematics rather than in luck on the night.
Cost also scales with speed, and speed is where slots quietly outprice everything else in the building. A live roulette table deals perhaps thirty to forty rounds an hour. A slot at autoplay runs five to eight spins a minute, which is three hundred to nearly five hundred rounds an hour, and a fast player clicking manually can approach the same rate. Multiply the per-spin cost by that speed and the hourly figures arrive. At £0.20 a spin on a 96 percent game, 400 spins an hour expects to cost £3.20. At £1 a spin on a 93 percent game, the same hour expects to cost £28. Neither number is a prediction of tonight, because variance dominates any single session, but both are the honest expected price of the hour, and the second one is nine times the first for what looks on the surface like a similar activity.
The practical consequence is that stake, not game choice, is the biggest lever a slots player controls. The gap between the best and worst return-to-player figures commonly found in a lobby is three or four percentage points, which roughly doubles the hourly cost at a fixed stake. The gap between a £0.10 stake and a £1 stake is a factor of ten, which dwarfs anything the game choice can do. This page therefore spends more of its length on stake sizing than on game selection, not because the return-to-player figure does not matter, but because the order of operations matters: fix the stake to the bankroll first, then pick the game whose volatility and return suit the session you want, and only then think about features and themes, which are the part of the decision that costs nothing to get wrong.
One framing that helps is to think of the expected cost as an entertainment price rather than as a stake in a contest. If an evening of slots at £0.20 a spin expects to cost £3 an hour, that is cheaper than most paid entertainment, and a reader who treats it that way, with the stake set so that the expected cost fits the budget, is pricing the activity correctly. The trouble arrives when the stake is set so that the expected cost exceeds what the evening was meant to cost, which happens easily because variance disguises the price. A session can run above expectation for hours at a time, and a reader who judges the cost by the balance rather than by the arithmetic will routinely believe slots are cheaper than they are. The tables and calculator further down exist to keep the arithmetic in front of the balance.
A last housekeeping point before the maths. Return-to-player figures, spin speeds and feature descriptions are published by the game studios on the game’s own information pages, and they can be configured differently by different operators, so the same title can carry a slightly different return figure in two different lobbies. Where this page needs a return figure for its arithmetic it uses the range commonly published across popular families, states the assumption next to every calculation, and treats the specific game’s own information page as the only figure that matters for the session actually being played.
How return to player works, and how long it takes
Return to player, always abbreviated to RTP, is the share of everything wagered that a game pays back over a very large number of spins. It is a property of the game’s paytable and reel configuration, it is measured across millions of rounds by the studio, and it is the single most important number on the game’s information page. A 96 percent RTP means that if every spin in the world of that game were totalled up, the players collectively would hold 96p of every £1 they had wagered, and the operator would hold the remaining 4p. It is not a promise about any session, any evening or any player, and the distance between that long-run average and any single session is the entire subject of the section after this one.
The first thing to understand about the figure is how slowly it arrives. The long run is genuinely long. Simulation studies of slot mathematics, and the studio documentation that accompanies the games, put the number of spins needed before the realised return reliably approaches the stated RTP in the tens of thousands to hundreds of thousands of rounds. A player putting in an evening of 400 spins has played a fraction of a percent of the sample the figure describes, which is why session outcomes scatter enormously: a single bonus round on a volatile game can return several hundred times the stake, and a cold hour can return almost nothing. Over a year of regular play, the realised return of a heavy player will drift toward the stated figure; over an evening, it is noise, and pleasant or unpleasant noise at that.
The second thing to understand is what the figure does not include. Published RTP describes the game’s base mathematics, and it usually assumes the player takes every spin the paytable offers at its standard configuration. It does not describe the effect of playing with a bonus attached, where the operator’s terms sit on top of the game’s maths and can convert a winning spin into a forfeited balance if a condition is breached. It does not describe progressive jackpot contributions, where a slice of each stake feeds the jackpot pool and the base game’s return is correspondingly lower unless the jackpot share is counted in. And it does not describe the difference between the version of a game an operator has configured and the studio’s default, which is why the figure worth checking is the one printed in the lobby you are actually playing in, not the one you remember from a review.
The third thing is the spread of figures across a lobby. The bulk of modern five-reel video slots publish between 95 and 96.5 percent. Classic three-reel games often sit lower, around 93 to 95 percent, compensated historically by simpler play. Megaways-style games cluster around 96 to 96.5 percent. Jackpot slots sit lowest of the regular families, commonly 92 to 94 percent on the base mathematics, because a share of every stake funds the pool. Feature-buy variants, where the player purchases direct entry to the bonus round, often publish a slightly different figure for the buy itself than for the base game. The spread from the best regular game in a lobby to the worst is three or four percentage points, which at a fixed stake roughly doubles the expected hourly cost between the two extremes, and it costs nothing to check the information page before choosing.
Converting RTP into a per-spin cost is one line of arithmetic. Expected cost per spin equals the stake multiplied by one minus the RTP. At 96 percent and a £0.20 stake, that is £0.20 multiplied by 0.04, or 0.8p a spin. Multiply by spins per hour and the hourly price falls out: 400 spins at that rate expects to cost £3.20. The same arithmetic at 93 percent gives 1.4p a spin and £5.60 an hour, and at 97 percent it gives 0.6p and £2.40. The table in the next section but one runs the full grid of returns and stakes so the numbers can be read off rather than recalculated, but the one-line formula is worth internalising, because it lets a reader price any game in any lobby in ten seconds with the information page open.
Finally, a word on where the figure sits relative to the licensed baseline, because it is one of the few places British regulation touches slots directly. Games offered by GB-licensed operators must publish their RTP and must comply with Commission rules on slot design, including caps on spin speed and the removal of features that celebrate losses or accelerate play. Those rules shape the experience of playing on a licensed site in ways that are easy to feel once named: spins cannot be accelerated past the cap, losses cannot be dressed as wins, and the pace of an hour is bounded. The figure itself, though, is set by the studio and the operator’s configuration, and similar titles with similar returns appear across licensed and offshore lobbies alike. What differs between the two markets is not usually the game mathematics, it is the frame of protections and terms wrapped around the same reels, which the section near the foot of this page costs out.
Volatility: why two games at the same RTP feel nothing alike
Two slots can both publish 96 percent return and produce utterly different evenings. The difference is volatility, sometimes called variance or, loosely, hit profile, and it describes how the return is distributed across spins rather than how large it is in total. A low-volatility game pays small amounts frequently, trickles its 96 percent back in a steady drip of minor wins, and produces long flat sessions with shallow swings. A high-volatility game pays rarely and pays large when it does, concentrates most of its return into bonus rounds that may land once in several hundred spins, and produces sessions that run down slowly and then lurch. Both cost the same per spin in expectation. They do not feel remotely alike, and choosing between them is a decision about the shape of an evening, not about its price.
The practical marker of volatility that most players notice first is hit frequency, the share of spins that return anything at all. Low-volatility games commonly hit on a quarter to a third of spins, with most hits smaller than the stake. High-volatility games commonly hit on fewer than one spin in five, with most of the return packaged into features. Studios publish hit frequency on some games’ information pages and leave it implicit on others, but it is easy to feel within a hundred spins, and it is the number that decides whether a given bankroll survives long enough to see the game’s interesting moments. A high-volatility game with a bonus that lands once in 300 spins needs a bankroll sized for 300 spins of mostly-missing before the feature has its chance; funded at 250 spins’ worth of stake, the reader is more likely than not to end the session without having seen it.
That observation produces the single most useful rule in slots economics: volatility and bankroll depth must be matched, because a bankroll too thin for the game’s variance converts the game’s own design into a bad experience. The arithmetic is simple. Divide the bankroll by the stake to get the number of spins funded. On a low-volatility game, 150 spins’ depth is usually enough for the session to resemble the game’s average behaviour, because the frequent small returns keep replenishing the balance. On a high-volatility game, the same depth is usually gone before the feature arrives, and the player has effectively bought a ticket to the first third of an experience whose payoff sits at the end. The game was not the problem and the luck was not the problem; the sizing was. Playing the same game at a stake one notch lower doubles the depth and transforms the session.
The shape of the balance over a session differs too, and it is worth picturing because it explains a common complaint. On a low-volatility game the balance declines in a rough sawtooth, drifting down at close to the expected cost per hour with small recoveries, and a reader who checks the clock will find the evening matched the arithmetic quite closely. On a high-volatility game the balance declines in steps, long slides punctuated by single jumps, and the reader’s memory of the session is dominated by the jumps rather than the slides. Both sessions expected the same cost. The volatile one feels like it nearly paid, because at one point the balance was briefly up, and the feeling survives even though the ending matched the mathematics. Anyone who finds that pattern pulls at them harder than they would like should treat it as a reason to prefer steadier games and lower stakes, and the resources at the foot of this page exist for exactly that conversation with oneself.
Volatility also changes the arithmetic of wagering a bonus, which is the subject of its own section below, but the short version belongs here: a wagering schedule measured in thousands of pounds of play is a fundamentally different proposition on a low-volatility game, where the balance erodes predictably toward the expected loss, than on a high-volatility game, where the whole schedule can be won or lost in one feature. Readers clearing bonuses tend to prefer low-volatility titles for exactly that reason, and where an operator restricts the games that count fully toward wagering, the restriction is worth reading precisely because it removes the steadiest option from the schedule.
How to find a game’s volatility when the information page does not state it: the paytable tells the story. A top prize in the hundreds or thousands of times the stake signals high volatility. A top prize of a few hundred at most, with a long ladder of mid-sized line wins beneath it, signals lower volatility. The frequency of features advertised on the splash screen, the presence of a feature-buy button, and the size of the maximum win all corroborate. None of this is secret; it is all on the game’s own pages, and reading it before the first spin is the cheapest minute in slots.
Expected loss per 100 spins, at every combination that matters
This section is the page’s own model, built on stated assumptions rather than any operator’s current configuration. The model takes five return-to-player figures, 93, 94, 96, 96.5 and 97 percent, which bracket the range commonly published across the popular slot families, and four stake levels, £0.10, £0.20, £0.50 and £1, which cover most careful play. The expected loss per 100 spins is the stake multiplied by 100 multiplied by one minus the return figure. Spins per session of 100 is deliberately modest, about a quarter of an hour at autoplay speed, so the table reads as a short-burst price rather than an evening one; multiply by four for an hour, by twelve for a long evening, and so on. The model assumes the published return applies at the stated stake, takes no account of jackpot contributions or feature buys, and treats variance as noise around the expectation rather than modelling it, because variance is the reader’s own evening and the expectation is the price of the ticket.
| RTP | House edge | £0.10 stake | £0.20 stake | £0.50 stake | £1 stake |
|---|---|---|---|---|---|
| 97% | 3% | 30p | 60p | £1.50 | £3.00 |
| 96.5% | 3.5% | 35p | 70p | £1.75 | £3.50 |
| 96% | 4% | 40p | 80p | £2.00 | £4.00 |
| 94% | 6% | 60p | £1.20 | £3.00 | £6.00 |
| 93% | 7% | 70p | £1.40 | £3.50 | £7.00 |
Read the table as a price list and three habits fall out of it. The first is that the stake column dominates. Moving down the table from the best to the worst return at a fixed stake at most doubles the cost; moving across from £0.10 to £1 multiplies it by ten at every return figure. A reader on a budget gets more control from the stake button than from any amount of game hunting, which is why the next section is about sizing the stake rather than about picking titles.
The second habit is to convert the table into hourly terms, because 100 spins is a burst rather than an evening. At the autoplay speeds most games run, roughly 400 to 500 spins an hour, the 96 percent row at £0.20 becomes about £3.20 to £4.00 an hour, and the 93 percent row at £1 becomes £28 to £35 an hour. The difference between a careful setup and a careless one is the difference between a cheap evening and a genuinely expensive one, and both feel identical from inside the session, because variance hides the rate. The clock and the arithmetic are the only honest meters.
The third habit is to notice what the jackpot row of the market implies. At a 93 percent return the house edge is 7 percent, more than twice the edge on a typical 96 percent game and fourteen times the edge on basic-strategy blackjack. Jackpot slots are not a mistake, and someone chasing a life-changing pool is buying a lottery-shaped product with entertainment attached, but the price of that shape should be chosen knowingly. A reader who plays jackpot titles at £1 stakes is paying £7 per 100 spins in expectation for the chance structure, and a reader who plays a 96.5 percent game at £0.20 is paying 70p for a very different one. Both are legitimate evenings. They are just priced differently, and the information page prices them in advance.
The model’s honest limitation is that 100 spins is too short a sample for the expectation to assert itself. Over 100 spins a high-volatility game at 96 percent can return anything from near nothing to several hundred times the stake, and the table’s middle cells are the centre of a very wide distribution, not a forecast. The table is a costing tool: it says what the spins cost in expectation, which is the number a budget needs. It is not a predictor of any particular hundred, and a reader whose sessions habitually land far from the expectation on either side is simply seeing the variance that the model, by design, does not draw.
Sizing the stake to the bankroll
Stake sizing is the part of slots economics the player actually controls, and it has a simple governing rule: the stake should be set so the bankroll funds enough spins for the session to resemble the game, rather than ending before the game has shown what it does. The rule converts directly into arithmetic. Bankroll divided by stake equals spins funded. For a low-volatility game, 150 to 200 spins of depth is a workable floor. For a high-volatility game, 250 to 350 spins of depth is the honest minimum, because the game’s return is concentrated in rare events and a shallow bankroll never reaches them. A £30 bankroll on a high-volatility title therefore argues for a stake of 10p to 12p, not the 50p the same bankroll comfortably supports on a steadier game. The bankroll did not change; the game’s variance set the price of admission.
The session budget belongs to the same calculation. If the evening’s entertainment budget is £10, and the chosen game at the chosen stake expects to cost £4 an hour, the session is priced at two and a half hours, which is a good evening. If the stake is set two notches higher, the expectation doubles or triples and the same £10 is an hour, and the reader has accidentally bought a more expensive product than they meant to. The expected-cost table in the previous section exists precisely so this multiplication can be done before depositing rather than inferred from a declining balance afterwards. Nothing about the game changes between the two setups; only the price per hour does, and the price is the reader’s to set.
Loss limits and session structure deserve plain treatment here, because stake sizing without a limit is only half a system. A loss limit is a figure decided while calm, written down, and treated as the end of the session when reached, not a target to be defended by raising stakes. The licensed market enforces such limits through account tools, and a reader playing anywhere can enforce one themselves with a note on the phone and the same discipline. Chasing a losing session by lifting the stake is the most reliable way to turn a small expected loss into a large actual one, because it multiplies the house edge’s per-spin take at the exact moment the balance is thinnest. The mathematics of the game is indifferent to the stake’s history; the player’s budget is not.
Autoplay deserves its own paragraph, because it is the feature that industrialises the cost. Manual play at a thoughtful pace runs perhaps 150 to 250 spins an hour, with natural pauses that give judgement a chance to operate. Autoplay at default speed runs 400 to 600, and on fast configurations more. The expected cost per hour is therefore two to three times higher at the same stake, and the number of decisions per hour, which is where a player notices the balance and reconsider, drops to near zero. A reader who uses autoplay should set its spin count to a figure that matches the session budget, not leave it at the maximum, and should treat the pause at the end of each batch as the intended checkpoint. Licensed games cap autoplay and spin speed by rule; elsewhere the cap is whatever the configuration allows, which makes the player’s own batch-setting the only speed limiter in the room.
Feature buys invert the sizing question and are worth costing separately. Where a game sells direct entry to its bonus round, the price is typically 50 to 150 times the base stake, and the bought feature’s own return figure is usually published separately and is often slightly lower than the base game’s. A £0.20 stake with a 100-times buy is a £20 single event with the volatility of the bonus round concentrated into it. The expected cost arithmetic still applies, the house edge still takes its share, but the variance is extreme: the buy is either a substantial win or, far more often, a substantial immediate loss. Readers who buy features should size them as individual events within the budget, not as spins, and should check the buy’s own return figure on the information page rather than assume the base game’s travels with it.
Finally, the stake interacts with the maximum-bet clauses that come with bonuses, and this is where slots differ from every other game in the lobby. A wagering schedule played at slots has a stake ceiling printed in the bonus terms, commonly £2 to £5 a spin, and exceeding it even once, often including through an autoplay batch or a feature buy set before the bonus was credited, commonly voids the bonus and anything won with it. The next section takes that interaction at length, because it is the single most expensive piece of small print a slots player meets.
Wagering a bonus on slots, and the maximum-bet clause
Slots are the default game for clearing a bonus because they contribute fully to wagering, where table games typically contribute a fraction or nothing, and that full contribution is what makes the slots-and-bonus combination the market’s standard package. The arithmetic of the schedule itself is covered in detail elsewhere on this network; the short form is that a bonus of £50 at a forty-times requirement, which is what this part of the market commonly asks, means £2,000 of qualifying play, and at a £0.50 stake that is 4,000 spins, which at a steady pace is ten or more hours of reels. The schedule is feasible on slots precisely because slots run fast and count fully. What the schedule costs in expectation is a separate question, and the answer is unwelcome: £2,000 of play at a 4 percent house edge expects to cost £80, which on a £50 bonus means the expected value of the exercise is negative before any conversion cap is applied. A bonus is best read as purchased playing time, not as money with a delay attached.
The maximum-bet clause is where that playing time can be forfeited outright, and it deserves to be read word by word. Bonus terms at this end of the market commonly cap the stake while wagering is live at £2 to £5 per spin, and the standard consequence of exceeding the cap is forfeiture of the bonus and any winnings attached to it. The clause bites slots players in three unglamorous ways. First, the stake button: a reader who has been wagering at £0.50 and nudges to £6 for a few spins of interest has breached it, and some terms apply the breach retroactively to the whole bonus. Second, autoplay: a batch configured at a legal stake before the bonus credited, or configured with a bet adjustment that lands above the cap, can breach the clause without a single conscious decision. Third, feature buys: the purchase price is a multiple of the stake, and a buy priced above the cap-equivalent is commonly treated as a breach even though the base stake was legal.
The defence is procedural rather than clever, and it costs a minute. Before wagering begins, set the stake comfortably below the printed cap, not at it, leaving margin for an accidental nudge. Reconfigure or disable autoplay so no batch can exceed the cap while unattended. Leave feature buys alone for the duration of the wagering unless the terms expressly permit them. Check whether the terms cap the stake per spin or per round, and whether the cap differs during a bonus round from the base game, because a few operators word it differently and the difference decides whether a feature triggered at a legal stake is safe to play out. None of this changes the expected cost of the schedule; all of it protects the playing time the schedule buys from a single-spin error.
Two further terms shape the slots side of a bonus and are worth naming. Game weighting is the first: most offers count slots at 100 percent but exclude or down-weight a list of titles, usually the highest-return games and the feature-buy family, because a player clearing a schedule on the lowest-edge games costs the operator the most. A reader whose plan was to grind the schedule on the steadiest 97 percent title in the lobby should check that title against the excluded list before starting, because the substitution to an allowed game at 94 percent raises the expected cost of the same £2,000 schedule from £60 to £120. The second is the conversion cap, the maximum the bonus funds can turn into withdrawable cash, commonly £100 to £500 on offers of this kind. A cap converts the schedule into a bounded trade: the maximum upside is fixed while the expected cost is not, and a reader costing a bonus should write both numbers down before the first spin.
Withdrawal during wagering is the last trap in the chain. Withdrawing before the wagering completes typically forfeits the bonus and winnings, and some terms extend forfeiture to the deposit itself, which is a clause to read twice. For a slots player, the practical shape is this: once the bonus is credited, the balance is committed for the length of the schedule, at a stake under the cap, on games that count fully, until the wagering completes or the bonus expires. A reader who does not want that commitment should decline the bonus at the deposit step, where most cashiers offer the choice, and play their own money free of any wagering attachment. Declining a bonus that does not fit the session one wanted is the most profitable single decision available in the slots lobby, and it costs nothing.
The licensed comparison, briefly and without a second regulatory section: GB-licensed operators work under Commission rules on how bonus promotions must be presented, which require the significant terms to be shown prominently and limit some of the practices that made bonus small print genuinely hazardous. Offers without a British licence sit outside those presentation rules, which means the burden of reading shifts entirely to the reader, and the maximum-bet clause, the weighting list and the conversion cap are all still there, just less prominently flagged. The arithmetic does not change between the two markets; the prominence of the small print does, and this page’s method assumes the reader does the reading.
Work out your own session numbers
The calculator below runs this page’s session model on whatever figures suit the reader. It takes a bankroll, a stake, a house edge assumption and a number of sessions, and it returns the expected cost across them, using the same arithmetic as the tables above: expected cost per spin equals stake multiplied by edge, and the total scales from there. The prefill matches this page’s model throughout: a £30 bankroll, a £0.20 stake, a 4 percent edge, which corresponds to a 96 percent return-to-player game, and eight sessions, because that combination is a realistic careful setup for a regular slots player and it prices an evening at roughly £2.56.
- Rounds
- 150
- Per ses
- 0.01
- Total
- 0.1
Expected cost assumes the house edge of 0.04 percent applies to every round at a flat stake.
The assumptions deserve stating in full, because the calculator is a model and not a forecast. It assumes the house edge is constant across the stake and the session, which is true for fixed-odds slots within a game’s stake range. It assumes the reader plays to the expectation rather than stopping at a limit, so a real session with a stop-loss will usually cost less than the model and a real session that runs to the budget exactly will land wherever variance puts it. It takes no account of bonuses, wagering, caps or game weighting, all of which sit on top of the game mathematics and are costed in the section above. And it rounds nothing in the reader’s favour or against; it is the same one line of arithmetic the table uses, run at whatever scale the reader plugs in.
Three runs of the calculator answer most practical questions before a deposit. Run it at the intended stake and the chosen game’s published return to price the evening. Run it at the next stake setting up to see what one notch of ambition costs per hour, which is usually enough to settle the stake at the lower figure. And run it at the return figure of a jackpot variant being considered, to see the premium the pool structure charges, which at 93 percent against 96 percent is nearly double. Two minutes with the calculator and the lobby’s information pages turns game selection from a feeling into a price, which is the whole method of this site.
One caution on reading the output. The figure the calculator returns is an expectation across the sessions entered, not a guarantee of any of them, and the site does not use that word because no arithmetic delivers one. Variance will put individual sessions above and below the expectation, sometimes far, and the model’s value is that the average across many sessions should approach it. A reader who finds their realised results running consistently well past the expectation, in the expensive direction, session after session, is either playing more spins than they are counting or playing at a higher effective edge than they think, and both discoveries are worth making with the calculator rather than the balance.
Where slot families sit on return to player
The chart below shows the return-to-player figures most commonly published across the four broad slot families, as a range rather than a point, because individual titles within each family vary with the studio’s design and the operator’s configuration. The figures are compiled from published game information across the popular families and treated as category-typical, not as quotes of any single title. Classic three-reel games sit around 94 percent, five-reel video slots around 96 percent, the Megaways family around 96.5 percent, and jackpot slots around 93 percent, with the shortfall against the other families funding the progressive pools.
Reading the chart, the gap that matters most is between the Megaways family and the jackpot family, three and a half percentage points, which at a fixed stake is nearly a doubling of the expected hourly cost between the two. A reader moving from a 96.5 percent title to a 93 percent one at the same £0.20 stake lifts the expected cost from £1.40 to £2.80 per 100 spins, and the only compensation is the pool structure the shortfall funds. That trade is legitimate when chosen and expensive when stumbled into, which is why the return figure belongs on the checklist next to the theme and the features rather than being discovered afterwards.
The second reading is that the best regular families cluster tightly. Five-reel video slots and Megaways titles mostly sit within half a point of each other, so within that cluster the choice can be made on volatility, theme and features with almost no cost consequence, and the cost decision is then made entirely by the stake. This is the division of labour this page has argued throughout: the family sets the return within a narrow band, the game’s volatility sets the shape of the session, and the stake sets the price. A reader who fixes the stake first and chooses freely within the tight cluster afterwards has got the economics right without sacrifice.
The third reading concerns configuration. Studios publish default figures, and operators configure games within permitted ranges, so the same title can appear at slightly different returns in different lobbies. The figure printed on the game’s information page in the lobby being played is the one that applies, and where two versions of a family sit side by side at different figures, the difference is real money at the stake in question: half a point on £1 stakes is 50p per 100 spins, £2 an hour at autoplay pace. Checking the information page takes seconds and is the only verification that costs nothing.
What a licence buys, and how the market compares
This is the page’s one regulatory section, and it exists because the frame around the reels, rather than the reels themselves, is where the two halves of the market differ most. The game mathematics are broadly shared: the same studios, the same families and similar published returns appear in licensed and offshore lobbies alike. What travels with a British licence and does not travel with an offshore offer is a set of protections that a slots player feels directly, and the honest commercial comparison is what each of them would have bought.
For a slots player specifically, three of those protections bite hardest. The first is the slot design rules: licensed games are capped on spin speed, cannot celebrate losses as wins, and cannot offer features that accelerate play, so the pace and the honesty of the experience are regulated rather than configured. The second is the limit framework: deposit limits that take effect immediately, reality checks at set intervals, and a self-exclusion scheme covering every licensed operator at once. The third is the dispute route: a withheld withdrawal at a licensed operator can be taken to an independent adjudicator and the regulator stands behind the process. At an offer with no British licence, none of those travel, the pace is whatever the configuration allows, the limits are the operator’s own policy if any, and a dispute ends where the operator’s support team says it ends.
The comparison table below puts the offshore-facing rows alongside licensed British operators so the differences carry faces. The figures are category-typical ranges compiled from operator-published terms at the time of writing, not quotes of any single promotion, and they describe what each kind of operator commonly publishes rather than what any one of them is promising today. The UKGC rows are the British baseline shown for comparison and none of them are claimable at the offshore offers; the two halves of the table are different markets, and the table’s job is to make the line between them visible row by row.
| Brand | Licence | Welcome range | Wagering | Min deposit | Payout window | Rails |
|---|---|---|---|---|---|---|
| Betfred | UKGC licence | Modest matched offer or spins | Light on spins | £5 to £10 | 1 to 3 days | Debit cards, PayPal, e-wallets |
| Ladbrokes | UKGC licence | Low-match or spins package | Light, often 10x to 30x | £5 to £10 | 1 to 3 days | Debit cards, PayPal, Apple Pay |
| Sky Vegas | UKGC licence | Spins-led welcome | Light on the spins | £5 to £10 | Same day to 2 days | Debit cards, PayPal, e-wallets |
| Virgin Games | UKGC licence | Spins with free-play framing | Light on spins | £10 | Same day to 2 days | Debit cards, PayPal |
| MrQ | UKGC licence | Spins with no wagering messaging | None on the spins offer | £10 | Same day to 2 days | Debit cards, PayPal |
| Videoslots | UKGC licence | Matched offer with spins | 20x to 40x typical | £10 | 1 to 2 days | Debit cards, e-wallets, Trustly |
| Casumo | UKGC licence | Matched offer with spins | 30x typical | £10 | 1 to 2 days | Debit cards, e-wallets |
| PartyCasino | UKGC licence | Matched offer or spins | 20x to 40x typical | £10 | 1 to 3 days | Debit cards, PayPal, e-wallets |
| Rolletto | Offshore licence — verify on register | Matched package with a high ceiling | 35x to 45x | £10 to £20 | 1 to 5 days | Cards, e-wallets, crypto |
| NineWin | Offshore licence — verify on register | Matched package with spins | 35x to 45x | £10 to £20 | 1 to 5 days | Cards, e-wallets, crypto |
| Fat Pirate | Offshore licence — verify on register | Matched package, theme-led | 35x to 50x | £10 to £20 | 1 to 5 days | Cards, e-wallets, crypto |
| Gamdom | Offshore licence — verify on register | Crypto-leaning matched offer | 35x to 50x | Lower on crypto rails | Hours to 2 days on crypto | Crypto, cards |
Note: the figures above 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 and none of them are claimable at the offshore offers. Rows are descriptive text, not links, and every figure should be verified on the operator’s own terms page before any deposit.
Read against this page’s subject, the table says three things. Wagering on the offshore rows sits at thirty-five to fifty times where several licensed rows publish lighter structures or none on spins, which changes the cost of the bonus section’s arithmetic by an order of magnitude. Minimum deposits and payout windows overlap between the two halves, with crypto at the offshore rows the fastest rail anywhere on the table. And the columns that would matter most to a slots player, the spin-speed protections, the limit tools and the dispute route, are not columns at all, because they are not terms an operator publishes; they are consequences of the licence, which the aside above has already costed. The game mathematics are the same on both sides of the table. The frame around them is not, and the stake button, the information page and the terms sheet are where the reader’s own decisions all live.
What return-to-player figure should I look for on a slot?
The bulk of modern five-reel games publish between 95 and 96.5 percent, and the Megaways family clusters near the top of that band. Anything at 96 percent or above is at the better end of the regular market; jackpot titles sit lower, commonly 92 to 94 percent, because a share of each stake funds the pool. The figure on the game’s information page in the lobby you are playing in is the one that applies.
How much does a session of online slots actually cost?
Expected cost per spin is the stake multiplied by one minus the return figure. At a £0.20 stake on a 96 percent game that is 0.8p a spin, about £3.20 an hour at autoplay pace. The same hour at £1 stakes on a 93 percent game expects to cost £28. The table in the fourth section runs the full grid.
Does a higher RTP mean I will win more tonight?
No. The return figure describes the average across enormous numbers of spins, far more than any session contains, so single evenings are dominated by variance. A higher RTP lowers the expected cost of the hour, which matters over months of play, but it predicts nothing about an individual evening on either side of the average.
What does volatility mean and why does it matter?
Volatility describes how a game distributes its return: low-volatility games pay small amounts often, high-volatility games pay rarely and large. It matters because the bankroll must fund enough spins to reach the game’s interesting moments. On a high-volatility title, 250 to 350 spins of depth is the honest minimum, which at a £30 bankroll argues for a 10p to 12p stake.
What is the maximum-bet clause on a bonus?
While bonus wagering is live, terms at this end of the market commonly cap the stake at £2 to £5 per spin, and exceeding the cap, often including through an autoplay batch or a feature buy, commonly voids the bonus and any winnings attached to it. Set the stake comfortably below the printed cap, reconfigure autoplay, and leave feature buys alone until the wagering completes.
Is it better to clear a bonus on slots or table games?
Slots almost always, because they contribute 100 percent to wagering where table games contribute a fraction or nothing, so a £2,000 schedule at slots is £2,000 of play and at blackjack can be £20,000 of play. Check the weighting list too, because the steadiest high-return slots are often excluded from full contribution.
Do autoplay and fast spin modes change the cost?
They change the speed, and the speed is the multiplier on the per-spin cost. Manual play runs 150 to 250 spins an hour and autoplay runs 400 to 600, so the same stake costs two to three times as much per hour under autoplay. Setting the batch count to match the session budget restores a checkpoint that fast modes otherwise remove.
Are slot RTP figures the same on licensed and offshore sites?
Broadly yes, because the same studios and titles appear in both, though operators configure games within permitted ranges so the same title can carry slightly different figures in different lobbies. What differs is the frame: licensed sites carry regulated spin-speed caps, enforced limits and a dispute route that do not travel with an offshore offer.
Gambling involves risk and there is no way to play it without it. Free, confidential help is available from BeGambleAware at begambleaware.org, from the National Gambling Helpline on 0808 8020 133, run by GamCare, and from Gamblers Anonymous, which runs local meetings across the country. GamStop at gamstop.co.uk is the free self-exclusion scheme covering all GB-licensed online operators, and the Gambling Commission’s public register at gamblingcommission.gov.uk lets you check any operator’s licence before you play. If gambling has stopped feeling like a choice, start with the helpline.