The four rails: BTC, ETH, USDT and LTC

A crypto casino is decided at the payment rail before it is decided at the reels. Four rails carry the great majority of British-facing crypto traffic: Bitcoin (BTC), Ethereum (ETH), Tether (USDT) and Litecoin (LTC). Each has its own fee level, its own confirmation clock and its own habits inside bonus terms, and the differences are wide enough to change what an offer is worth. Before comparing welcome banners, it pays to know which coin is moving your money, because the rail sets the floor, the delay and the fee before a single spin is counted.

Bitcoin is the oldest rail and the one every desk supports. Its base layer produces a block roughly every ten minutes, and a transfer is not safely finished until the network has confirmed it, which most desks require two or three times before crediting. On a quiet day the whole trip takes ten to twenty minutes; when the mempool fills, fees and waits stretch together. Fees are quoted per unit of data and set by auction, so a £100 deposit might carry £1 of fee at midnight and £5 at a weekend peak. Where a desk offers the Lightning extension, small amounts move in seconds for pennies, but Lightning support is the exception rather than the rule at this end of the market.

Ethereum is the programmable rail. Its blocks arrive about every twelve seconds, so a plain transfer of ETH clears in two to six minutes at a typical confirmation count, and the fee, called gas, is auctioned the same way as Bitcoin's. The important second act is tokens: Tether mostly travels as an ERC-20 token on this rail, and the gas for moving USDT on Ethereum costs what an ETH transfer costs, regardless of the token's own value. In busy spells that makes ERC-20 the most expensive way to move a small deposit in the whole four-rail set, and it is the one rail where the fee can approach the deposit itself.

Tether is the stablecoin, pegged to the United States dollar, and it behaves like dollars in an envelope rather than like an investment. It runs on several networks, and the two that matter here are Tron, where a TRC-20 transfer costs pennies and lands in one to three minutes, and Ethereum, where ERC-20 inherits the gas bill described above. Because the peg holds the value steady, USDT is the settlement currency of choice for desks that quote balances in coin: what you send is what arrives, with no coin-price lottery in the gap between deposit and credit. The peg itself is a market construct rather than a promise, but in ordinary weeks it holds within a penny or two of the dollar.

Litecoin is the utility rail. Its technology mirrors Bitcoin's, but blocks arrive every two and a half minutes and fees sit at pence, which makes it the cheapest full-size rail for a modest deposit. It draws less attention than the other three, and that is its charm: fewer queues, fewer spikes, a predictable few minutes from send to credit. Desks that quote a minimum deposit of £10 to £20 in coin often accept LTC as the low-friction route, and small cashouts leave it with more of the balance intact than any on-chain Bitcoin withdrawal would.

The rail decision lands in three places. The deposit floor, because fees do not scale down with the amount sent. The clock, because a session that starts with a forty-minute Bitcoin wait is a different shape from one that starts in two minutes. And the terms, because some deals of this shape attach the welcome offer to specific coins and quietly exclude the rest. The model table later on this page puts numbers on all three, and the sections in between explain where each number comes from.

Why offshore lobbies lead with crypto

Start with what a British licence does to payments. Since 14 April 2020, operators licensed by the Gambling Commission have been barred from taking credit cards for gambling. Crypto is not named in that ban, but the acquiring arrangements behind the licensed market rarely carry it, and the practical result is that a GB-licensed casino payment page reads as a short list: debit cards, a few wallets, bank transfer, sometimes pay-by-phone. The rails are familiar, the ceilings are modest, and every one of them runs through a banking partner with a say in the matter.

Offshore desks sit outside that system twice over. They hold no Gambling Commission licence, so the British payment rules do not bind them; and they settle, where they settle at all, through channels that never touch a British acquiring chain. Credit cards therefore appear in their rails, and crypto is usually the headline rail rather than a footnote. This is the licence gap expressed in plumbing. It is not a secret trick and not a loophole to be worked; it is simply what the absence of a British licence looks like at the deposit screen, and the same absence removes the protections on the other side of the transaction.

The economics explain the enthusiasm. A card deposit costs an operator an acquiring fee and carries chargeback risk; a bank transfer needs a banking partner willing to serve the sector; crypto settlement is final, cheap and needs nobody's permission. A payment desk at the offshore end of the market can move a player's deposit to its own wallet in minutes, with no reverse gear and no intermediary holding the funds in between. Some of that saving funds the larger headline offers; the player pays the rest of it back through wagering multiples, which at this end tend to start where the licensed market's stop.

For the reader, the trade is explicit. On one side: faster settlement, bigger banners, credit cards back on the menu, and a deposit that never asks a bank what it thinks. On the other: no Gamstop, no IBAS adjudication, no affordability checks, and a regulator with no reach over the desk holding the funds. Neither side of that list is hidden; both follow from the same missing licence. The honest way to price a crypto offer is to count both columns, the convenience and the missing protections, in the same sum, and to notice that no banner at the top of a lobby will do that counting for you.

There is also a quieter reason crypto leads the lobby: it travels where other rails cannot. A player whose debit card is declined by a gambling block, or whose bank applies its own filters to gambling payments, often finds that crypto is the only rail still working. The lobby presents that as a feature. It is worth reading it as a warning light too, because a rail that outlasts every block a player has set up is not neutral plumbing, and the moment to notice that is before the deposit, not after it.

Network fees and where they land

Network fees are set by the coin's own market, not by the desk, and they move hour by hour. On a normal day this site's model takes Bitcoin at £1 to £5 for a standard transfer, Ethereum gas at £2 to £10, Litecoin at pence, Tether on Tron at pennies to about £1 and Tether on Ethereum at the same gas price as ETH. Every one of those can spike: Bitcoin when the mempool fills, Ethereum when a popular token sale opens. A fee table is a snapshot, and yesterday's snapshot is not today's price, which is why everything on this page is written as a range rather than a figure.

On a deposit, the sender pays, and the fee comes off the top. Send £50 of Bitcoin with £3 attached and £47 arrives; whether the desk credits £47, rounds to its floor, or treats the shortfall against the minimum deposit as a failed deposit is printed in its own terms. That line is one of the few places where a desk's wording, not the network, decides the outcome, and it is worth reading before the first transfer rather than after it. Desks that credit the net amount are the norm; desks that bounce a deposit for arriving a pound short of the floor exist, and their terms usually say so in a sentence most readers skip.

On a withdrawal the boot is on the other foot. The desk is the sender, and it recovers the network's cost one of three ways: it deducts the fee from the withdrawal, it prints a flat per-coin withdrawal charge, or it absorbs the cost and markets free withdrawals. A flat Bitcoin withdrawal charge of £2 to £5 is common at this end of the market; on a £60 cashout that is up to 8% gone before the exchange spread is counted. Small cashouts are where withdrawal fees bite hardest, and they are also the cashouts most likely to be subject to a minimum withdrawal floor that sits above the deposit floor.

The percentage maths explains the floors. A £3 fee on a £20 deposit is 15% gone before a spin; on £200 it is 1.5%; on £2,000 it is noise. Desks know this, which is why crypto minimum deposits sit at £10 to £20 rather than at the £5 floors the licensed card market manages, and why the cheap rails, Litecoin and TRC-20 Tether, exist at all. Match the rail to the amount: the smaller the deposit, the cheaper the rail needs to be, and the more a flat withdrawal charge matters on the way back out.

Timing is the last lever. Fees clear at quiet hours for a fraction of peak prices, and some desks batch withdrawals to save themselves the same money, which is one reason a payout window reads in hours rather than minutes. None of this is concealed; it is all in the fee schedule the desk prints beside its cashier-free deposit screen. The player's real defences are reading that schedule and choosing the hour, and both cost nothing.

Confirmation times and the shape of a session

A crypto deposit is not finished when the send button is pressed. It finishes when the network has confirmed the transaction enough times for the desk's risk rules, and only then does the balance move. The confirmation clock is a property of the coin: Bitcoin produces a block about every ten minutes, Ethereum about every twelve seconds, Litecoin every two and a half minutes, and Tether inherits the clock of whichever network it rides. Most desks credit after one to two confirmations on the fast rails and two to three on Bitcoin, which is why the same £100 can take forty times longer on one rail than on another.

The chart below gathers the modelled medians into one picture. The spread is the point: the rails are not interchangeable, and the difference is measured in spins.

What the wait does to a session depends on when it lands. A deposit sent before sitting down arrives invisibly; one sent after the player has chosen a game turns into forty minutes of watching a mempool. Bonus expiry is the sharper edge: welcome terms usually start their clock at credit, not at send, so a seven-day window is only ever as long as the first deposit allows it to be. On the withdrawal side the same physics runs in reverse, with the desk's processing stacked on top of the network's own delay, which is why the printed window, not the marketing clock, is the number to price.

Large wins add a second layer. Bigger withdrawals attract more confirmations, batch cuts and, frequently, the first real identity check the account has seen, because verification that was light at signup tends to arrive when money leaves. That is not unique to crypto desks, but the offshore setting gives it less oversight: there is no UK regulator to complain to if a review drags, only the desk's own support channel and whatever complaints route its licence class provides. The payout window in the terms is the honest expectation; everything faster is advertising, and everything slower is the part the advertising leaves out.

Volatility class and the wrapped bonus maths

A bonus is not cash; it is cash wrapped in turnover, and the wrap has a measurable price. Take a game returning 96%, which is the typical published return of an online slot. Every £1 staked returns 96p on average, so after each full multiple of the bonus the surviving share of the wrapped money is multiplied by 0.96. After m multiples the surviving share is 0.96^m of the original bonus. That one line is the whole conversion model, and everything else on this page's costings is built from it.

The worked example: a £100 bonus at 40x asks £4,000 of turnover. The expected conversion, 0.96^40, comes to about £19.50 of the original £100. The other £80.50 is the average price of the wrap, the amount the wagering itself is expected to consume before the balance is withdrawable. It is an expectation across many sessions, not a prediction for one; a lucky night beats it and a normal night does not, and neither outcome changes the arithmetic that produced it.

The multiple does the damage. At 35x the model converts about £24.00 of a £100 bonus; at 45x it converts about £15.90. Each extra multiple costs roughly 80p of expected value per £100 wrapped, which is why a desk's move from 35x to 45x is a real price change even though the headline never moves. Crypto offers sit at the top of that range: wagering of 35x to 45x on the bonus is category-typical at the offshore end of the market, against the 30x to 40x more usual for a standard licensed match.

Two wordings stretch the ladder further. Wagering on deposit plus bonus doubles the base: 40x on a £100 deposit and a £100 bonus is £8,000 of turnover, and the same decay applies to the doubled sum. Game weighting stretches it in the other direction: when tables count 10% toward wagering, a 40x term on a £10 table stake is effectively 400x on that game. Max bet caps, commonly £5 a spin while wagering, set the pace, and a seven-day expiry sets the deadline. All four numbers belong to the same sentence, and an offer is only costed when all four are in it.

Volatility class is the last layer. The 0.96 model is an average, and averages are made of swings: a high-volatility slot carries the same 4% house edge and distributes it in rarer, larger movements, so a short session can land far from the mean in either direction. The model says nothing about which way one night goes. What it does say is what the wrap costs before variance is applied, and that is the number a reader needs when comparing a 35x offer with a 45x one at the same face value, or an offer in coin with one fixed in pounds.

Stability coins, growth coins and the terms that follow

Coins divide into two families, and the bonus terms follow the split. Floating coins, Bitcoin, Ethereum and Litecoin, carry their own price risk: the sterling value of a BTC-denominated balance moves with the market, and 5% in a day is an ordinary move rather than a shock. Pegged coins, Tether above all, hold to the dollar, so their sterling value drifts only with the exchange rate, a percent or two in a quiet week. Which family a desk credits decides whether the wagering arithmetic is fixed or floating, and the two are not the same offer even at the same headline number.

The terms come in two shapes. Some desks fix the bonus in sterling at the moment of deposit and settle any surplus in coin at the end; under that wording the conversion model in the previous section runs unchanged. Others credit the bonus in coin and let it float while the player clears 35x to 45x across several days. Floating wording stacks a second variable on the wrap: a coin that rises makes the wrap cheaper in sterling terms, and a coin that falls makes it dearer. Neither direction is controllable, and only one of the two wordings says plainly in the terms which one applies.

Withdrawal runs the same film backwards. A deposit made in BTC is normally withdrawn in BTC, at the network's speed and the desk's printed withdrawal fee, and turning it back into pounds adds the off-ramp spread, which runs from about half a percent to 2% depending on the route, plus whatever the receiving bank charges. That spread belongs in the same column as the network fees: it is a cost of the rail, not of the games, and it lands whether the session won or lost. Desks that quote balances in sterling but settle in coin usually print their own conversion rate, and it is worth checking that rate against the market rate before assuming the two are the same.

Matching coin to purpose is the practical summary. Tether on Tron for predictable sizing at the lowest fee; Bitcoin for availability and the deepest support when the amount is large enough to wear the fee; Litecoin for small deposits where every pound of fee is a percent of the total; Ethereum or ERC-20 Tether where the desk takes nothing else. The games do not care which rail funded them. The terms sometimes do: a few promotions in this bracket attach the welcome offer to named coins and exclude the rest, which is a one-line check worth making before the deposit rather than after it.

One more habit worth forming: price everything in pounds at the moment of decision. A bonus quoted as a fraction of a Bitcoin, a floor quoted in USDT and a withdrawal fee in LTC are three currencies pretending to be one page. Convert each to sterling at the desk's own published rate, write the numbers down in a column, and the comparison becomes arithmetic instead of advertising. Any offer that resists being converted into a single currency is telling the reader something about itself.

The model table: fees, floors and speed by rail

The table below gathers this page's model into one grid: six rails, with the fee a £100 deposit typically carries, the typical wait to a credited balance, the floor a desk tends to print, whether the rail usually counts toward bonus wagering, and how the rail's price behaves against sterling. The assumptions are the ones used throughout: typical confirmation counts, mid-range network conditions, and sterling equivalents at recent rates. Every figure is a modelled range, not a quote from any operator, and the desk's own schedule outranks all of it.

Figures are category-typical and modelled; a payment desk's own terms govern.

Read the grid as a matching exercise rather than a chart of winners. A £20 deposit on ERC-20 Tether can wear a quarter of its value in gas at the wrong hour; the same £20 on TRC-20 or Litecoin keeps almost all of it. A £500 deposit can afford Bitcoin's fee and buys the deepest support if a withdrawal ever needs chasing. The row that suits the deposit size is the row to use, and the fee schedule checked on the day of the deposit beats every figure printed here, including the ones in this sentence.

Crypto casino desks against the British baseline

The comparison below sets eight rows carrying a UK licence beside four rows carrying none, so the licence gap can be read in the same columns as the prices. The British rows are the baseline: debit-only card rails, wallets, floors around £10, payout windows counted in working days. The offshore rows are the crypto end of the market: coin rails, credit cards back on the list, faster windows and larger headline matches, paid for through the wagering and through everything the licence column no longer covers. Ranges are category-typical and compiled from operator-published terms at the time of writing, not quotes of any single promotion.

Figures are category-typical ranges compiled from operator-published terms at the time of writing, not quotes of any single promotion. Rows carrying a UK licence are the British baseline shown for comparison; none of them are claimable at the offshore offers this page discusses.

Three columns do the work. Licence: the top eight rows print UK licence, the bottom four print No UK licence, and everything Gamstop, IBAS and the Gambling Commission provide attaches only to the first group. Payout window: the offshore crypto rows promise hours where the licensed rows print working days, and the difference is real; so is the difference in what happens when a payout stalls. Payment rails: the licensed rows never list credit cards, because they may not; the offshore rows do, because nothing stops them. Welcome range and wagering move together: the bigger the headline, the higher the multiple, and the model in the previous section prices that trade in pounds.

What a UK licence would have bought you

A UK licence is a bundle of protections, and this section is what it would have bought you at a desk that does not hold one. The clearest is Gamstop, the national self-exclusion scheme: one free registration excludes a player from every GB-licensed gambling site at once. Offshore desks sit outside it. Their own exclusion tools exist and vary, but the national safety net has no mesh where they stand, and a self-exclusion that has to be rebuilt site by site is a weaker thing than one that holds everywhere by default.

Second is IBAS, the Independent Betting Adjudication Service, which adjudicates disputes between players and GB-licensed operators when the operator's own complaints process stalls. A complaint at an offshore desk has no equivalent British route. The recourse is the desk's own support process and whatever procedure its licence class provides, and if that ends unsatisfactorily there is no regulator above it with powers over the outcome. A payout window is only as enforceable as the forum behind it, and at a desk with no British licence the forum is the desk itself.

Third is the credit card ban. Since 14 April 2020, Gambling Commission licensees have been barred from accepting credit cards for gambling, a rule aimed squarely at betting with borrowed money. The offshore rows in the comparison above list credit cards among their rails, because no British rule reaches them. Whatever a reader thinks of the ban, it exists to close a specific harm channel, and a licence-free desk reopens that channel by default. That reopening is part of the price printed invisibly on every larger headline, and it belongs in the same sum as the wagering multiple.

Fourth is the duty and what it sits beside. Remote Gaming Duty takes 21% of a GB-licensed operator's online gaming profits, rising to 22% from April 2027. The licensed market carries that cost and still supplies the protections above; the offshore model skips the duty and the protections in the same stroke. A reader comparing a 35x offer with a 45x one is comparing prices inside one model. Comparing a licensed desk with an unlicensed one is comparing two different models, and only one of them pays for Gamstop, IBAS and a regulator with enforcement powers.

Fifth is affordability. British licence conditions require checks designed to catch harm while it is still small, and they are friction by design: a licensed account may ask about source of funds at thresholds that feel intrusive at the time. Offshore, the pattern tends to run the other way: light checks at signup, heavier document requests at withdrawal, when the balance is on the table and the player's leverage is at its lowest. None of this makes a licensed account frictionless. It makes the friction arrive early, in daylight, with a regulator standing behind it, and that difference is worth more than any single number in a comparison table.

Read the outputs the way they are built: expected values across many sessions, not predictions for tonight. If the conversion figure looks small against the headline, that is the model doing its job. The comparison that matters is not offer against offer but offer against its own turnover, and the calculator's job is to put both numbers on the same line where they can be seen together.