Cryptocurrency acceptance and licensing: An analysis of 2,554 casino brands across 24 markets
Nobody agrees on quite how big crypto gambling is. The estimates currently in circulation differ by a factor of eight, and every one of them is produced by a party with an interest in the answer. So we measured something else: not how much money moves, but how many brands are built to accept crypto, and which ones. The answer is far more orderly than the argument about size.
In short
Half of all casino brands accept cryptocurrency. In Sweden, Romania, Italy and the Netherlands, not one nationally licensed brand does. The industry didn't adopt crypto; it was repopulated.
What this study covers
We looked at the payment infrastructure at 2,554 online casino brands across 24 markets, recorded per brand per market. The question is always whether a brand offers a given method to players in a specific country, not whether it offers it somewhere in the world. We cross-reference that against licensing, brand launch year, corporate ownership and game supply to see if there are any notable trends.
All figures in the study are shares, rates and brand counts drawn from our own structured dataset. The study includes no monetary totals, traffic or deposit volumes, or named individual brands or operators. Our methodology, robustness tests and limitations are set out in full at the end, including findings that did not survive scrutiny.
Scale
50.2%
of casino brands accept at least one cryptocurrency
1,283 of 2,554 brands accept cryptocurrency. Bitcoin appears at 98% of them.
The dividing line
0%
of nationally licensed brands in Sweden, Romania, Italy and the Netherlands accept it
Not one of 319 locally licensed brands across those four markets. Against 19–92% of unlicensed brands in the same markets.
What actually changed
2%
of the market-wide rise in crypto comes from licensed brands adopting it
EEA-licensed brands went from 11.4% to 13.9% across 12 years. The rest is offshore brands and a shifting mix.
The real growth story
82.8%
of new offshore brands now accept the cryptocurrency Tether, up from 22%
Tether is the only stablecoin in our data. Stablecoin uptake is genuine adoption, not composition.
Why measure infrastructure instead of money
The public numbers on crypto gambling are irreconcilable. The most-cited estimate puts crypto-denominated gross gaming revenue at $81.4bn for 2024, which is a figure larger than most independent estimates of the entire global online gambling market. The best-known challenge, from a blockchain analytics firm tracking wallets directly, puts the same figure at around $10–11bn [7]. Forbes, reviewing the dispute, sided closer to the lower number [8]. Both camps have a stake in the answer: the higher estimate originates from work commissioned by an anti-gambling campaign, while the lower figure comes from a firm that consults for operators.
An eight-fold disagreement isn’t a measurement problem. Instead, it signals that the underlying quantity — money moved through wallets that nobody is obliged to disclose — isn’t reliably observable from the outside. So we stopped trying to measure it.
What is observable is the plumbing. Every casino brand publishes all of the payment methods it accepts, because it has to: a player needs to know this information before depositing. That makes payment infrastructure a rare thing in the online casino industry. We get to see an operator-published, externally verifiable, censusable fact. It tells us how much of the market is built for cryptocurrencies, and precisely where the boundary falls.
“We can’t tell you how much crypto is gambled, and neither can anyone else who has published a number. What we can tell you is which brands have wired it in, and that turns out to be almost perfectly predicted by one thing.”
Finding 1: Half the market accepts crypto, and the halves are not random
Of 2,554 brands with payment data, 1,283 (50.2%) accept at least one cryptocurrency. That figure alone would support the “crypto is taking over gambling” headline that trade coverage has run for three years. It is also the least informative number in this study.
Break it down by market and an inverse relationship appears immediately: the markets where crypto is most common are the ones where domestic bank rails are least common, and the reverse is also true. Australia sits at 90.6% crypto against 4.4% bank-ID. Sweden and Denmark, with their more ubiquitous domestic bank rails, sit at the opposite corner.
- Accepts cryptocurrency
- Accepts a Nordic bank-ID rail
Share of brands present in each market offering each rail type
Show data table Hide data table
| Market | Accept crypto | Accept a bank-ID rail |
|---|---|---|
| Australia | 90.6% | 4.4% |
| Germany | 68.4% | 23.3% |
| Czech Republic | 54.2% | 5.1% |
| Switzerland | 50.5% | 42.9% |
| Italy | 49.3% | 30.1% |
| Ireland | 49.1% | 37.1% |
| Finland | 45.9% | 59.2% |
| Canada | 45.9% | 36.4% |
| New Zealand | 43.4% | 19.7% |
| Austria | 42.6% | 52.2% |
| Norway | 41.3% | 47.8% |
| Estonia | 31.8% | 54.5% |
| Netherlands | 27.7% | 58.6% |
| Spain | 25.4% | 40.8% |
| Romania | 20.4% | 5.6% |
| United Kingdom | 11.4% | 49.2% |
| Sweden | 10.2% | 81.5% |
| Denmark | 7.8% | 65.7% |
Source: Kasinohai casino dataset, brand × market grain. Markets with fewer than 40 brands suppressed..
It’s tempting to read this as a map of national payment culture. For example, Swedes use Trustly and Swish, so crypto has no room to penetrate, while Australians have no equivalent, so crypto fills the gap. That reading is wrong, however, and the rest of this study explains why.
Finding 2: The dividing line is the licence, not the country
Market-level figures conflate two very different populations. In any regulated market, some brands hold that country’s national licence, while others serve the same players without one. If we split them, however, then the picture stops being a gradient and becomes a wall.
- Holds the local national licence
- No local licence
Show data table Hide data table
| Market | Locally licensed brands | Accept crypto | Brands without local licence | Accept crypto |
|---|---|---|---|---|
| Sweden | 195 | 0% | 216 | 19.4% |
| Romania | 84 | 0% | 24 | 91.7% |
| Denmark | 80 | 1.3% | 22 | 31.8% |
| Germany | 41 | 4.9% | 1195 | 70.6% |
| Spain | 36 | 2.8% | 35 | 48.6% |
| Italy | 26 | 0% | 47 | 76.6% |
| Netherlands | 14 | 0% | 242 | 29.3% |
Source: Kasinohai casino dataset. Crypto measured within the same market as the licence, at brand × market grain..
In Sweden, Romania, Italy and the Netherlands, zero nationally licensed brands accept crypto — across 319 brands. Denmark has only one, and Spain has only one. Germany has two, out of 41. Meanwhile, in those same markets, between 19.4% and 91.7% of the brands without a local licence do accept cryptocurrencies.
A finding this absolute invites suspicion, so we tried to break it. The most obvious explanation is that licensed brands simply list fewer payment methods. But we found that they do not — holding the number of listed providers constant, the gap barely moves, running at roughly 13% for licensed brands against 78% for unlicensed ones at equivalent documentation depth. The second explanation would be market composition, since Germany’s pool contains 1,195 unlicensed brands against 41 licensed ones. But the effect holds in Sweden, where the licensed population is larger.
The third test was the most revealing. If crypto acceptance were a property of the company, brands would carry it everywhere. Instead, 17 of the 369 nationally licensed brands in our data offer crypto in some other market while keeping it out of the one where they hold a licence. The number is small, but the behaviour is unambiguous: these are the same brands, running a crypto-free product where a regulator is watching and a crypto-enabled product where no one is.
“Zero is a very unusual number to find in a dataset of 2,500 brands. It doesn’t mean crypto is banned, and in most of these countries it isn’t, explicitly. It means every compliance department in Europe has independently reached the same conclusion about what a regulator would or would not tolerate.”
Finding 3: The industry did not adopt crypto. It was repopulated.
This is the finding we think matters most, and it inverts the standard narrative about crypto acceptance.
Group brands by launch year, and crypto acceptance climbs steeply across cohorts, from 30.6% of brands launched in 2014–16 to 70.9% of those launched since 2023. Read on its own, that is a textbook technology adoption curve, and it’s how the trend is almost always reported.
Now split the same cohorts by licence, and the EEA-licensed line does not move: 11.4% in 2014–16, 13.9% in 2023–26, wandering in between without direction. The offshore line rises from 53.6% to 86.6%. Neither line individually resembles the steep market-wide curve.
- Offshore or unlicensed brands
- All brands
- EEA-licensed brands
Share of each launch cohort accepting cryptocurrency
Show data table Hide data table
| Launch cohort | All brands | EEA-licensed | Offshore or unlicensed | Offshore share of cohort |
|---|---|---|---|---|
| 2014–16 (n=242) | 30.6% | 11.4% (n=132) | 53.6% (n=110) | 45.5% |
| 2017–19 (n=489) | 33.3% | 11% (n=263) | 59.3% (n=226) | 46.2% |
| 2020–22 (n=687) | 59.5% | 23% (n=244) | 79.7% (n=443) | 64.5% |
| 2023–26 (n=764) | 70.9% | 13.9% (n=165) | 86.6% (n=599) | 78.4% |
Source: Kasinohai casino dataset. Cohorts pooled into multi-year buckets because single-year EEA cells fall as low as n=10..
The resolution is in the final column of that table. Offshore-licensed brands made up 45.5% of brands launched in 2014–16 and 78.4% of those launched since 2023. The market did not change its mind about crypto. Instead, it changed its composition.
Breaking down the 40.3-point market-wide rise into its three sources makes the point clear. 18.9 points (47%) come from the shift in where new brands are licensed. 20.4 points (51%) come from offshore brands themselves adopting crypto more heavily. And 1 point (2% of the total) comes from EEA-licensed brands adopting it. 12 years of the most-discussed payment shift in the industry, and the licensed segment of the market contributed only one percentage point.
This is a Simpson’s paradox in its purest form, and it has a concrete consequence for policymakers. If crypto were spreading through gambling as a technology, you would regulate the technology, i.e., restrict the payment method and player behaviour changes. If, instead, the licensed segment never adopted it and growth is in the number of unlicensed brands, then crypto is not what changed. It marks what changed: where new brands choose to be licensed. In that reading, restricting crypto payments at licensed operators regulates the one population that was never using it in the first place.
Finding 4: Tether is the actual growth story
One measure breaks the composition pattern, and it should interest anti-money-laundering supervisors most.
Within the offshore segment alone — holding composition constant by construction — Tether acceptance went from 21.8% of the 2014–16 cohort to 82.8% of the 2023–26 cohort. That is real adoption by brands that already existed in the same regulatory category, not a mix effect. Among EEA-licensed brands, it also rose from 3% to 10.3%. This is a smaller shift, but one of the few crypto measures that moves at all inside the licensed segment.
There is a detail here worth stating plainly, and it surprised us: among the brands we catalogue, Tether is the only stablecoin. USD Coin does not appear at a single brand in our data. Neither does Solana, Binance Coin, Polygon or any of the other assets that dominate general crypto-market coverage. That is a statement about the payment methods brands publish and we record — not about the whole market: a casino can accept a coin it never lists, and our coverage of the offshore fringe is partial by construction, as the method sets out. What the data does show is that the advertised crypto stack is narrow and old (primarily Bitcoin, Ethereum, Litecoin, Tether and Dogecoin), and it has barely changed in composition even as crypto acceptance has spread.
- Volatile asset
- Stablecoin
- Niche or privacy asset
Share of crypto-accepting brands supporting each asset Scale runs to 100%.
Show data table Hide data table
| Asset | Share of crypto-accepting brands |
|---|---|
| Bitcoin | 98.2% |
| Ethereum | 83.6% |
| Litecoin | 81.3% |
| Tether (USDT) | 74.7% |
| Dogecoin | 59.2% |
| Ripple (XRP) | 52.5% |
| Tron | 49.8% |
| Cardano | 34.8% |
| Stellar | 5.8% |
| Monero | 2.7% |
Source: Kasinohai casino dataset; base = 1,283 brands accepting at least one crypto asset or gateway..
The stablecoin shift matters more than the headline crypto figure. That’s because it changes what the payment method is for. A player depositing Bitcoin is exposed to price movement between deposit and withdrawal, which makes it a poor medium for routine gambling but a good one for speculation. A player depositing Tether is not. Stablecoins turn crypto from a wager on the asset into a functioning parallel banking rail. Tether transactions settle in minutes, cross borders without a correspondent bank and leave no chargeback trail.
Monero, the privacy coin whose presence would most directly indicate deliberate anonymity engineering, appears at 2.7% of crypto-accepting brands. That is a small number, but it’s worth reporting because it undercuts the strongest version of the anonymity argument.
Finding 5: Crypto now outranks the domestic bank rail in much of Europe
We compared Bitcoin with each market’s leading domestic payment rail, using the same measure for both, i.e., the share of brands in that market that offer it.
- Bitcoin ahead of the domestic rail
- Domestic rail ahead of Bitcoin
How many times more brands offer the leading method than the trailing one Scale runs to 10x.
Show data table Hide data table
| Market and comparison | Ratio | Penetration |
|---|---|---|
| Germany — Bitcoin vs GiroPay | 5.9x | Bitcoin ahead: 67.1% vs 11.4% |
| Finland — Bitcoin vs Euteller | 4.1x | Bitcoin ahead: 44.6% vs 11.0% |
| Finland — Bitcoin vs Zimpler | 2x | Bitcoin ahead: 44.6% vs 22.0% |
| Finland — Bitcoin vs Trustly | 1.3x | Bitcoin ahead: 44.6% vs 34.0% |
| Netherlands — Bitcoin vs iDeal | 1.2x | Bitcoin ahead: 27.0% vs 23.4% |
| Canada — Interac vs Bitcoin | 1.5x | Rail ahead: 66.7% vs 44.0% |
| Sweden — Trustly vs Bitcoin | 7.4x | Rail ahead: 70.3% vs 9.5% |
| Denmark — Trustly vs Bitcoin | 9x | Rail ahead: 61.8% vs 6.9% |
Source: Kasinohai casino dataset, brand × market grain. Only rails with records in the dataset are compared..
Germany is the starkest case: 67.1% vs 11.4%. Nearly six times as many German-facing brands offer Bitcoin as offer GiroPay. Finland shows the same pattern across all three of its bank rails, including Trustly, the most widely supported conventional method in the Nordic market.
The countries where the domestic rail still dominates (Sweden, Denmark and Canada) are precisely the countries with the highest proportion of locally licensed brands. Once again, the apparent payments story resolves into a licensing story.
Across launch cohorts, the two move as mirror images: bank-ID rail acceptance fell from 62.4% in the 2014–16 cohort to 36.9% in the 2023–26 cohort, while crypto rose. But that says more about which brands are launching, not which brands are swapping one rail for another.
Finding 6: Crypto and bank rails are not opposites, and the licence only makes them look that way
The obvious hypothesis after five findings is that crypto and regulated bank rails are substitutes: a brand picks a lane. However, when we tested it, we found it to be a false assertion.
448 brands (17.5% of the dataset) run both crypto and a Nordic bank-ID rail, and 81.5% of those offer both in the same market. This means it is not an artefact of a brand behaving differently across countries.
More importantly, the association reverses when you control for licensing. Pooled across all brands, crypto and bank-ID rails are negatively associated, with an odds ratio of 0.34. This is what produces the mirror-image appearance in Figure 1. Within EEA-licensed brands alone, the odds ratio is 1.49 (positive). It stays positive under Malta (1.25), UK (1.67) and Estonian (4.47) regulators when taken separately.
Said another way, European licenced brands with the best conventional payment coverage are slightly more likely to also offer crypto, not less. There is no technical or commercial antagonism between the two rails. The appearance of one exists entirely because the licensing regime sorts brands into groups that happen to differ on both measures at once.
Finding 7: The crypto brand is a small, young, wide-catalogue operation
Our earlier Finnish study suggested that offshore operators often used large fleets of low-cost sites on shared infrastructure. We expected crypto brands to show the same pattern. In this sample, however, they didn’t. The difference is sufficiently clear that we have revised that expectation rather than treating crypto brands as another example of the broader offshore pattern.
Crypto-accepting brands sit in fleets averaging 5.9 brands per operator, against 10 for non-crypto brands. Measured by concentration, the crypto population is less consolidated than the non-crypto one. The decisive test is among the newest brands, where any age effect is removed. Taking brands launched since 2023, 91.9% run by single-brand operators accept crypto, versus 48.6% of those in fleets of ten or more.
Large fleets, in other words, still avoid crypto even when launching brand-new offshore sites. This is consistent with fleet operators protecting a licensed business elsewhere in the group. The same compliance logic visible in Finding 2, applied at the portfolio level.
Crypto acceptance among brands launched 2023–2026, by the size of the fleet they belong to Scale runs to 100%.
Show data table Hide data table
| Measure | Crypto-accepting | Non-crypto |
|---|---|---|
| Median launch year | 2022 | 2018 |
| Median game studios supported | 58 | 28 |
| Mean brands per operator | 5.9 | 10 |
| Median brands per operator | 3 | 5 |
| Operator concentration (Herfindahl, max 10,000) | 38 | 80 |
Source: Kasinohai casino dataset. Operators are counted, never named..
The game-catalogue gap is the widest structural difference we measured, giving us a median of 58 game studios at crypto brands against 28 at non-crypto ones. A licensed operator must contract each studio individually and give evidence of the relationship to its regulator. Whereas, an unlicensed casino can aggregate hundreds of studios through a single turnkey platform. That means the game catalogue is not a feature decision. Instead, it’s a fingerprint of how the business was created.
What we looked for, but didn’t find
We tested whether players behave differently at crypto-accepting brands, using 18 months of observed activity. We report the result as null because it is one, and because the opposite claim is widely made without evidence.
Pooled together, crypto brands appear to convert visitors around 30% worse than non-crypto brands. That gap is entirely market mix: crypto brands cluster in card-dominated markets, non-crypto brands in Nordic bank-ID markets that convert better for structural reasons. Standardising across markets erases it and marginally reverses it. Within individual markets, the direction is inconsistent: crypto brands convert substantially better in Germany, Canada, Australia and New Zealand, and worse in Finland and Sweden.
Deposit sizes are also larger at crypto brands cross-sectionally, but a balanced panel of brands observed in both an earlier and a later window kills the causal reading. Within the same brands over time, deposit sizes fell faster at crypto brands than at non-crypto brands. The premium reflects which brands are in each group, not the payment method. An apparent survival penalty for crypto brands proved equally fragile: it shrank to near-nothing after controlling for brand scale, and it failed to replicate in two other entry cohorts, one of which ran the other way.
Therefore, we make no claim that crypto changes how players gamble. Every difference we could find in player behaviour dissolved into selection once we controlled for it properly. That is a useful finding on its own: the case for regulating crypto in gambling rests on money-laundering and self-exclusion arguments, not on evidence that crypto players lose differently.
The regulatory position, and the gap in it
Europe has spent three years building crypto regulation, and it is worth being precise about what it does and doesn’t reach.
The Markets in Crypto-Assets Regulation is now fully applicable, with its transitional periods having ended on 1 July 2026, after which unauthorised crypto-asset service providers serving EU clients are required to wind down [2]. MiCA does not mention gambling, gaming, betting or casinos anywhere in its text [1]. No EU authority has published guidance on whether an operator accepting crypto deposits is itself performing a regulated crypto-asset service. That question is, as far as we can establish, genuinely left open.
The Anti-Money Laundering Regulation is where the real force lies, and its timing is remarkable for anyone following the Finnish reform. It applies from 10 July 2027, i.e., just nine days after Finland’s licensed market officially opens. It makes gambling operators obliged entities across the EU under one directly-applicable regulation, sets a €2,000 customer due diligence trigger, and, under Article 79, prohibits anonymous crypto-asset accounts and anonymity-enhancing coins [3].
That prohibition, however, binds credit institutions, financial institutions and crypto-asset service providers. On the flip side, it doesn’t bind gambling operators directly. Meanwhile, the EU travel rule has applied to crypto transfers since December 2024 with no de minimis threshold [4], which is a stricter standard than applies to equivalent fiat transfers.
National regulators have moved faster than Brussels, it seems. The Kansspelautoriteit (Ksa) cited cryptocurrency and anonymous payments among the breaches when it fined an unlicensed operator €24.8m in March 2026 — its record fine against an illegal operator [5]. The UK has moved in the opposite direction: in February 2026 the Gambling Commission began exploring whether to permit crypto at licensed operators, on the reasoning that prohibition is pushing players toward illegal sites [6].
Our data suggests the UK instinct is at least coherent. If licensed operators have independently reached zero crypto acceptance without being told to, then a formal ban changes nothing about licensed behaviour and everything about where the argument sits. The binding constraint on crypto gambling in Europe is not the payment rule. It is whether an operator wants a licence in the first place.
“Regulators keep asking whether to allow crypto. Our data says the licensed market answered that question years ago and moved on. The unanswered question is what to do about the growing share of brands that never applied for a licence in the first place, as no payment rule constrains them.”
What this means for Finland in 2027
Finland licenses its market on 1 July 2027 and inherits the AMLR nine days later. Finnish law will require verified customer identity and enforceable self-exclusion, and our earlier study of the Finnish market found that crypto onboarding and Finnish bank identification are close to mutually exclusive among the brands serving Finnish players today.
This study sharpens that finding. Finland currently sits at 45.9% crypto penetration, which is mid-table and higher than every regulated market in Europe. More brands offer Bitcoin to Finnish players than offer Zimpler, Euteller or Trustly. If Finland follows the pattern of every other licensed European market, its licensed segment will settle at or near zero crypto acceptance without any specific rule requiring it, and the crypto-accepting population will simply continue operating outside the licensed perimeter.
The policy question is, therefore, not whether to permit crypto at licensed Finnish operators. Based on evidence from seven other European markets, licensed operators will not use it regardless. The question is what proportion of Finnish players follow the brands that do allow crypto payments.
Methodology
The dataset is our structured record of casino brands and their payment, licensing, ownership and game-supply relationships, captured on 2026-07-30. The payment analysis covers 2,554 brands with at least one payment-method record across 24 markets, each with 40 or more brands.
Grain. Payment methods are recorded per brand per market. Every penetration figure in this study is computed at that grain, which gives us a meaningful question: whether or not a brand offers a method to players in a given country. This is not the same as asking whether a brand offers a method anywhere, and the two yield materially different answers. Measured at the brand level across all markets, thirteen Swedish-licensed brands appear to accept crypto, but measured within the Swedish market alone, none do. The difference is Finding 2’s dual-face behaviour, and getting this grain wrong is the single easiest way to misread this dataset.
Classification. “Crypto” means one of ten direct crypto assets present in the data (Bitcoin, Ethereum, Litecoin, Tether, Dogecoin, Ripple, Tron, Cardano, Stellar, Monero) or one of three crypto-native gateways. We derived the classification by enumerating every distinct payment provider in the dataset rather than pattern-matching names, which produces false positives (such as several conventional e-wallets carrying crypto-sounding names). Nordic bank-ID rails are Zimpler, Brite, Trustly, Euteller and Swish. “EEA-licensed” means a licence from a regulator inside the European Economic Area; the UK Gambling Commission is classified as non-EEA, reflecting the UK’s withdrawal from the EU in 2020.
Cohorts. The time dimension is brand launch year, not the date a payment record was created. We tested the latter and rejected it: record-creation timestamps track our editorial pipeline rather than operator behaviour, with a median gap of −143 days relative to the brand record itself. Any “crypto adoption over time” chart built on record dates would be an artefact of how we assembled the dataset. We pooled cohorts into multi-year buckets because single-year cells within the EEA-licensed segment fall as low as ten brands.
Robustness. We tested the licence finding against documentation depth, market composition, and same-brand cross-market behaviour. We tested the cohort finding by restricting to records refreshed since 2025 and to brands with recorded activity since 2024; both preserved the gradient. Behavioural comparisons restrict every ratio to rows reporting both of its inputs, because reporting completeness varies by field and dividing a densely reported measure by a sparsely reported one produces a meaningless number.
What we don’t publish. This dataset is commercially sensitive, and the study is an industry statistic, not a disclosure about our own business. Every figure is a share, rate, brand count or per-deposit average. We publish no monetary totals, traffic or deposit volumes, per-website breakdowns, or figures identifying an individual brand or operator.
Limitations
- This is a census of documented acceptance, not of usage. A brand listing Bitcoin may process very little of it, and a brand not listing it may well accept it on request. We measure what operators publish, which is a good proxy for how a business is built and a poor one for how much money moves.
- Coverage is not the whole internet. The dataset covers brands we track across our markets. Brands with no affiliate presence are systematically absent and disproportionately likely to be unlicensed and crypto-native. That means the true crypto share of the global market is higher than the 50.2% reported here, not lower.
- Licence records are point-in-time. A brand that recently gained or lost a national licence may be classified by its current status while its payment stack still reflects the previous one. This would blur the Finding 2 boundary rather than sharpen it, so the true separation is at least as sharp as reported.
- Small licensed cells. Several national licensed populations are small, with the Netherlands at 14 brands, Italy at 26, Spain at 36, and Germany at 41. The zero and near-zero results are consistent across all of them and across the much larger Swedish (195) and Romanian (84) populations, but individual market figures should not be read as precise.
- The decomposition is descriptive. Splitting the crypto rise into composition and adoption components describes what changed, but it doesn’t establish why operators choose offshore licensing in the first place. Crypto may be one motivation for that choice rather than merely a consequence, and our data cannot separate the two.
- Behavioural nulls are not proof of no effect. Finding no reliable difference in player behaviour after controlling for market and scale means our data cannot detect one, not that none exists. A dataset with player-level rather than brand-level granularity might.
Reuse
Journalists and researchers are welcome to reproduce any figure or chart from this study with attribution and a link to this page. As far as we can establish, no regulator, academic or named analyst house has published a methodologically transparent census of payment-method prevalence across online casino brands, so we would particularly welcome attempts to replicate or contradict these figures against other datasets. For the underlying aggregate tables, or for questions about method, contact the author.
References
- [1] Official Journal of the European Union. Regulation (EU) 2023/1114 on markets in crypto-assets (MiCA) (accessed 2026-07-30).
- [2] European Securities and Markets Authority. Public statement: unauthorised crypto-asset service providers must wind down as MiCA transitional periods end (accessed 2026-07-30).
- [3] Official Journal of the European Union. Regulation (EU) 2024/1624 (Anti-Money Laundering Regulation), Articles 3, 19 and 79 (accessed 2026-07-30).
- [4] Official Journal of the European Union. Regulation (EU) 2023/1113 on information accompanying transfers of funds and certain crypto-assets (accessed 2026-07-30).
- [5] NL Times. Dutch gambling regulator slaps record €24.8 million fine on illegal online casinos (accessed 2026-10-06).
- [6] CoinDesk. UK’s gambling watchdog explores allowing gamblers to pay bets with crypto (accessed 2026-07-30).
- [7] Tanzanite. The $81B myth debunked (accessed 2026-07-30).
- [8] Forbes. How stablecoins unleashed the $10 billion global crypto casino boom (accessed 2026-07-30).
- [9] International Center for Gaming Regulation, University of Nevada Las Vegas. An assessment of AML risks linked to accepting crypto-payments in the gaming sector (accessed 2026-07-30).
- [10] Forbes. Crypto casinos exploiting the UK’s gambling self-exclusion loopholes (accessed 2026-07-30).