No regulator or supplier publishes a cross-operator leaderboard of live dealer streaming latency in milliseconds, so a literal “ranked by latency” list cannot be verified. What can be measured from public filings is table count, studio footprint and protocol disclosure — and on that basis, Evolution and Playtech remain the only two live casino suppliers large enough to independently verify, with Evolution operating roughly four times as many tables as its nearest rival.
Key takeaways
- No independent body audits comparative glass-to-glass latency across live casino suppliers, so marketing claims of “sub-second” or “fastest” streaming cannot be verified against a public benchmark.
- Evolution reported live casino revenue of approximately EUR 1,772.6 million for 2025 and ran about 2,000 live tables with dealers speaking 24 languages, giving it an estimated 62% of the live casino market by revenue.
- Playtech operated roughly 480 tables across 20 studios as of mid-2026, with live revenue up 8% year-on-year and live margin improving to 26.6%.
- Table variety and studio scale are measurable proxies for product depth; streaming latency, by contrast, is a vendor-asserted figure with no standardized third-party test.
- UK regulation treats live dealer fairness as a compliance floor, not a competitive differentiator — every licensed studio must meet the same auditability and surveillance standard.
Table of contents
Why latency can’t be ranked like a leaderboard
Every live casino supplier markets low latency, and every vendor of the underlying streaming technology makes similar promises. A patent filing describing low-latency media delivery for casino applications defines the category loosely:
“The low-latency application may be an application having latency in the range of 0-5 seconds, particularly in the range of 0-3 seconds and preferably in the range of 0-2 seconds.”
That is a wide band, and it is set by a patent applicant, not a regulator or an independent lab.
Commercial streaming vendors go further in their own marketing. One infrastructure provider advertises
“sub-250ms latency, ensuring fair play between in-person and online players while maintaining broadcast-quality streams across any device or network condition”
, contrasting this with
“traditional streaming solutions that can introduce 2-3 seconds of delay”
. That figure describes the vendor’s own infrastructure claim, not a measurement taken independently across competing live casino lobbies in production. No gambling regulator we reviewed — including the UK Gambling Commission — publishes comparative latency data for licensed live dealer studios. Given that gap, a genuine “ranked by streaming latency” list would have to invent numbers. Instead, this article uses the data points that are independently verifiable: table counts, studio footprints, revenue scale and regulatory disclosure, which function as reasonable proxies for production quality and infrastructure investment.
Market structure: what the filings actually show
Two of the three largest live casino suppliers are publicly listed and file audited results, which makes their numbers checkable in a way that private competitors’ claims are not. Evolution’s own interim filings show
net revenues of EUR 434,883 thousand for live games in the first quarter of 2026, against EUR 448,655 thousand a year earlier, with full-year 2025 live revenue at EUR 1,772,579 thousand
. Independent market analysis built on those filings estimates
Evolution earned EUR 1,772.6 million of live revenue in 2025, about 62% of the market by that analysis, with the top five studios holding about 84%
.
Playtech, the next largest listed supplier, reported in its H1 2026 results that
live revenue increased 8% year-on-year, with continued focus on table optimisation and cost efficiency initiatives improving margins, and c.480 Live tables globally
. Its investor presentation for the same period specifies
c.480 tables and 20 studios (including dual-play)
. A year earlier, Playtech’s full-year 2025 results stated that
total Live tables globally reached c.500 (from more than 450 in FY24) across 17 studios
— meaning the supplier actually trimmed table count between periods while improving margin, a sign of deliberate capacity management rather than runaway expansion.
Pragmatic Play, the third major supplier, does not publish standalone audited live casino financials because it is not a separately listed entity, which limits what can be independently verified about its table count or studio scale relative to the two listed suppliers above. That asymmetry matters for anyone trying to compare suppliers on hard numbers: two can be checked against regulatory filings, one cannot.
| Supplier | Latest reported live revenue | Approx. live tables | Approx. studios | Disclosure basis |
|---|---|---|---|---|
| Evolution | EUR 1,772.6 million (FY2025) | ~2,000 | Multiple, global | Audited annual/interim reports |
| Playtech | Up 8% YoY (H1 2026) | ~480 | 20 (incl. dual-play) | Audited interim results |
| Pragmatic Play | Not separately disclosed | Not independently verifiable | Not independently verifiable | Privately reported figures only |
The scale gap is directly relevant to table variety. A supplier running roughly 2,000 tables with
presenters speaking 24 languages at the end of 2025
can support far more side-bet variants, speed formats and regional game shows than one running a few hundred. Table count is not the same as streaming quality, but it is the clearest publicly disclosed proxy for lobby depth.
The technical latency budget
Where latency claims can be assessed at all, it is at the protocol level rather than the operator level. The industry has broadly settled on WebRTC for the player-facing video feed because it avoids the segment-based delay inherent to HTTP streaming. The same patent filing referenced above frames the entire category as operating in a
“range of 0-5 seconds, particularly in the range of 0-3 seconds and preferably in the range of 0-2 seconds”
— a reminder that even “low latency” in this industry spans a multi-second band depending on protocol choice, device, and network path, not a single fixed number any operator can claim across its entire user base.
This is also why dual-play tables — where an online feed streams the same physical wheel or shoe used by players on a land-based casino floor — are a meaningful structural detail rather than a marketing flourish. Playtech’s expansion with MGM Resorts included
the launch of 7 tables in H1 in the Live studio on the floor of the MGM Grand in Las Vegas
, a format that requires the online stream to stay synchronized with real-time physical play rather than a studio-only setup with no external timing constraint. Any latency claim tied to a dual-play table is testable in principle — a delayed online feed relative to the physical table would be visible to anyone watching both — which is not true of a studio-only stream where the “ground truth” timing is whatever the operator’s own systems record.
Table variety and studio footprint
Table variety breaks into two measurable dimensions: breadth of game types (blackjack, baccarat, roulette variants, game shows) and depth of localization (language-specific dealers, regional rule sets, market-specific side bets). Evolution’s scale advantage shows up clearly here: operating about 2,000 tables with dealers across two dozen languages supports a lobby structure that smaller suppliers cannot replicate purely through software updates — it requires physical studio capacity and trained staff in each market.
Playtech’s growth pattern illustrates the alternative strategy: fewer total tables, but targeted regional expansion. Its half-year 2026 results noted the business is
“well-positioned going forward following the completion of our São Paulo studio”
, and its US operation has scaled from a standing start to
more than 60 tables across 3 US studios
, serving over a dozen operators and more than 50 brands. That is table variety built through geographic targeting rather than raw volume — a different route to lobby depth that a pure table-count comparison would understate.
The trade-off between breadth and margin is visible in Playtech’s own numbers: total live tables actually fell slightly between periods — from
c.500 across 17 studios
at the end of 2025 to
c.480 tables and 20 studios
by mid-2026 — while studio count rose and margin improved. Fewer, better-utilized tables spread across more studios suggests a deliberate efficiency play rather than expansion for its own sake, which matters for players evaluating whether a growing lobby reflects genuine new content or table consolidation under a different label.
Regulation as a floor, not a differentiator
In Great Britain, live dealer fairness is governed by a specific technical standard rather than left to operator discretion. The Gambling Commission’s remote technical standards require that
“Live dealer operations must be fair and independently auditable”
, with implementation guidance specifying that
equipment and consumables should be of commercial casino quality, and croupiers need to undergo adequate training to provide the gambling in a fair way according to documented procedures and game rules
. The same standard mandates that
video surveillance to record all dealer activity should be in place, enough to cover the predefined gaming areas with sufficient detail to confirm whether dealing procedures and game rules were followed
.
This means every GB-licensed live studio, regardless of supplier, operates under the same baseline fairness and surveillance requirement — it is a floor every licensee must clear, not a basis for ranking one supplier above another on trustworthiness. It is a reasonable explanation for why none of the major suppliers publish comparative fairness or latency metrics: the regulatory framework asks for auditability on demand, not public benchmarking against competitors. For related context on how the overall UK market is structured around this remote casino category, see the Gambling Commission’s own industry statistics, which show
total Remote Casino GGY equating to 69.9 percent of Total RCBB GGY for the quarter to September 2025
— a figure that covers RNG and live tables together, since the regulator does not break out live dealer revenue as its own public category.
The lack of a separate live dealer reporting line in regulatory statistics is itself informative: it means independent, regulator-sourced data on live dealer-specific metrics (including latency) simply does not exist in the public domain, in the UK or in most other licensing jurisdictions we reviewed.
How GamblScout scores live dealer lobbies
Given the absence of a verifiable cross-operator latency metric, our algorithm weights live dealer lobbies on signals that can actually be checked: scraped table counts and game-type variety per operator lobby, the breadth of language and regional variants on offer, supplier disclosure quality (audited filings versus unverified press claims), and sentiment extracted from player discussion of buffering, freezing or disconnects during live sessions. That sentiment layer draws on the same natural language processing approach we apply elsewhere, including in our chat sentiment analysis of bingo sites, where live chat responsiveness is scored using comparable text-mining methods. For a genre where real-time interaction is similarly central to the player experience, our review of crash games like Aviator and Spaceman applies the same scepticism toward unverified “instant” or “real-time” marketing claims.
We treat vendor-claimed millisecond figures as marketing inputs rather than scored data points unless they come from an audited or regulator-published source. This is consistent with our broader stance, explained on our core principles hub, that algorithmic scoring should rely on data that can be re-scraped and re-checked rather than operator-supplied superlatives. The underlying scraping infrastructure that collects lobby table counts and studio metadata across operators is described in more detail on our data scraping and technical engine page, and the relative weighting we apply to disclosure quality versus table variety is set out on our scoring system and algorithmic weights page.
For players comparing genres rather than individual live tables, it is worth noting that live dealer remains a smaller slice of overall casino activity than slot content in most mature markets — a pattern explored in our analysis of why slots still dominate European casino revenues. Live dealer’s smaller revenue share does not make its technical claims less important to verify — if anything, the higher stakes typically wagered at live tables raise the cost of an unverified latency or fairness claim.
Frequently asked questions
What counts as “good” latency for a live dealer table?
There is no single regulator-set threshold. Industry and vendor sources describe low-latency live streaming as falling anywhere in a
range of 0-5 seconds, particularly 0-3 seconds and preferably 0-2 seconds
, which is a wide band. Treat any supplier’s specific millisecond claim as a marketing figure unless it is tied to an independently auditable source, such as a dual-play table synced to a visible physical game.
Why isn’t there an independent latency benchmark across operators?
No gambling regulator we reviewed, including the UK Gambling Commission, publishes comparative live dealer latency data; its technical standards focus on fairness and auditability rather than streaming speed metrics. Streaming latency claims therefore originate almost entirely from the suppliers and their technology vendors, not from third-party testing.
Does a bigger table count mean a better live dealer lobby?
Not necessarily. Evolution’s roughly 2,000 tables and Playtech’s roughly 480 tables across 20 studios represent different strategies — one maximizing breadth and language coverage, the other optimizing table utilization and regional targeting while margins improve. Variety and production quality are not strictly proportional to raw table count.
Are live dealer games riskier than other online casino formats?
Peer-reviewed research on online gamblers found that
problem gambling and over-indebtedness were markedly higher among online casino gamblers, with online casino gambling combined with live sports betting associated with remarkably increased risk
. This doesn’t single out live dealer tables specifically, but it supports treating real-time, fast-paced formats with the same caution recommended for any rapid betting cycle.
Methodology note
This article’s structural data was pulled from audited supplier filings (Evolution and Playtech interim and annual reports) and UK Gambling Commission technical standards and industry statistics rather than operator marketing pages. Our algorithm cross-references scraped lobby data — table counts, game types, and language variants — against these filings, and layers in NLP-derived player sentiment on streaming reliability, as described on our NLP and sentiment analysis hub, to flag discrepancies between advertised performance and reported player experience.
Gambling involves risk. Only play with money you can afford to lose and use the deposit limits and self-exclusion tools available in your jurisdiction.
