Methodology

Data source

Channel-level and category-level daily viewership aggregates from Streams Charts, an independent streaming-analytics provider covering Twitch, Kick, YouTube and other platforms. Streams Charts polls each platform's public API at a fixed interval and stores per-stream snapshots; our figures are derived from those aggregates.

This site currently uses Twitch and Kick only. YouTube has no comparable category taxonomy for casino content, so a YouTube figure would not be like-for-like and we do not publish one.

Reporting window

Unless a page says otherwise, all figures cover H1 2026 — 2026-01-01 to 2026-06-30 inclusive. Year-on-year comparisons use the same calendar window in 2025.

Which categories count as iGaming

"Casino and betting content" on this site means these platform categories, which together make up the 380.4M-hour figure:

Deliberately excluded: casino-management and casino-simulation video games (Casino Simulator, Grand Casino Simulator, Dungeons & Degenerate Gamblers and similar), party games that involve wagering as a mechanic, and gambling-adjacent minigames inside other titles. These are video games, not iGaming products, and including them would inflate the vertical by a few million hours of irrelevant content.

A known boundary problem: platform categories are chosen by the streamer. Some sponsored casino segments run under Just Chatting or IRL and are therefore invisible to a category-based count. Our figures are a floor for the vertical, not a ceiling.

Metric definitions

• Hours watched — average concurrent viewers × airtime, summed across every stream-day in the window. The primary ranking metric because it reflects attention actually delivered rather than historic popularity.
• Average concurrent viewers (avg CCV) — airtime-weighted: Σ(avg viewers × airtime) ÷ Σ(airtime). A simple mean across stream-days would over-weight short sessions.
• Peak viewers — the single highest concurrent-viewer reading recorded in the window, in the relevant categories.
• Airtime — minutes broadcast, converted to hours. Casino airtime counts only time spent in the categories listed above.
• Casino % — casino airtime ÷ total airtime for that channel in the window. Airtime-based rather than hours-watched-based, because it answers "how much of this channel is casino content" rather than "which of its content performed best".
• Peak-to-average ratio — peak viewers ÷ avg CCV. A volatility measure, not a quality measure.
• Followers — the last recorded follower count in the window. Followers gained is the sum of daily changes, so it can be negative.

How the ranking is built

Channels are ordered by hours watched in the listed categories over the window, across Twitch and Kick combined. A channel appears once, on the platform where it produced more casino hours; where a channel is active on both, the secondary platform is noted in the copy rather than given a second row. Platform-specific pages re-rank the same rows and keep the global rank number so the two views stay reconcilable.

Channels are not excluded for having a low Casino % — a variety streamer with 20% casino airtime is a legitimate part of this market, and the column exists so readers can filter mentally rather than have the decision made for them.

What we do not claim to know

Being explicit about this is the point of the page.

• Viewer geography. Neither platform exposes reliable audience-location data publicly. Broadcast language is our proxy and it is imperfect — Spanish-language audiences span two continents.
• Viewer authenticity. We derive metrics from platform-reported viewer counts. We can test whether those counts behave plausibly — variance, chat ratio, follower-growth alignment — but we cannot audit a platform's counting. Any audience-quality output is a flag, not a verdict.
• Sponsorship values. Deal terms are private. We benchmark from delivery and comparables and label estimates as estimates. We do not publish claimed rates.
• Which operator sponsored which stream. Stream titles rarely name the operator: across H1 2026 we found operator names in the titles of only a small fraction of casino streams. Sponsor attribution requires video or overlay analysis, not metadata, and we do not present title matches as a sponsorship census.
• Individual viewer identity or demographics. Not measured, not collected, not for sale.

Refreshes and corrections

The public dataset is refreshed twice a year, after each half-year closes, and the refresh date appears in the footer of every page. Client deliverables use rolling windows — typically the last 30 or 90 days — because a six-month-old average is not a useful basis for a live negotiation.

If you believe a figure here is wrong, tell us what and why through the contact form — say which page and which number. We correct errors and note substantive corrections on the affected page.

Independence

iGaming Watch is a measurement and research service. We do not operate gambling products, run affiliate links, or take revenue share from operators, affiliates or streamers. We have no commercial relationship with any channel covered on this site. Clients pay for data and analysis, and a paid engagement does not change a channel's position in a public ranking — the ranking is generated from the dataset, not curated.

There is one figure we will not produce: an estimate of how much money any individual streamer has won or lost. It is not measurable from viewership data, and publishing a guess would be exactly the kind of unsourced number this page exists to argue against.

Reproducing a figure

Take 78,000+ channels, filter to the categories in the table above, restrict to 2026-01-01–2026-06-30, group by channel, and sum hours watched. Any analytics platform with per-stream category data for Twitch and Kick should land within a small margin of our figures; where it does not, the likely cause is a different category basket. Ours is listed above precisely so that can be checked.