$31B in wagering volume. 286.6M bets captured. Here are the results.
In-house games, also known as casino originals, are games built and operated directly by casinos.
At Gamstat, we have captured $31B in wagering volume across 286.6M bets and here are some interesting statistics showing which casinos have delivered the highest observed RTP across individual games.
A quick reminder, all of these bets are captured directly from publicly available live casino feeds. We collect, analyse and make this data publicly accessible. These are observed results, not necessarily the theoretical RTP advertised by casinos.
1. Blackjack, leader: @Roobet
99.63% observed RTP, the highest among the tracked casinos.
gamstat.io/games/originals/r…
2. Plinko, 3 leaders: @Stake, @yeet & @Thrill_com
All three have recorded over 110% observed RTP, meaning players collectively received more than they wagered during the captured sample.
gamstat.io/games/originals/s…
gamstat.io/games/originals/y…
gamstat.io/games/originals/t…
3. Keno, leader: Duel
109.74% observed RTP. Some casinos have recorded even higher numbers but we have excluded them from this comparison due to known high roller bet placement affecting their live feeds.
gamstat.io/games/originals/d…
4. Mines, leader: @shufflecom
96.95% observed RTP. An interesting result considering how volatile this game can be, similarly to Keno, where individual betting patterns and large payouts can significantly influence the observed numbers.
gamstat.io/games/originals/s…
5. Limbo, leader: @rainbetcom
One of the closest results to the declared RTP we have observed.
1.1M rounds captured with a declared RTP of 99.00% against observed RTP of 99.08%. Just 0.08% difference.
gamstat.io/games/originals/r…
6. Crash, leader: Duel
120% observed RTP across more than $105M wagered.
gamstat.io/games/originals/d…
One important distinction: observed RTP above 100% does not mean a game has a positive expected return, it simply means players have collectively won more than they wagered within the captured sample.
We are continuously collecting data across casino live feeds, tracking millions of bets and comparing actual outcomes against advertised RTPs.
What other games should we look into next? Slot providers?
We have been measuring Gamstat’s success quite simply by looking at our daily active visitors and the growth has been massive.
Over the past months, we have put a lot of work into tracking streamers, their sessions, and RTP observations based on hundreds of millions of rounds to make our data as accurate as possible.
Now we are looking for feedback from the community.
What do you think is still missing on Gamstat? What features or improvements would you like to see next?
We are building this for the players. Every suggestion matters!
Update on how we collect data from Kick:
Gamstat scans Kick’s live feeds across Slots, Poker, and Sports Betting every 2 minutes, continuously discovering and tracking streamers automatically.
There is no manually maintained streamer list. If someone goes live in any of the tracked categories, they enter our dataset on the next scan.
Once discovered, their activity continues to be monitored and refreshed. Recently active streamers are checked every cycle, while inactive ones are periodically refreshed.
If someone disappears for more than 14 days, we stop polling them. The moment they return to one of the tracked categories, they are automatically picked up again.
Fully automated, directly from Kick.
1,334 streamers are now listed on gamstat.io/streamers
tag your favorite streamer listed on Gamstat in the comments, get a reply from them, and get a gift from me
prize amount in SOL depends on the streamer’s rank
Streamers? Custom profiles?
gamstat.io/streamers
Point 1 below is done.
Lots of new features are coming soon.
Good morning
We are building a full gambling streamer intelligence layer:
1. Streamer Tracker
— Track gambling streamers across Kick and Twitch in real time: live status, viewers, casino played, hours watched, stream history, and audience growth.
2. Verified Gambling Data
— Connect a streamer identity to their casino activity and verify what actually happens behind the stream: wagered volume, real profit and loss, average bet size, favorite games, biggest multipliers and more.
No more debating whether someone is playing with real or fake money. If their bets appear in the casino live feed and match what is shown on stream, they are verified. If not, they are marked as unverified.
We can also compare a streamer observed RTP against regular players playing the same game during the same period.
3. Measured Casino Impact
— Measure what actually happens to a casino while a streamer is live: changes in betting volume, player activity, new users, and how many continue playing afterward.
Instead of estimating how much a streamer is worth to a casino, we want to measure their actual impact.
4. On-Chain Payments
— Where attribution is possible, track verified blockchain transfers between casinos and streamers, replacing estimated sponsorship numbers with actual on-chain payments.
5. Transparency
— Show whether sponsorships are disclosed, whether casino provided funds are being used, whether the streamer casino identity is verified and which promo codes or affiliate relationships are connected to them.
The goal is simple: turn gambling streaming from an industry built around estimates and assumptions into something measurable and transparent.
Today’s casino rankings by volume:
1. @Stake
2. @Roobet
3. @bitstarz
4. @rainbetcom
5. @gamdom
6. @shufflecom
7. Due
8. @1win
9. @Bitcasinoio
10. @Cloudbet
And the most important update:
Now every single slot’s observed RTP is calculated only from casinos that display losing bets on their public feed.
Observed RTP accuracy should now be as close to perfect as possible.
For some slots with high winning activity, you will see the following mark:
NOT SETTLED
Real arithmetic over X captured rounds (e.g. 406,538) — but it has not yet settled on a value the game can consistently pay. Slot payouts are driven by rare large wins, so a single bonus round can significantly move the ratio. Read it as a sample, not as the game’s true return; it settles as more rounds are captured.
gamstat.io/providers/pragmat…
Major release:
Now every casino that doesn’t display the losing bets for the high rollers have players flagged, you can still fully analyze their bets and sessions,
Yellow mark will show:
* casino’s public bet feed publishes essentially no losing bets at all, so a profit here is a property of what the feed leaves out rather than of how the player did. There is no size at which this figure becomes a result.
Example:
gamstat.io/players?w=7d&b=wi…
















