Skip to content
Clash Watchdoglab count
Menu

Keno

Spribe Keno: pick numbers, 10 of 36 are drawn

Spribe logoSpribe· Self-play game

FairnessProvably fair
Keno cover art by Spribe
Illustration, not data

Specs

FairnessProvably fair
Lab statusMeasured
Provider
Spribe logoSpribe
Category
Self-play game
Provider claim
97%
Rounds counted
228,562 rounds
Last round recorded
17 Oct 2026, 10:17 UTC
Lab status
Measured
Rounds counted
228,562rounds
Provider claim
Provider claim
What is Provider claim?

Provider claim

The return the game provider states. We show it as a claim and compare it with what we count; we do not repeat it as fact.

97%

Provider source

Last round recorded
17 Oct 2026, 10:17 UTC

Past rounds describe behaviour. They do not predict the next round.

Short answer

No bot can predict Keno: the Keno game is designed so each round's result is independent of the last, and the lab will test this. Past results should therefore carry no information about the next one. The provider's stated Keno RTP, a long-run return, is 97%. On the headline line (Pick 5 numbers of 36, 10 drawn, stake 1 · USD demo), the return per round was 96.8% (95% CI 95.9% to 97.7%) across 228,562 rounds, as of 17 Oct 2026, 10:17 UTC.

Measured return

Return per round on the headline line (Pick 5 numbers of 36, 10 drawn, stake 1 · USD demo), against the provider claim.

Return per round
What is 95% interval?

95% interval

The range that would contain the true long-run value in 95 of 100 repeated samples of this size. It narrows as more rounds are counted.

96.8%

n
228,562
95% CI
95.9% to 97.7%
as of
17 Oct 2026, 10:17 UTC

Return per round · Pick 5 numbers of 36, 10 drawn, stake 1 · USD demo

Self-play lines

Each line is a fixed way of playing. It shows a figure once it passes 10,000 rounds, is up to date, and its 95% interval is within ±5 points.

Keno: self-play lines, each with its stake, rounds and status
SettingStakeRoundsStatusReturn per round
Pick 5 numbers of 36, 10 drawnstake 0.1 · USD demo (Stake below 1, see the note under the table)268,696 roundsMeasured

96.4%

95% CI 95.6%–97.2%

as of 17 Oct 2026, 10:17 UTC

Pick 5 numbers of 36, 10 drawnstake 1 · USD demo228,562 roundsMeasured

96.8%

95% CI 95.9%–97.7%

as of 17 Oct 2026, 10:17 UTC

At this stake Spribe rounds payouts down to 2 decimals, which lowers the long-run return slightly.
Where the rounds so far spread less than the paytable does, the interval uses the paytable's spread.

Myth checks

What players believe about Keno, what the game's design and maths say, and what the lab has counted so far.

  1. Can Keno be predicted?

    A bot, app or signal group can predict the next result.

    Not true

    No. Games are designed so each round is independent of the last; the lab will test this. Past results should therefore carry no information about the next one. Predictor apps show their hits and hide their misses; some are marketed alongside referral links.

    How the lab will test it Lag-1 autocorrelation of consecutive outcomes and the hit rate of any published 'signal' against recorded rounds.

    Read the full check

  2. Are there hot hours or patterns in Keno?

    Games pay more at certain hours (jam gacor) or follow readable patterns (pola).

    Not true

    No. The game does not know the time or your history. Any hour or pattern that looks hot is the normal clustering of random results; it should not repeat on fresh data, and the lab will test this.

    How the lab will test it Outcome distribution by hour-of-day compared with a chi-square test across hours.

    Read the full check

  3. Is Keno rigged?

    The game is rigged against players beyond its stated return.

    Open: measured here

    This is what the lab measures. Every game has a built-in house edge; the question is whether actual results match the provider's stated return. We publish measured return with its margin of error so you can see the answer, not take ours.

    Measured so far (return per round · Pick 5 numbers of 36, 10 drawn, stake 1 · USD demo): 96.8% (95% CI 95.9% to 97.7%) from 228,562 rounds as of 17 Oct 2026, 10:17 UTC, against 97% provider claim.

    How the lab tests it Measured RTP with a 95% confidence interval compared with the provider claim.

    Read the full check

Rules & math

Pick 1 to 10 numbers from 36; the game then draws 10. The payout multiplier depends on how many numbers you picked and how many of them were drawn; too few matches loses the stake.

Fairness

Provably fairThe provider publishes how a revealed round can be recomputed from its seeds.

Before play the provider publishes a hash of its server seed. Once the seed is revealed, our tool checks that it matches that commitment. Recomputing the round itself follows the provider's own published steps.

Check a hash commitment

How the badges are assigned

Questions

Can Keno be predicted?
No. Games are designed so each round is independent of the last, and the lab will test this. Past results should therefore carry no information about the next one. Apps that claim to show the next safe tile, path, dice roll or drawn numbers show their hits and hide their misses; some are marketed alongside referral links.
Is Keno rigged?
Keno has a built-in house edge, like every casino game. The real question is whether results match the provider's stated return. The lab measures exactly that: return with its 95% interval, compared with the provider claim. The headline line's measured return is published on this page with its 95% interval. Measured so far (return per round · Pick 5 numbers of 36, 10 drawn, stake 1 · USD demo): 96.8% (95% CI 95.9% to 97.7%) from 228,562 rounds as of 17 Oct 2026, 10:17 UTC, against 97% provider claim.
What is the RTP of Keno?
The provider claims 97% (source: https://spribe.co/games/keno). That is the provider's own long-run figure, not a lab measurement; this page labels it as a claim and compares it with what the lab counts.
Are there hot hours or patterns in Keno?
No. The game does not know the time or your history. Streaks and clusters appear in any random sequence, and they should not repeat on fresh data. The lab will check results by hour of day against the recorded rounds.

Data & citation

Daily fingerprints

Each UTC day of this game's raw records is sealed after 00:30 UTC the next day, and its row count and SHA-256 fingerprint are published here. The last 30 sealed days are listed.

Day (UTC)RowsSHA-256
48,331ac3712a3b44fb61dba0c99d31cda63d0b8f80ae1b312747dab36d68859d7ff67
36,936e6f37f702051e6abc53f5665bdccb75b528a59f5e840693a4fa9d2e2554f81d6
46,582fc4587ac95e65b706003b6a7face0762a75dd6678e0e75b96dba0b1b2dd2f816
47,804208bc994087654ce4f040de1b0e2146d98a87fadfd3e407cd9a6d74152e56439
47,693c99aeab6e66e306495e25ac769ab1ba1bf440185c024a7a2dd1e8b0e941ef5e1
48,269ddfdf9d4f3781116b97fe0de6b2a242c66d28ec4758945ac25734bfcb2690a16
48,063b122f70eaef622f5191da67a888cd263f56f5b588164a827dfa080da1d65d793
48,0056198c41cdfe4847a53412b8a22ba2148404117e52aa557f8a8473ac21c24147e
48,1120cef289d2d6be5fcf9561be09a2d6cce3a84dcd53b430b35ae983223067c5353
46,7770e71431922aaf5c1eba12baac82571b33e9db804ba02698bcea424ad900ea97f
9,8012ab6a2dd689dc2bf767d50b52d0f428d2c50a0b1433495edfc95f1072a8e951b

Raw per-round rows are not published for download. Raw rows are released on request to logged-in users through the data page (Telegram login), up to 5 requests in any 24 hours; a file sent for a day matches its published SHA-256. Raw data on request

Citation id
cw:keno:2026-10:v1
Cite as
Clash Watchdog (2026). Keno (Spribe): Return per round · Pick 5 numbers of 36, 10 drawn, stake 1 · USD demo: 96.8% (95% CI 95.9% to 97.7%), n = 228,562 rounds, as of 17 Oct 2026, 10:17 UTC. Methodology lab-v2.0 (site), crashlab-v1 (feed). cw:keno:2026-10:v1. https://clashwatchdog.com/games/kenoA self-play line's return per round is the return of that one fixed way of playing over this sample, not the game's overall RTP.
Licence
CC BY 4.0. Cite Clash Watchdog and link the dataset.
Methodology
lab-v2.0 (site), crashlab-v1 (feed). How we count