Crypto casino win rates: 25.56% to 80.83% of bets paid out, depending on the feed
Feed research · 7-day windows ending Sep 23, 2026Every figure is payout ÷ stake for each bet on the 13 public feeds we can measure, out of the 19 crypto casinos we track. Pick a casino below to see what 100 of its bets did.

Crypto casino win rates by casino: pick one, see what 100 bets did
Each square is one bet in 100, scaled from the feed’s bets over the week and shaded from paid nothing to 10x or more. Next to the grid you see how rare a bigger hit was and the feed’s three biggest hits.
Gamdom
9,894,905 bets- Paid nothing (0x)49.4%
- Some back (under 1x)18.8%
- Stake back (1x–2x)8.8%
- 2x–10x16.8%
- 10x or more6.2%
Scaled from 9,894,905 bets over the week. The legend shows each exact share.
- reached 10x
- 1 in 16
- reached 100x
- 1 in 170
- reached 1,000x
- 1 in 7,800
- median hit multiplier
- 1.6x
- 5,148,297xMineson a $0.10–$1 bet
101,435.6xSan Quentin 2: Death Rowon a $100–$1,000 bet
15,428.55xFire in the Hole 3on a $1–$10 bet
Crypto casino win rates on every feed: 19 to 75 of every 100 bets paid nothing
All 13 feeds get the same 100 squares, sorted by bets that paid nothing. Each grid is one casino’s own week.
- Cloudbet
75 of 100 paid nothing
- BC.Game
73 of 100 paid nothing
- GOATED
68 of 100 paid nothing
- BetFury
56 of 100 paid nothing
- Thrill
51 of 100 paid nothing
Display-weighted - Gamba
51 of 100 paid nothing
- Shuffle
50 of 100 paid nothing
- Gamdom
49 of 100 paid nothing
- Shock
49 of 100 paid nothing
- Roobet
43 of 100 paid nothing
- Qzino
43 of 100 paid nothing
- Chips
43 of 100 paid nothing
Display-weighted - 500 Casino
19 of 100 paid nothing
Cashed-out rounds
Cloudbet has 75 squares that paid nothing, and 500 Casino has 19, from a board of cashed-out rounds. A lighter grid reflects what a board lists as much as how its games pay.
At Gamdom, 10x or more fills 6 of 100 squares, the most of any feed. At 7 of the 13 feeds it fills one square or none.
Crypto casino win rates at 100x: 1 bet in 170 at Gamdom, 1 in 18,300 at BetFury
Each cell is the share of a feed’s bets over the week that reached the multiple, written as 1 in N. Stronger color means more often, and gray means none.
| Casino | ≥2x | ≥10x | ≥100x | ≥1,000x |
|---|---|---|---|---|
| Gamdom | 1 in4.3 | 1 in16 | 1 in170 | 1 in7,800 |
| ThrillDisplay-weighted | 1 in4.9 | 1 in33 | 1 in590 | 1 in25,100 |
| Gamba | 1 in9 | 1 in43 | 1 in240 | 1 in3,900 |
| 500 CasinoCashed-out rounds | 1 in4.2 | 1 in44 | 1 in1,200 | 1 in26,800 |
| BC.Game | 1 in11 | 1 in51 | 1 in1,200 | 1 in145,300 |
| ChipsDisplay-weighted | 1 in9 | 1 in57 | 1 in750 | 1 in13,100 |
| GOATED | 1 in7.9 | 1 in85 | 1 in4,100 | 1 in286,600 |
| BetFury | 1 in7.6 | 1 in89 | 1 in18,300 | None |
| Cloudbet | 1 in7.6 | 1 in110 | 1 in5,900 | 1 in481,200 |
| Roobet | 1 in6.9 | 1 in110 | 1 in2,800 | 1 in146,600 |
| Shuffle | 1 in7.3 | 1 in110 | 1 in2,700 | 1 in267,900 |
| Qzino | 1 in11 | 1 in120 | 1 in3,000 | 1 in59,800 |
| Shock | 1 in13 | 1 in160 | 1 in2,400 | None |
Rare cells rest on few bets. At GOATED, the 1 in 286,600 that reached 1,000x is just 2 bets out of 573,155.
A cell counts past bets, not the chance that the next one reaches the multiple. The same multiple on two feeds comes from different games.
Crypto casino biggest multipliers: the 12 largest across 13 feeds, led by 5,148,297x on “Mines” at Gamdom
Each card is one bet, with its payout ÷ stake, the game as the feed names it and the stake range it fell in.
- No. 1
No. 2- No. 3
- No. 4
No. 5
No. 6
No. 7- No. 8
- No. 9
- No. 10
No. 11- No. 12
Gamdom holds seven of the twelve and shows the most bets, 9,894,905. More bets mean more chances at a big one.
Two of the twelve come from display-weighted feeds.
Crypto casino win rates by game: 38 titles on at least three feeds, casino by casino
Each line runs from the share of a title’s bets that returned the stake (filled dot) to the share that paid anything (open dot). House games like Keno and Dice are each casino’s own game under a shared name.
Plinko
13 feeds
- Shock1x · $0.80
- Gamdom0.2x · $30
- BetFury0.3x · $1.99
- Qzino0.2x · $0.06
- Shuffle0.2x · $8.00
- Gamba0.3x · $0.08
- GOATED0.2x · $0.95
- 500 CasinoCashed-out rounds0.2x · $2.86
- ChipsDisplay-weighted0.2x · $0.07
- Cloudbet1.68x · $0.50
- ThrillDisplay-weighted1x · $0.18
- BC.Game0.2x · $0.20
- Roobet0.2x · $40
- The gap between the dots is bets that paid back less than the stake, like a slot spin that returned part of the bet. Where the dots meet, every bet paid nothing or at least the stake.
- Titles match across feeds ignoring case, punctuation and spacing. The full cross-casino table, with the ≥10x share and payout ÷ stake, is in the Full data section.
Crypto casino win rates data: every table behind the charts
Hit rates, the multiplier ladder and bands, stake buckets, the biggest multipliers, every game, the feeds’ own multipliers and the withheld feeds, one row per casino.
Hit rate by feed, with payout ÷ stake
The hit rate runs from 25.56% at Cloudbet to 80.83% at 500 Casino. Every feed is listed highest first, with the share of bets that paid anything, the share that returned the stake and payout ÷ stake.
- A hit is a bet with a USD payout above zero, and a bet returned the stake at 1x or more. Basis: “Paired rows with a positive USD stake and a non-negative USD payout. The multiplier is recomputed as payout_usd / bet_usd.”
- The gap between the two shares is bets that paid back less than the stake, like a slot spin that returned part of the bet. It is widest at 500 Casino (39.43 pts) and narrowest at Cloudbet (0.00 pts). Cloudbet shows every bet as 0x or at least 1x, so its two shares match.
- The median hit multiplier runs from 1x at Roobet to 2x at Cloudbet. It is the middle multiplier among bets that paid anything, and under 1x most paying bets returned less than the stake.
- A hit rate tracks what a board lists as much as the games on it. 500 Casino’s third-busiest game, “Crash”, returned at least the stake on 99.93% of its bets. A board of cashed-out rounds or bonus settlements is a list of hits. Cloudbet’s busiest game, “Mines”, paid on 20.24% of its bets.
- Payout ÷ stake on the bets counted is 100 × their total payout over their total stake. Every share here comes from the same bets. On the whole feed it is captured payout over captured stake for the window, the ratio the rankings print. The two differ where the feed carries rows this count leaves out. A high hit rate can sit next to a ratio under 100%, because one counts how often bets paid and the other how much.
- AceBet, CoinCasino, DegenCity, Rakebit, Whale.io and Winna have no hit rate. The withheld block gives each reason.
How the whole-feed ratio is computed, on the payout ratio page
The multiplier ladder: shares at 2x, 10x, 100x and 1,000x
At Gamdom, 6.18% of bets reached 10x, against 0.62% at Shock. Each feed’s share is listed at 2x, 10x, 100x and 1,000x. A 100x bet also counts at 10x and 2x.
- Shares never rise from left to right. A 100x bet also counts at 10x and 2x, and every 2x bet paid something, so the hit rate is never below the 2x column.
- Each bar is scaled to the largest share in its column, not to 100%. Rows are ordered by the ≥10x column.
- Shares are rounded to 4 decimals of a fraction, so the smallest one printed is 0.01% and anything under 0.005% prints as 0.00%. A 0.00% at ≥1,000x means fewer than 1 bet in 20,000 reached 1,000x in the window. Of the 13 feeds, 11 showed a bet at 1,000x or more, and at 8 of them the share still prints as 0.00%. BetFury and Shock showed none.
Every bet in nine multiplier bands
Each feed’s bets split into nine bands by payout ÷ stake, from 0x (paid nothing) and under 1x (some back) up to 1,000x and more. Every bet sits in exactly one band.
- 0xpaid nothing
- 0x–1xsome back
- 1x–2xstake back
- 2x–5x
- 5x–10x
- 10x–50x
- 50x–100x
- 100x–1,000x
- ≥ 1,000x
Exact shares and bet counts for every band
- Each band includes its lower bound and excludes its upper one, so a bet at exactly 2x sits in 2x–5x. The 0x band is exactly zero. A feed’s nine shares add up to 100%.
- The 0x band is 100% minus the hit rate, and the 0x–1x band is the gap between paid anything and returned the stake. Add the bands from the top down and you get the ladder. Rails follow the hit-rate table’s order.
- The most common paying band is 0x–1x at 500 Casino, Roobet, Gamdom and BC.Game. It is 1x–2x at Chips, Qzino, Shock, Shuffle, Gamba, Thrill, BetFury, GOATED and Cloudbet.
- Shares show two decimals, so a band under 0.005% prints as 0.00%. Its bet count shows whether it is empty.
Hit rate and multiplier shares by stake bucket
The same shares split across seven stake buckets: paid anything, returned the stake, ≥2x, ≥10x and the biggest multiplier. A bucket with fewer than 1,000 bets is withheld.
500 Casino
Sep 16–23, 2026 · 535,615 bets- $1–$10 · 256,414 bets
- 21.99 pts
BC.Game
Sep 16–23, 2026 · 581,018 bets- $0.10–$1 · 368,943 bets
- 32.36 pts
BetFury
Sep 16–23, 2026 · 494,402 bets- $1–$10 · 464,648 bets
- 11.65 pts
Chips
Sep 16–23, 2026 · 875,421 bets- < $0.10 · 533,068 bets
- 40.41 pts
Cloudbet
Sep 16–23, 2026 · 2,405,834 bets- $0.10–$1 · 1,197,318 bets
- 9.24 pts
Gamba
Sep 16–23, 2026 · 699,842 bets- < $0.10 · 329,674 bets
- 47.28 pts
Gamdom
Sep 16–23, 2026 · 9,894,905 bets- $10–$100 · 5,453,934 bets
- 57.93 pts
GOATED
Sep 16–23, 2026 · 573,155 bets- $1–$10 · 228,135 bets
- 29.91 pts
Qzino
Sep 16–23, 2026 · 478,579 bets- < $0.10 · 216,589 bets
- 25.53 pts
Roobet
Sep 16–23, 2026 · 586,444 bets- $100–$1,000 · 297,549 bets
- 32.52 pts
Shock
Sep 16–23, 2026 · 120,120 bets- $1–$10 · 53,500 bets
- 31.73 pts
Shuffle
Sep 16–23, 2026 · 535,704 bets- $10–$100 · 205,525 bets
- 25.88 pts
Thrill
Sep 16–23, 2026 · 1,206,944 bets- $0.10–$1 · 517,424 bets
- 27.01 pts
- Buckets are the same seven USD stake ranges the bet sizes page uses, and each share is of the bets in that bucket. A bucket with fewer than 1,000 bets is withheld with its count, never estimated (23 cells across the 13 feeds).
- The hit rate can swing between buckets (57.93 pts at Gamdom, highest to lowest) without marking a trend. A bucket at 100% means every bet the feed showed at that stake paid something.
- A bucket’s biggest multiplier is one bet. The feed’s biggest multiplier is the top of these unless its bucket is withheld.
The biggest multiplier on each feed, and its top ten
The biggest multiplier on any feed is 5,148,297x, at Gamdom. Each row is one feed’s biggest bet by payout ÷ stake, with its game and stake bucket.
- Each multiplier is one bet’s payout ÷ stake, from the feed’s USD figures. It shows what the feed listed, not the most the game can pay.
- The stake bucket is the USD range of that bet’s stake. Its exact stake, player and time are not published here.
- Gamdom’s 5,148,297x and BetFury’s 710x are one bet each. More bets mean more chances at a big one, so each figure sits next to its feed’s bet count.
The ten biggest multipliers on each feed
Each feed’s ten bets with the biggest payout ÷ stake, largest first, with game and stake bucket. The first is the feed’s biggest multiplier above.
Gamdom5,148,297x · “Starlight Princess 1000” holds two of the ten
Gamba22,750x · “Dice” holds six of the ten
Chips20,000x · “Limbo” holds nine of the ten
500 Casino9,900x · “Dice” holds nine of the ten
Thrill9,900x · “Dice” holds three of the ten
Qzino9,500x · “limbo” holds all ten
Cloudbet8,767x · “Don't Get Cancelled” holds two of the ten
GOATED3,134.87x · “DICE” holds two of the ten
Shuffle2,292.4x · “Keno” holds four of the ten
Roobet2,152.8x · “Wanted Dead or a Wild” holds eight of the ten
BC.Game1,536x · “Wanted Dead or a Wild” holds two of the ten
Shock942.5x · “Keno” holds four of the ten
BetFury710x · “keno” holds seven of the ten
- “limbo” fills Qzino’s whole top ten. A repeat only points to a game with a high ceiling. All 13 feeds have a game that repeats in their top ten.
Game by game, and the titles listed at three or more casinos
Each feed’s games with at least 1,000 bets, busiest first, then every title listed at 3 or more casinos. Game names appear as each feed prints them.
500 CasinoSep 16–23, 2026 · 40 of 74 games with 1,000+ bets, busiest first · 5,198 games in the feed
BC.GameSep 16–23, 2026 · 40 of 107 games with 1,000+ bets, busiest first · 6,901 games in the feed
BetFurySep 16–23, 2026 · All 14 games with 1,000+ bets · 21 games in the feed
ChipsSep 16–23, 2026 · 40 of 41 games with 1,000+ bets, busiest first · 1,006 games in the feed
CloudbetSep 16–23, 2026 · 40 of 363 games with 1,000+ bets, busiest first · 1,241 games in the feed
GambaSep 16–23, 2026 · 40 of 41 games with 1,000+ bets, busiest first · 3,219 games in the feed
GamdomSep 16–23, 2026 · 40 of 595 games with 1,000+ bets, busiest first · 5,628 games in the feed
GOATEDSep 16–23, 2026 · 40 of 48 games with 1,000+ bets, busiest first · 790 games in the feed
QzinoSep 16–23, 2026 · All 7 games with 1,000+ bets · 7 games in the feed
RoobetSep 16–23, 2026 · 40 of 99 games with 1,000+ bets, busiest first · 2,525 games in the feed
ShockSep 16–23, 2026 · All 12 games with 1,000+ bets · 465 games in the feed
ShuffleSep 16–23, 2026 · 40 of 46 games with 1,000+ bets, busiest first · 3,176 games in the feed
ThrillSep 16–23, 2026 · 40 of 90 games with 1,000+ bets, busiest first · 2,028 games in the feed
Titles at 3 or more casinos
One row per title, one column per casino. Each cell holds that casino’s figures for the title: paid anything, ≥1x, ≥10x and payout ÷ stake. “Not listed” means the title is missing from that casino’s list, though the casino may still carry the game.
- Each list is a feed’s games with at least 1,000 bets in the window, busiest first, up to 40. BetFury, Qzino and Shock had fewer than 40 such games, so their lists are complete. The rest show their top 40.
- Basis: “Same rows as multipliers, grouped by game label.” A game’s figures cover its bets at that casino only.
- Titles match across casinos ignoring case, punctuation and spacing. So “Hilo”, “hilo”, “HiLo” and “HILO” share one row under the spelling most feeds use (the busiest on a tie), with the others listed. A renamed variant, such as a 1000 or Super Scatter edition, keeps its own row.
- A title’s range runs from its lowest to its highest figure among the casinos that list it.
The same games with each casino’s stake, payout and net, on the games page
Does each feed’s own multiplier match payout ÷ stake?
Most feeds print their own multiplier next to each bet. At 5 of 13 feeds it differs from payout ÷ stake by more than 1% on some bets, most often at BetFury (25.06%).
- Definition: “Rows whose source multiplier column differs from payout_usd / bet_usd by more than 1% of the larger of the two.” The share is of bets that carry a feed multiplier. Where some don’t, the casino’s row says how many do.
- A disagreement doesn’t mean the feed got its math wrong. Its multiplier can be the game’s figure before a fee or rounding, or on a different base than the USD stake. We record the gap without ruling on which figure is the game’s.
- At 8 of the 13 feeds the two agree on every bet: 500 Casino, BC.Game, Cloudbet, Gamdom, GOATED, Roobet, Shock and Thrill.
- Every multiplier and share here is payout ÷ stake as we compute it. A feed’s own multiplier never enters a figure.
Withheld feeds and why
No hit rate or multipliers for a feed that may show only a selection of bets, shows only bets that paid out, doesn’t link most payouts to a stake or counts money in units with no confirmed currency. Each feed’s reason is quoted below.
- Rule: “Money is measured only for USD-comparable feeds whose own snapshot does not describe the captured rows as a selection.” Each reason in the table is quoted from that feed’s snapshot.
- A payouts-only feed has no losses to count. Its hit rate would read 100% by design, and its multiplier shares would have nothing to divide by. A feed that may be a selection could produce both figures, but they would describe the selection.
- We still capture and count their rows. Only the money figures are withheld.
How we measure crypto casino win rates, and what a hit rate can’t tell you
Every rule below is checked at build time. A snapshot that breaks one stops the build before a number goes out.
Method
- Scope
- The 19 casinos in the current ranking scope, taken from the rankings, never listed by hand. Stopped collectors are out of scope. The latest snapshot is dated Sep 23, 2026, 21:27 UTC.
- Bets counted
- Every captured bet with both a USD stake and a USD payout. A bet missing either figure, or with a negative one, is left out, and the snapshot counts every exclusion. Basis: “Paired rows with a positive USD stake and a non-negative USD payout. The multiplier is recomputed as payout_usd / bet_usd.”
- Hit rate
- Bets with a USD payout above zero, divided by bets counted, per feed and window.
- Returned the stake
- Bets at 1x or more, divided by bets counted. The gap to the hit rate is bets that paid back less than the stake. The median hit multiplier is the middle multiplier among bets that paid anything. Payout ÷ stake is 100 × total payout over total stake on the same bets.
- Multiplier shares
- Bets that reached 2x, 10x, 100x or 1,000x, each divided by bets counted. A bet counts at every threshold it reached. Shares are rounded to 4 decimals of a fraction.
- Bands
- Every bet goes in one of nine multiplier bands, lower bound included, upper bound excluded. The 0x band is exactly zero. The build checks that a feed’s bands add up to its bets.
- Stake buckets
- The same shares and the biggest multiplier within each of seven USD stake buckets. A bucket with fewer than 1,000 bets is withheld, never estimated.
- Top multipliers
- Each feed’s ten bets with the biggest multipliers, with game label and stake bucket.
- Games
- Games with at least 1,000 bets, busiest first, up to 40 per feed. Titles match across casinos ignoring case, punctuation and spacing, and never merge with a variant. Basis: “Same rows as multipliers, grouped by game label.”
- Feed multiplier
- Counted where the feed prints its own multiplier: “Rows whose source multiplier column differs from payout_usd / bet_usd by more than 1% of the larger of the two.”
- Money
- Rule: “Money is measured only for USD-comparable feeds whose own snapshot does not describe the captured rows as a selection.” A withheld feed shows its snapshot’s reason and no figures.
- Comparison
- Shares are compared across feeds, never added, because each feed is a different display with its own rules. Where we count across feeds, we count casinos.
- Windows
- Each casino has its own capture window. The windows overlap but don’t line up exactly.
- Privacy
- No player name, alias or ID is read or shown.
Limits
- A bet feed is not the casino’s ledger. The casino decides what it shows, and a feed can leave bets out.
- A hit rate describes what a feed shows. A board of cashed-out rounds shows a hit on almost every bet, so the figure describes the board, not the game’s odds.
- A multiplier is one bet’s payout ÷ stake. The biggest one says nothing about a game’s math, and payout ÷ stake is not a game’s RTP.
- One week of one feed predicts nothing about the next bet. Every casino game carries a house edge.
- Two casinos’ Dice or Keno are different games with different math, and the same slot on two feeds follows two display rules.
- We didn’t place these bets, and our real-money tests are coming soon. Nothing here is a verdict on a casino.
Crypto casino win rates sources: 19 feeds, each in its own window
Below are every casino’s data date, capture window, money status and row count, with the checks each snapshot passed.
The checks each feed passed before its figures were written
- Casinos in the ranking scope: every committed snapshot whose capture window overlaps the freshest window, minus the manual ranking exclusions. Each window is re-resolved from the database and must equal the committed snapshot period.
- dataAsOf equals the window end
- captured rows = rows in window + excluded lane rows
- no lane exclusion applies
- activity grid cells sum to rows in window
- hour-profile games are ordered by rows
- minute rows sum to rows in window
- currency rows sum to labeled rows
- balance row shares sum to 1
- rows in window = money rows + null-stake rows + negative-stake rows
- money rows = paired rows + null-payout rows + negative-payout rows
- paired rows = multiplier rows + zero-stake paired rows
- stake bucket rows sum to money rows
- stake bucket totals sum to total stake
- stake bucket shares sum to 1
- activity grid stake sums to total stake
- stake percentiles are monotone and at most the max stake
- histogram rows sum to money rows
- largest-0.1% row count matches ceil(n × share)
- largest 0.1% of bets hold at least 0.1% of stake
- largest-1% row count matches ceil(n × share)
- largest 1% of bets hold at least 1% of stake
- largest-10% row count matches ceil(n × share)
- largest 10% of bets hold at least 10% of stake
- concentration shares grow with the share of bets
- stake-by-game percentiles are monotone and the games are ordered by rows
- multiplier query rows match paired positive-stake rows
- multiplier shares are monotone
- multiplier bands sum to multiplier rows
- the 0x band equals the rows that paid nothing
- ladder-by-bucket rows sum to multiplier rows
- top multipliers are ordered
- per-game ladders are monotone
- games meet the row floor and are ordered by rows
- game list length equals min(limit, games at floor)
- return bucket rows sum to multiplier rows
- quietest cell and empty-cell count exclude the shared capture gaps
- silent minutes exclude the collector pauses and count full minutes only
- dataAsOf equals the window end
- captured rows = rows in window + excluded lane rows
- no lane exclusion applies
- activity grid cells sum to rows in window
- hour-profile games are ordered by rows
- minute rows sum to rows in window
- currency rows sum to labeled rows
- balance row shares sum to 1
- quietest cell and empty-cell count exclude the shared capture gaps
- silent minutes exclude the collector pauses and count full minutes only
- dataAsOf equals the window end
- captured rows = rows in window + excluded lane rows
- activity grid cells sum to rows in window
- hour-profile games are ordered by rows
- minute rows sum to rows in window
- currency rows sum to labeled rows
- balance row shares sum to 1
- rows in window = money rows + null-stake rows + negative-stake rows
- money rows = paired rows + null-payout rows + negative-payout rows
- paired rows = multiplier rows + zero-stake paired rows
- stake bucket rows sum to money rows
- stake bucket totals sum to total stake
- stake bucket shares sum to 1
- activity grid stake sums to total stake
- stake percentiles are monotone and at most the max stake
- histogram rows sum to money rows
- largest-0.1% row count matches ceil(n × share)
- largest 0.1% of bets hold at least 0.1% of stake
- largest-1% row count matches ceil(n × share)
- largest 1% of bets hold at least 1% of stake
- largest-10% row count matches ceil(n × share)
- largest 10% of bets hold at least 10% of stake
- concentration shares grow with the share of bets
- stake-by-game percentiles are monotone and the games are ordered by rows
- multiplier query rows match paired positive-stake rows
- multiplier shares are monotone
- multiplier bands sum to multiplier rows
- the 0x band equals the rows that paid nothing
- ladder-by-bucket rows sum to multiplier rows
- top multipliers are ordered
- per-game ladders are monotone
- games meet the row floor and are ordered by rows
- game list length equals min(limit, games at floor)
- return bucket rows sum to multiplier rows
- quietest cell and empty-cell count exclude the shared capture gaps
- silent minutes exclude the collector pauses and count full minutes only
- dataAsOf equals the window end
- captured rows = rows in window + excluded lane rows
- no lane exclusion applies
- activity grid cells sum to rows in window
- hour-profile games are ordered by rows
- minute rows sum to rows in window
- currency rows sum to labeled rows
- balance row shares sum to 1
- rows in window = money rows + null-stake rows + negative-stake rows
- money rows = paired rows + null-payout rows + negative-payout rows
- paired rows = multiplier rows + zero-stake paired rows
- stake bucket rows sum to money rows
- stake bucket totals sum to total stake
- stake bucket shares sum to 1
- activity grid stake sums to total stake
- stake percentiles are monotone and at most the max stake
- histogram rows sum to money rows
- largest-0.1% row count matches ceil(n × share)
- largest 0.1% of bets hold at least 0.1% of stake
- largest-1% row count matches ceil(n × share)
- largest 1% of bets hold at least 1% of stake
- largest-10% row count matches ceil(n × share)
- largest 10% of bets hold at least 10% of stake
- concentration shares grow with the share of bets
- stake-by-game percentiles are monotone and the games are ordered by rows
- multiplier query rows match paired positive-stake rows
- multiplier shares are monotone
- multiplier bands sum to multiplier rows
- the 0x band equals the rows that paid nothing
- ladder-by-bucket rows sum to multiplier rows
- top multipliers are ordered
- per-game ladders are monotone
- games meet the row floor and are ordered by rows
- game list length equals min(limit, games at floor)
- return bucket rows sum to multiplier rows
- quietest cell and empty-cell count exclude the shared capture gaps
- silent minutes exclude the collector pauses and count full minutes only
- dataAsOf equals the window end
- captured rows = rows in window + excluded lane rows
- no lane exclusion applies
- activity grid cells sum to rows in window
- hour-profile games are ordered by rows
- minute rows sum to rows in window
- currency rows sum to labeled rows
- balance row shares sum to 1
- rows in window = money rows + null-stake rows + negative-stake rows
- money rows = paired rows + null-payout rows + negative-payout rows
- paired rows = multiplier rows + zero-stake paired rows
- stake bucket rows sum to money rows
- stake bucket totals sum to total stake
- stake bucket shares sum to 1
- activity grid stake sums to total stake
- stake percentiles are monotone and at most the max stake
- histogram rows sum to money rows
- largest-0.1% row count matches ceil(n × share)
- largest 0.1% of bets hold at least 0.1% of stake
- largest-1% row count matches ceil(n × share)
- largest 1% of bets hold at least 1% of stake
- largest-10% row count matches ceil(n × share)
- largest 10% of bets hold at least 10% of stake
- concentration shares grow with the share of bets
- stake-by-game percentiles are monotone and the games are ordered by rows
- multiplier query rows match paired positive-stake rows
- multiplier shares are monotone
- multiplier bands sum to multiplier rows
- the 0x band equals the rows that paid nothing
- ladder-by-bucket rows sum to multiplier rows
- top multipliers are ordered
- per-game ladders are monotone
- games meet the row floor and are ordered by rows
- game list length equals min(limit, games at floor)
- return bucket rows sum to multiplier rows
- quietest cell and empty-cell count exclude the shared capture gaps
- silent minutes exclude the collector pauses and count full minutes only
- dataAsOf equals the window end
- captured rows = rows in window + excluded lane rows
- no lane exclusion applies
- activity grid cells sum to rows in window
- hour-profile games are ordered by rows
- minute rows sum to rows in window
- currency rows sum to labeled rows
- balance row shares sum to 1
- rows in window = money rows + null-stake rows + negative-stake rows
- money rows = paired rows + null-payout rows + negative-payout rows
- paired rows = multiplier rows + zero-stake paired rows
- stake bucket rows sum to money rows
- stake bucket totals sum to total stake
- stake bucket shares sum to 1
- activity grid stake sums to total stake
- stake percentiles are monotone and at most the max stake
- histogram rows sum to money rows
- largest-0.1% row count matches ceil(n × share)
- largest 0.1% of bets hold at least 0.1% of stake
- largest-1% row count matches ceil(n × share)
- largest 1% of bets hold at least 1% of stake
- largest-10% row count matches ceil(n × share)
- largest 10% of bets hold at least 10% of stake
- concentration shares grow with the share of bets
- stake-by-game percentiles are monotone and the games are ordered by rows
- multiplier query rows match paired positive-stake rows
- multiplier shares are monotone
- multiplier bands sum to multiplier rows
- the 0x band equals the rows that paid nothing
- ladder-by-bucket rows sum to multiplier rows
- top multipliers are ordered
- per-game ladders are monotone
- games meet the row floor and are ordered by rows
- game list length equals min(limit, games at floor)
- return bucket rows sum to multiplier rows
- quietest cell and empty-cell count exclude the shared capture gaps
- silent minutes exclude the collector pauses and count full minutes only
- dataAsOf equals the window end
- captured rows = rows in window + excluded lane rows
- no lane exclusion applies
- activity grid cells sum to rows in window
- hour-profile games are ordered by rows
- minute rows sum to rows in window
- currency rows sum to labeled rows
- balance row shares sum to 1
- quietest cell and empty-cell count exclude the shared capture gaps
- silent minutes exclude the collector pauses and count full minutes only
- dataAsOf equals the window end
- captured rows = rows in window + excluded lane rows
- no lane exclusion applies
- activity grid cells sum to rows in window
- hour-profile games are ordered by rows
- minute rows sum to rows in window
- currency rows sum to labeled rows
- balance row shares sum to 1
- quietest cell and empty-cell count exclude the shared capture gaps
- silent minutes exclude the collector pauses and count full minutes only
- dataAsOf equals the window end
- captured rows = rows in window + excluded lane rows
- no lane exclusion applies
- activity grid cells sum to rows in window
- hour-profile games are ordered by rows
- minute rows sum to rows in window
- currency rows sum to labeled rows
- balance row shares sum to 1
- rows in window = money rows + null-stake rows + negative-stake rows
- money rows = paired rows + null-payout rows + negative-payout rows
- paired rows = multiplier rows + zero-stake paired rows
- stake bucket rows sum to money rows
- stake bucket totals sum to total stake
- stake bucket shares sum to 1
- activity grid stake sums to total stake
- stake percentiles are monotone and at most the max stake
- histogram rows sum to money rows
- largest-0.1% row count matches ceil(n × share)
- largest 0.1% of bets hold at least 0.1% of stake
- largest-1% row count matches ceil(n × share)
- largest 1% of bets hold at least 1% of stake
- largest-10% row count matches ceil(n × share)
- largest 10% of bets hold at least 10% of stake
- concentration shares grow with the share of bets
- stake-by-game percentiles are monotone and the games are ordered by rows
- multiplier query rows match paired positive-stake rows
- multiplier shares are monotone
- multiplier bands sum to multiplier rows
- the 0x band equals the rows that paid nothing
- ladder-by-bucket rows sum to multiplier rows
- top multipliers are ordered
- per-game ladders are monotone
- games meet the row floor and are ordered by rows
- game list length equals min(limit, games at floor)
- return bucket rows sum to multiplier rows
- quietest cell and empty-cell count exclude the shared capture gaps
- silent minutes exclude the collector pauses and count full minutes only
- dataAsOf equals the window end
- captured rows = rows in window + excluded lane rows
- no lane exclusion applies
- activity grid cells sum to rows in window
- hour-profile games are ordered by rows
- minute rows sum to rows in window
- currency rows sum to labeled rows
- balance row shares sum to 1
- rows in window = money rows + null-stake rows + negative-stake rows
- money rows = paired rows + null-payout rows + negative-payout rows
- paired rows = multiplier rows + zero-stake paired rows
- stake bucket rows sum to money rows
- stake bucket totals sum to total stake
- stake bucket shares sum to 1
- activity grid stake sums to total stake
- stake percentiles are monotone and at most the max stake
- histogram rows sum to money rows
- largest-0.1% row count matches ceil(n × share)
- largest 0.1% of bets hold at least 0.1% of stake
- largest-1% row count matches ceil(n × share)
- largest 1% of bets hold at least 1% of stake
- largest-10% row count matches ceil(n × share)
- largest 10% of bets hold at least 10% of stake
- concentration shares grow with the share of bets
- stake-by-game percentiles are monotone and the games are ordered by rows
- multiplier query rows match paired positive-stake rows
- multiplier shares are monotone
- multiplier bands sum to multiplier rows
- the 0x band equals the rows that paid nothing
- ladder-by-bucket rows sum to multiplier rows
- top multipliers are ordered
- per-game ladders are monotone
- games meet the row floor and are ordered by rows
- game list length equals min(limit, games at floor)
- return bucket rows sum to multiplier rows
- quietest cell and empty-cell count exclude the shared capture gaps
- silent minutes exclude the collector pauses and count full minutes only
- dataAsOf equals the window end
- captured rows = rows in window + excluded lane rows
- no lane exclusion applies
- activity grid cells sum to rows in window
- hour-profile games are ordered by rows
- minute rows sum to rows in window
- currency rows sum to labeled rows
- balance row shares sum to 1
- rows in window = money rows + null-stake rows + negative-stake rows
- money rows = paired rows + null-payout rows + negative-payout rows
- paired rows = multiplier rows + zero-stake paired rows
- stake bucket rows sum to money rows
- stake bucket totals sum to total stake
- stake bucket shares sum to 1
- activity grid stake sums to total stake
- stake percentiles are monotone and at most the max stake
- histogram rows sum to money rows
- largest-0.1% row count matches ceil(n × share)
- largest 0.1% of bets hold at least 0.1% of stake
- largest-1% row count matches ceil(n × share)
- largest 1% of bets hold at least 1% of stake
- largest-10% row count matches ceil(n × share)
- largest 10% of bets hold at least 10% of stake
- concentration shares grow with the share of bets
- stake-by-game percentiles are monotone and the games are ordered by rows
- multiplier query rows match paired positive-stake rows
- multiplier shares are monotone
- multiplier bands sum to multiplier rows
- the 0x band equals the rows that paid nothing
- ladder-by-bucket rows sum to multiplier rows
- top multipliers are ordered
- per-game ladders are monotone
- games meet the row floor and are ordered by rows
- game list length equals min(limit, games at floor)
- return bucket rows sum to multiplier rows
- quietest cell and empty-cell count exclude the shared capture gaps
- silent minutes exclude the collector pauses and count full minutes only
- dataAsOf equals the window end
- captured rows = rows in window + excluded lane rows
- no lane exclusion applies
- activity grid cells sum to rows in window
- hour-profile games are ordered by rows
- minute rows sum to rows in window
- currency rows sum to labeled rows
- balance row shares sum to 1
- rows in window = money rows + null-stake rows + negative-stake rows
- money rows = paired rows + null-payout rows + negative-payout rows
- paired rows = multiplier rows + zero-stake paired rows
- stake bucket rows sum to money rows
- stake bucket totals sum to total stake
- stake bucket shares sum to 1
- activity grid stake sums to total stake
- stake percentiles are monotone and at most the max stake
- histogram rows sum to money rows
- largest-0.1% row count matches ceil(n × share)
- largest 0.1% of bets hold at least 0.1% of stake
- largest-1% row count matches ceil(n × share)
- largest 1% of bets hold at least 1% of stake
- largest-10% row count matches ceil(n × share)
- largest 10% of bets hold at least 10% of stake
- concentration shares grow with the share of bets
- stake-by-game percentiles are monotone and the games are ordered by rows
- multiplier query rows match paired positive-stake rows
- multiplier shares are monotone
- multiplier bands sum to multiplier rows
- the 0x band equals the rows that paid nothing
- ladder-by-bucket rows sum to multiplier rows
- top multipliers are ordered
- per-game ladders are monotone
- games meet the row floor and are ordered by rows
- game list length equals min(limit, games at floor)
- return bucket rows sum to multiplier rows
- quietest cell and empty-cell count exclude the shared capture gaps
- silent minutes exclude the collector pauses and count full minutes only
- dataAsOf equals the window end
- captured rows = rows in window + excluded lane rows
- no lane exclusion applies
- activity grid cells sum to rows in window
- hour-profile games are ordered by rows
- minute rows sum to rows in window
- currency rows sum to labeled rows
- balance row shares sum to 1
- quietest cell and empty-cell count exclude the shared capture gaps
- silent minutes exclude the collector pauses and count full minutes only
- dataAsOf equals the window end
- captured rows = rows in window + excluded lane rows
- no lane exclusion applies
- activity grid cells sum to rows in window
- hour-profile games are ordered by rows
- minute rows sum to rows in window
- currency rows sum to labeled rows
- balance row shares sum to 1
- rows in window = money rows + null-stake rows + negative-stake rows
- money rows = paired rows + null-payout rows + negative-payout rows
- paired rows = multiplier rows + zero-stake paired rows
- stake bucket rows sum to money rows
- stake bucket totals sum to total stake
- stake bucket shares sum to 1
- activity grid stake sums to total stake
- stake percentiles are monotone and at most the max stake
- histogram rows sum to money rows
- largest-0.1% row count matches ceil(n × share)
- largest 0.1% of bets hold at least 0.1% of stake
- largest-1% row count matches ceil(n × share)
- largest 1% of bets hold at least 1% of stake
- largest-10% row count matches ceil(n × share)
- largest 10% of bets hold at least 10% of stake
- concentration shares grow with the share of bets
- stake-by-game percentiles are monotone and the games are ordered by rows
- multiplier query rows match paired positive-stake rows
- multiplier shares are monotone
- multiplier bands sum to multiplier rows
- the 0x band equals the rows that paid nothing
- ladder-by-bucket rows sum to multiplier rows
- top multipliers are ordered
- per-game ladders are monotone
- games meet the row floor and are ordered by rows
- game list length equals min(limit, games at floor)
- return bucket rows sum to multiplier rows
- quietest cell and empty-cell count exclude the shared capture gaps
- silent minutes exclude the collector pauses and count full minutes only
- dataAsOf equals the window end
- captured rows = rows in window + excluded lane rows
- no lane exclusion applies
- activity grid cells sum to rows in window
- hour-profile games are ordered by rows
- minute rows sum to rows in window
- currency rows sum to labeled rows
- balance row shares sum to 1
- rows in window = money rows + null-stake rows + negative-stake rows
- money rows = paired rows + null-payout rows + negative-payout rows
- paired rows = multiplier rows + zero-stake paired rows
- stake bucket rows sum to money rows
- stake bucket totals sum to total stake
- stake bucket shares sum to 1
- activity grid stake sums to total stake
- stake percentiles are monotone and at most the max stake
- histogram rows sum to money rows
- largest-0.1% row count matches ceil(n × share)
- largest 0.1% of bets hold at least 0.1% of stake
- largest-1% row count matches ceil(n × share)
- largest 1% of bets hold at least 1% of stake
- largest-10% row count matches ceil(n × share)
- largest 10% of bets hold at least 10% of stake
- concentration shares grow with the share of bets
- stake-by-game percentiles are monotone and the games are ordered by rows
- multiplier query rows match paired positive-stake rows
- multiplier shares are monotone
- multiplier bands sum to multiplier rows
- the 0x band equals the rows that paid nothing
- ladder-by-bucket rows sum to multiplier rows
- top multipliers are ordered
- per-game ladders are monotone
- games meet the row floor and are ordered by rows
- game list length equals min(limit, games at floor)
- return bucket rows sum to multiplier rows
- quietest cell and empty-cell count exclude the shared capture gaps
- silent minutes exclude the collector pauses and count full minutes only
- dataAsOf equals the window end
- captured rows = rows in window + excluded lane rows
- no lane exclusion applies
- activity grid cells sum to rows in window
- hour-profile games are ordered by rows
- minute rows sum to rows in window
- currency rows sum to labeled rows
- balance row shares sum to 1
- rows in window = money rows + null-stake rows + negative-stake rows
- money rows = paired rows + null-payout rows + negative-payout rows
- paired rows = multiplier rows + zero-stake paired rows
- stake bucket rows sum to money rows
- stake bucket totals sum to total stake
- stake bucket shares sum to 1
- activity grid stake sums to total stake
- stake percentiles are monotone and at most the max stake
- histogram rows sum to money rows
- largest-0.1% row count matches ceil(n × share)
- largest 0.1% of bets hold at least 0.1% of stake
- largest-1% row count matches ceil(n × share)
- largest 1% of bets hold at least 1% of stake
- largest-10% row count matches ceil(n × share)
- largest 10% of bets hold at least 10% of stake
- concentration shares grow with the share of bets
- stake-by-game percentiles are monotone and the games are ordered by rows
- multiplier query rows match paired positive-stake rows
- multiplier shares are monotone
- multiplier bands sum to multiplier rows
- the 0x band equals the rows that paid nothing
- ladder-by-bucket rows sum to multiplier rows
- top multipliers are ordered
- per-game ladders are monotone
- games meet the row floor and are ordered by rows
- game list length equals min(limit, games at floor)
- return bucket rows sum to multiplier rows
- quietest cell and empty-cell count exclude the shared capture gaps
- silent minutes exclude the collector pauses and count full minutes only
- dataAsOf equals the window end
- captured rows = rows in window + excluded lane rows
- no lane exclusion applies
- activity grid cells sum to rows in window
- hour-profile games are ordered by rows
- minute rows sum to rows in window
- currency rows sum to labeled rows
- balance row shares sum to 1
- rows in window = money rows + null-stake rows + negative-stake rows
- money rows = paired rows + null-payout rows + negative-payout rows
- paired rows = multiplier rows + zero-stake paired rows
- stake bucket rows sum to money rows
- stake bucket totals sum to total stake
- stake bucket shares sum to 1
- activity grid stake sums to total stake
- stake percentiles are monotone and at most the max stake
- histogram rows sum to money rows
- largest-0.1% row count matches ceil(n × share)
- largest 0.1% of bets hold at least 0.1% of stake
- largest-1% row count matches ceil(n × share)
- largest 1% of bets hold at least 1% of stake
- largest-10% row count matches ceil(n × share)
- largest 10% of bets hold at least 10% of stake
- concentration shares grow with the share of bets
- stake-by-game percentiles are monotone and the games are ordered by rows
- multiplier query rows match paired positive-stake rows
- multiplier shares are monotone
- multiplier bands sum to multiplier rows
- the 0x band equals the rows that paid nothing
- ladder-by-bucket rows sum to multiplier rows
- top multipliers are ordered
- per-game ladders are monotone
- games meet the row floor and are ordered by rows
- game list length equals min(limit, games at floor)
- return bucket rows sum to multiplier rows
- quietest cell and empty-cell count exclude the shared capture gaps
- silent minutes exclude the collector pauses and count full minutes only
- dataAsOf equals the window end
- captured rows = rows in window + excluded lane rows
- no lane exclusion applies
- activity grid cells sum to rows in window
- hour-profile games are ordered by rows
- minute rows sum to rows in window
- currency rows sum to labeled rows
- balance row shares sum to 1
- quietest cell and empty-cell count exclude the shared capture gaps
- silent minutes exclude the collector pauses and count full minutes only
- dataAsOf equals the window end
- captured rows = rows in window + excluded lane rows
- no lane exclusion applies
- activity grid cells sum to rows in window
- hour-profile games are ordered by rows
- minute rows sum to rows in window
- currency rows sum to labeled rows
- balance row shares sum to 1
- quietest cell and empty-cell count exclude the shared capture gaps
- silent minutes exclude the collector pauses and count full minutes only
More from the data
ResearchCrypto Casino Bet Sizes 2026: Median Stake at 13 FeedsAt BC.Game the top 1% of bets carry 81.5% of the money. Median and biggest bets, the bet-size mix and payout ÷ stake by bet size at 13 crypto casinos.
ResearchCrypto Casino RTP 2026: Observed Payout vs TheoreticalCrypto casino payout ratios measured on public bet feeds, casino by casino, and how a few big wins swing them. A feed ratio is not a game’s RTP.
ResearchWhat a Casino’s Live Wins Ticker Really ShowsEvery crypto casino scrolls a live wins ticker. We tracked 19 for a week: what each feed says it shows, what it leaves out and what a lobby counter counts.










