
Plinko casino guide: how Stake Plinko really pays
Plinko is the simplest game on Stake and the one people ask about most. We read the full paytable for every risk level, recomputed the 99% RTP from the binomial distribution behind the pegs and show where the money actually sits. This Plinko casino guide covers the odds, the provably fair check and the strategies that do and do not change anything.
What a Plinko casino game is
A ball, pegs and a row of multipliers
Plinko is the drop balls game from The Price Is Right with a paytable bolted on: a ball falls through a triangle of pegs, bouncing left or right at every row, and lands in a slot whose multiplier times your bet is what comes back. The centre slots, where most balls end up, pay less than your stake; the edges, where almost none do, pay the big numbers.
Why Stake’s is the reference version
Stake’s Plinko is a Stake Original: built in-house, provably fair and playable only on Stake. The Plinko game panel shows an RTP of 99.00%, a house edge of 1%; a typical slot keeps three to five times as much. On 5 October 2026 the game page listed four risk levels (Low, Medium, High, Expert), 8 to 16 rows, a 10,000× ceiling, Autobet with a set ball count, a spacebar hotkey and Instant Bet (no animation). The other Originals: games and Originals review.
Risk and rows: what the settings change
Risk reshapes, rows stretch
Risk redistributes the same 99%: a higher level takes money out of the middle and piles it onto the edges, so centre multipliers drop below 1× and edge multipliers climb. Rows add pegs: more slots, a bigger ceiling and, since each extra row is one more coin flip the ball must win to reach the edge, a far rarer top hit. Neither touches the house edge; Low with 8 rows and Expert with 16 both return 99 cents per dollar and differ only in how those cents are spread.
The multiplier rows on Stake
Game board, 5 October 2026, left to right; symmetric.
| Risk | Rows | Multipliers |
|---|---|---|
| Low | 8 | 5.6, 2.1, 1.1, 1, 0.5, 1, 1.1, 2.1, 5.6 |
| Low | 12 | 10, 3, 1.6, 1.4, 1.1, 1, 0.5, 1, 1.1, 1.4, 1.6, 3, 10 |
| Low | 16 | 16, 9, 2, 1.4, 1.4, 1.2, 1.1, 1, 0.5, 1, 1.1, 1.2, 1.4, 1.4, 2, 9, 16 |
| Medium | 8 | 13, 3, 1.3, 0.7, 0.4, 0.7, 1.3, 3, 13 |
| Medium | 12 | 33, 11, 4, 2, 1.1, 0.6, 0.3, 0.6, 1.1, 2, 4, 11, 33 |
| Medium | 16 | 110, 41, 10, 5, 3, 1.5, 1, 0.5, 0.3, 0.5, 1, 1.5, 3, 5, 10, 41, 110 |
| High | 8 | 29, 4, 1.5, 0.3, 0.2, 0.3, 1.5, 4, 29 |
| High | 12 | 170, 24, 8.1, 2, 0.7, 0.2, 0.2, 0.2, 0.7, 2, 8.1, 24, 170 |
| High | 16 | 1,000, 130, 26, 9, 4, 2, 0.2, 0.2, 0.2, 0.2, 0.2, 2, 4, 9, 26, 130, 1,000 |
| Expert | 8 | 50, 4.6, 1.1, 0.1, 0.1, 0.1, 1.1, 4.6, 50 |
| Expert | 12 | 619, 30, 6, 1.5, 0.4, 0.1, 0.1, 0.1, 0.4, 1.5, 6, 30, 619 |
| Expert | 16 | 10,000, 216, 26, 7, 2.5, 1.1, 0.1, 0.1, 0.1, 0.1, 0.1, 1.1, 2.5, 7, 26, 216, 10,000 |
Top multiplier by row count
| Rows | Low | Medium | High | Expert |
|---|---|---|---|---|
| 8 | 5.6× | 13× | 29× | 50× |
| 9 | 5.6× | 18× | 43× | 100× |
| 10 | 8.9× | 22× | 76× | 201× |
| 11 | 8.4× | 24× | 120× | 324× |
| 12 | 10× | 33× | 170× | 619× |
| 13 | 8.1× | 43× | 260× | 1,012× |
| 14 | 7.1× | 58× | 420× | 2,369× |
| 15 | 15× | 88× | 620× | 5,000× |
| 16 | 16× | 110× | 1,000× | 10,000× |
Low does not rise in a straight line (8.9× at 10 rows, 7.1× at 14): Stake rebalances the middle at each step. Expert is the fourth level, added since our September notes: floor 0.1×, so most balls lose 90% of the bet, ceiling ten times High’s.
Claim the deal with code PACTBONUS
Plinko RTP, expected value and variance
Why the centre pays under 1×
Every peg is a coin flip, so after 16 rows the slot is simply the number of times the ball went right, 0 to 16. The paths to slot k number C(16, k) out of 216 = 65,536: 12,870 lead to the centre, which takes 19.6% of all balls, and one leads to each edge, 1 in 65,536 or 0.0015%, the figure Stake’s own probability guide quotes. The five middle slots collect 79% of balls; pay them even 1× and nothing is left for the edges, hence 0.1× to 1×.
Plinko RTP is the sum of probability times multiplier over the slots. For all twelve rows above it comes to 98.9% to 99.1%; the spread is rounding in the displayed multipliers. The published 99% holds.
Worked example: 1,000 balls, 16 rows, High
Expected return $990, expected cost $10, exactly as on Low; the setting shows in the spread. The standard deviation is about $208, so finishing anywhere between roughly $575 and $1,400 is ordinary. The two 1,000× slots hit once in 32,768 balls, so your chance of seeing one in 1,000 balls is about 3%, yet they carry 3.05 points of the 99%. In the 97% of sessions where neither lands you are effectively playing a 95.9% game, a 4.1% edge.
| Setting, 16 rows | Expected return per $1,000 wagered | Standard deviation of a 1,000-ball session | Balls returning less than 1× | Share of the RTP in the two edge slots |
|---|---|---|---|---|
| Low | $990 | about $10 | 19.6% | 0.05% |
| Medium | $990 | about $46 | 54.6% | 0.34% |
| High | $990 | about $208 | 79.0% | 3.05% |
| Expert | $990 | about $1,754 | 79.0% | 30.5% |
Expert with 16 rows is the extreme: nearly a third of the return sits in two slots hit once in 32,768 balls, so the typical session returns about 68.5% of turnover and the standard deviation of a 1,000-ball run exceeds the amount wagered: the price of a 10,000× ceiling.
Provably fair, step by step
Server seed, client seed and nonce
The server seed is Stake’s random value; you see its SHA-256 hash before you bet, which commits Stake to it. The client seed is yours to type. The nonce counts your bets with the current pair. The Fairness panel (Overview, Seeds, Verify) puts it as “fair result = operator’s input (hashed) + player’s input”.
The six steps
- Fairness → Seeds: set your own client seed, then copy the hashed server seed, the client seed and the nonce.
- Play; note the nonce of the bet to check.
- Rotate the seed pair; Stake reveals the server seed.
- SHA-256 the revealed seed; it must equal the hash from step 1, which proves the seed was fixed before you bet.
- Compute HMAC-SHA256 (key: server seed; message: client seed and nonce) in the Verify tab or any script; the output bytes become one float per row.
- Per row, floor(float × 2) is 0 for left and 1 for right, the rule Stake publishes. Count the rights: that is the slot you were paid.
What the proof covers
It proves the path was not changed after you bet, not that the paytable is generous; that is the arithmetic above. It exists only on Originals; a Spribe or Hacksaw Plinko relies on the studio’s certified RNG. The operator’s licence is covered in is Stake safe.
Plinko strategy and why Martingale fails
No setting beats the 1%
The best Plinko strategy question has a short answer: none changes the expected result, because every ball is independent, every setting returns 99%, and no sequence of bet sizes alters a fixed percentage of each bet. You decide only the shape of the session. Low with 16 rows returns the stake or more on 80% of balls and swings about $10 per $1,000; Medium with 16 rows puts a 41× within reach once in 2,048 balls; Expert with 16 rows is a lottery ticket priced at 1% of turnover plus the variance above. Set the ball count in Autobet before you start, as Stake’s page advises; its stop limits are a budget.
Why Martingale fails
Martingale, doubling after each loss, needs a bet that wins about even money or loses. Plinko has none: on Medium with 16 rows, 54.6% of balls return under 1× and most of the rest 1× to 1.5×, so a “win” rarely repairs the chain. Eight doublings take a $1 ball to $256 and the balance or bet limit ends the chain first; meanwhile turnover rises, and 1% of more turnover is more loss. With bonus money it is impossible anyway: the maximum bet with the welcome bonus active is about 1% of the deposit.
Plinko versions compared
Same board, different paytables
Several studios licence their own Plinko to hundreds of casinos, and the paytables differ far more than the boards.
| Version | RTP | Max multiplier | Risk levels / rows | Verification | Where |
|---|---|---|---|---|---|
| Stake Originals Plinko | 99% | 10,000× | 4 / 8–16 | Provably fair | Stake only |
| BGaming Plinko (2019) | 99% | 1,000× | 3 (low, normal, high) / 8–16 | Studio RNG | Licensed to many casinos |
| Spribe Plinko | 97%, as reported | 1,000×, as reported | 3 / 8–16, as reported | Studio RNG | Licensed to many casinos |
| Hacksaw Gaming Plinko | As shown in game | 3,800×, as reported | Studio design | Studio RNG | Licensed to many casinos |
Which one to play
A 97% Plinko costs three times as much per ball as a 99% one: $30 instead of $10 per $1,000 wagered. Between the two 99% games the difference is the ceiling and the proof; themed variants trade RTP or hit frequency for their features. For Plinko real money play the published RTP is the only number that compounds, and Stake’s is the highest we have seen.
Fake Plinko apps and “predictors”
App-store Plinkos are not casinos
Search for a Plinko casino and the first results are often Google Play and App Store apps: simulators funded by adverts or coin purchases, where nothing comes out, not Stake and not plinko gambling. Stake’s Plinko runs inside stake.com or Stake’s own app where offered; there is no “Stake Plinko APK”, and any site offering one or asking for your Stake login is phishing.
Predictor apps and hack scripts
A ball’s path is HMAC-SHA256 of the server seed, your client seed and the nonce, and the server seed is secret until you rotate it. A working predictor would mean a leaked seed, which the hash commitment exposes. Every “Plinko predictor” we have seen sells a random-number toy or asks for your credentials.
How the Pactbonus deal applies
Welcome bonus wagering
Register with code PACTBONUS and the first deposit, $50 or more, is matched 200% up to $1,000: $100 becomes $300. Wagering is 40× deposit plus bonus, typically within 30 days; Stake Originals count 100% toward it at the time of writing, so Plinko qualifies. The maximum bet with the bonus active is about 1% of the deposit: on $100 that is $1 balls, 12,000 of them to clear $300, at an expected cost of about $120. Conditions: bonus terms; the offer: the Stake deal.
Rakeback and the leaderboard
Rakeback, 3.5% of the house edge for the life of the account with no wagering (VIP and rakeback), is worth $0.035 per $100 of Plinko turnover: $100 wagered carries $1 of house edge, and 3.5% of that is 3.5 cents; $10,000 returns $3.50. Every Plinko bet also counts toward the $30,000 monthly Pactbonus leaderboard. Neither changes the 1%; both are paid on top.
Play it for what it is
Plinko is a 1% game with a volatility dial. Set deposit and loss limits in Stake Smart before the first ball, fix the ball count in Autobet, and treat Expert as a ticket, not a plan. Stake does not serve Ontario and is an offshore operator elsewhere in Canada; see is Stake safe. 18+. If play stops being play, our responsible gambling page lists limits and help lines.
Common questions
Is Plinko rigged?
Stake Plinko is provably fair: the server seed is hashed before you bet, you supply the client seed, and after rotating the seed you can recompute every ball’s path with HMAC-SHA256. The paytable is public and we recomputed its 99% RTP from the binomial distribution. Rigged, no; a 1% house edge, yes.
What is the RTP of Stake Plinko?
99.00%, shown in the game panel, on every risk level and row count. Our own sum of probability times multiplier for twelve risk and row combinations came to 98.9%–99.1%, the difference being rounding in the displayed multipliers.
Is there a Plinko strategy that works?
Not in the sense of beating the house edge. Risk and rows change how the 99% is distributed, not its size, and Martingale only raises turnover, which raises the expected loss. The useful decisions are the setting that matches your goal and a fixed ball count in Autobet.
Which Plinko casino game has the best RTP?
Stake Originals Plinko and BGaming’s Plinko both publish 99%. Spribe’s version is reported at 97% and themed variants vary. Stake’s is the only one of these with a 10,000× ceiling and provably fair verification.
Can I play Plinko for real money in Canada?
Stake accepts players in Canadian provinces other than Ontario, where it has no iGaming Ontario registration. It is an offshore, Curaçao-licensed operator; read our guide on whether Stake is safe before depositing.
Does Plinko count toward the Stake welcome bonus wagering?
At the time of writing Stake Originals count 100% toward the 40× rollover, so Plinko qualifies. The maximum bet while the bonus is active is about 1% of the deposit. Check the terms page for the current weighting.
Are Plinko predictor apps real?
No. The result depends on a server seed nobody sees until it is rotated. Predictor apps and APKs are random-number toys or phishing for your login. Stake Plinko has no separate download; play it on stake.com. 18+; see our responsible gambling page if play stops being play.
Sources: stake.com/casino/games/plinko and stake.com/fi/casino/games/plinko (game panel, paytable, description, 2026-10-05); stake.com/provably-fair/game-events (Plinko direction rule); stake.com/blog/plinko-probability-and-distribution-guide (2025-12-30); bgaming.com/games/plinko (2026-10-05); our own binomial calculation of the paytables; Stake bonus terms as published at the time of writing.
