Plinko at Shuffle

What Shuffle publishes about Plinko

Plinko is a Shuffle Original, built by Shuffle's in-house studio, so the payout tables and the fairness proof are Shuffle's own. Its game page gives a specification table and, unusually, prints the minimum and maximum payout of every one of its 27 reward tables. It does not print an RTP percentage. The wording is "Plinko at Shuffle carries a house edge of around 1%" (Shuffle Plinko game page, 12 Nov 2026).

The page is candid about what the animation is: "a physics style rendering of a path that was already decided the moment you dropped, so what you watch is a replay of the result, not the thing producing it".

The 27 reward tables, as Shuffle prints them

Shuffle's page lists, for each risk level and row count, the number of slots and the lowest and highest payout. The lowest always sits in the centre and the highest in the two outermost slots. The chance column is added here: a ball reaches one particular outermost slot along a single path, so the chance of either is 2 divided by 2 to the power of the row count.

Minimum and maximum payouts: Shuffle Plinko game page, 12 Nov 2026. Chance of the outermost slots is arithmetic from the 50/50 step. * See the note on the 16-row Medium table below.

High risk pays 0.2x in the centre at every row count, and Shuffle says so directly: "The center pays 0.2x at every row count." The 1,000x figure exists on one table only, High at 16 rows, where either outermost slot is reached once in 32,768 drops on average.

Return and hit rate of each table

"Around 1%" can be made exact. The return of a table is each slot's multiplier times its binomial probability, added up. Using the full multiplier tables the game carried on 1 Nov 2026, whose minimum and maximum match the printed figures for 26 of the 27 tables:

Arithmetic from the multiplier tables and binomial odds. "Drops under 1x" is the share of drops that return less than the stake.

Returns run from 98.906% (8 and 10 rows on Medium) to 99.160% (11 rows on High): an edge between 0.84% and 1.09%, which is what "around 1%" turns out to mean. The three right-hand columns show what the risk setting really controls. On Low at 16 rows about one drop in five returns less than the stake. On High at 12 rows it is more than four in five. The cost per dollar wagered is almost the same; the route is not.

The one table that does not match is Medium at 16 rows. Shuffle's page prints 0.5x as its minimum; the game's table has 0.3x in the centre slot with 0.5x in the slots either side. Only the 0.3x is consistent with an edge near 1%: with 0.5x in the centre that table would return about 102.9%, more than is staked. The returns above use 0.3x.

Where the return of a High board comes from

Arithmetic for the 16-row High table. The last column is multiplier times chance; the column adds up to that table’s 98.976% return.

On this board the two 1,000x slots, hit once in 32,768 drops, account for about three percentage points of the return. The 130x and 26x slots add about sixteen more. A session that never reaches those slots, which is the ordinary case over a few hundred drops, is running well below 99% while it lasts. Shuffle's own page makes the same point about clips of 1,000x hits: "the thousands of 0.2x drops around it do not get posted".

Bet and win limits

Shuffle's help centre gives Plinko a maximum bet of $10,000 and a maximum win of $1,000,000 per round, in USD whatever your display currency. The article explains how the two interact: "if the combination of bet size and multiplier you select on a game is greater than the maximum win, you will be prevented from placing your bet". On the 16-row High board, whose top slot pays 1,000x, a stake above $1,000 would put the potential win over the cap; on the 16-row Medium board (110x) the ceiling is about $9,090, and on every Low board the full $10,000 fits. The limit is applied to the bet up front.

At the other end, the page puts the minimum at "around 0.10, with the exact figure depending on your currency", and the bet panel shows the figure for your coin. The demo "runs the same board, the same pin count, the same payout tables and the same house edge as real play" on a balance that cannot be withdrawn.

How a drop is derived from your seeds

Shuffle documents its generator in four help-centre articles. The commitment article names three values: an active client seed "determined by the player", an active server seed that "is kept secret, and created by the server" with only its hash shown, and a nonce that "is originally set to 0" and "will increase with every gameplay". The implementation article gives the function: HMAC-SHA256 with the server seed as key over the text clientSeed:nonce:currentRound, four bytes per number. For Plinko the game-events article then says:

floats = generateFloats({ serverSeed, clientSeed, nonce, cursor, count })  // count = number of rows
dropDirections = floats.map(val => Math.floor(val * 2))  // 1 = right, 0 = left
finalPositionFromLeft = sum of dropDirections

In words: one number per row, doubled and rounded down to 0 or 1, and the slot is the count of 1s from the left edge. An 8-row drop uses the first eight numbers of one hash; 9 rows and up run into a second hash with the round counter raised by one.

The article's worked Plinko example contains a slip worth knowing about if you test a tool against it. It lists a first number of 0.832 for client seed abc, server seed def and nonce 1. The same article's Dice and Limbo examples give 0.4305 as the first number for those same seeds, and that is the value the documented function produces. With it, the 8-row path for abc / def / 1 is left, right, left, right, right, left, right, left: four rights, the centre slot. The Plinko example's numbers belong to some other input.

To check one of your own drops, Shuffle's steps are: open the bet, choose "rotate pair", set a new client seed, and return to the bet, where "the server seed has been unhashed and revealed to you". Then:

• Commitment: the SHA-256 of the revealed server seed must equal the hashed seed the bet showed before rotation.
• Path on Shuffle: the Verify button in the bet window recomputes the result from the client seed, server seed and nonce.
• Path outside Shuffle: our Shuffle seed checker runs the documented function in your browser for 8 to 16 rows and prints the left/right path and the slot number. It does not look up the multiplier, which depends on the risk level you played. The Plinko verifier uses the same code.

The check establishes that the path was fixed by a seed committed before the drop. It has no bearing on the payout table, which is where the edge lives. Background: what provably fair gambling proves and the general verifier.

Bonuses and Plinko

Shuffle's game page says Plinko play "counts toward the weekly race and earns XP toward the VIP program". It also carries a caution that applies to anyone clearing a bonus: "Originals and table style games generally contribute less toward bonus wagering requirements than slots do, so check the contribution rate before you decide to clear a bonus here", and exceeding the stake cap during a bonus "is the most common way players void one". The contribution rate and the cap are in the terms of the specific bonus, which are the ones that bind.

Shuffle Plinko against Stake, Roobet and Gamdom

Stake: fairness documentation and payout table 1 Nov 2026, help centre 12 Nov 2026. Roobet and Gamdom: game pages and fairness pages, 1 Nov 2026.

Shuffle's Low, Medium and High tables are the same numbers as Stake's tables of the same names (both as carried in the games on 1 Nov 2026), and the two sites document the same generator, so identical seeds produce the same path on both. What Stake adds is a fourth, steeper risk level. What Shuffle adds is a flat dollar cap on a single win and a help-centre page that states it.

Sources: Shuffle Plinko game page (12 Nov 2026); Shuffle help centre, maximum bet and win limits, commitment schemes, implementation, game events and how to verify bets (12 Nov 2026); full multiplier tables as carried in the game on 1 Nov 2026. Returns, hit rates and the abc / def / 1 path are our arithmetic from those sources.

The Plinko calculator converts any table into expected loss per session. Shuffle's licence, KYC and withdrawal terms are in the Shuffle review.

Frequently asked questions

Related pages