Plinko RTP and house edge: the odds for every board

Início » Plinko in the US: How It Works, What It Pays and the Law » Plinko RTP and house edge: the odds for every board
Last updated:

Plinko RTP is no mystery number. The game is a string of fair left-or-right turns sitting on top of a table, so the chance of every slot and the return of the whole board fall straight out of arithmetic. This guide does that arithmetic for all 27 boards in our tables, with three risk levels on boards between 8 and 16 rows tall, and then asks the harder question: how many drops must you watch before your own results tell you anything? We borrowed the multipliers from a published sample, keep the stake at 1 chip and run every figure ourselves in code. What stands out is that each board’s house edge hovers around 1%, and no amount of watching reveals it.

Key Takeaways: Plinko RTP and House Edge

  • Narrow House Edge: Every board from 8 to 16 rows across all risk levels holds an RTP between 98.91% and 99.16%, fixing the house edge near 1%.
  • Combinatorial Odds: Slot probabilities stem strictly from row count (2 to the power of R paths); changing risk modifies the multiplier table, not slot chances.
  • Sample Size Demands: Single sessions cannot confirm RTP; due to heavy variance, a 16-row high-risk board requires around 430,000 drops to narrow error to 1 point.
  • Hit Rate vs Return: Breaking even often does not mean higher overall return; 16-row boards hit 80.36% on low risk versus 21.01% on high risk, yet both yield ~99% RTP.

Where Every Slot Chance Comes From

A ball on a board with R rows makes R turns, so there are 2 to the power R paths, and each path is as likely as any other. A slot’s chance is the number of paths ending in it divided by the total. Slot k + 1 collects the paths with k right turns, and counting them is a combinations problem: choose which k of the R turns go right.

None of this touches the multipliers. The chances belong to the row count alone, which is why one set of chances serves all three risk settings.

A Nine-Row Board, Slot by Slot

Nine rows give 512 paths and 10 slots. Two slots share the middle, with 126 paths apiece, so each takes 24.61% of drops, while each end slot gets a single path, which makes the pair a 1-in-256 event.

SlotWays to land thereOddsLowMediumHigh
110.195%5.6×18×43×
291.76%2×4×7×
3367.03%1.6×1.7×2×
48416.41%1×0.9×0.6×
512624.61%0.7×0.5×0.2×
612624.61%0.7×0.5×0.2×
78416.41%1×0.9×0.6×
8367.03%1.6×1.7×2×
991.76%2×4×7×
1010.195%5.6×18×43×

The board is symmetrical, so slot 2 mirrors slot 9 and so on, and only the last three columns depend on the risk setting.

From Chances to a Return

RTP measures what a chip is worth on average: the mean amount returned for each chip staked. To get it for a board, multiply every slot’s chance by its payout and sum the results. On 9 rows at medium risk the total is 99.14%, which leaves the house with 0.859% and makes this one of the cheaper boards in our tables.

Any table yields to the same treatment, so a game that lists its row count and its multipliers gives you everything needed to test the return it claims.

RTP Across All 27 Boards

This table gives the return for every combination of rows and risk level.

Board rowsLowMediumHigh
898.98%98.91%99.06%
998.98%99.14%99.06%
1099.00%98.91%99.06%
1199.00%99.02%99.16%
1298.98%98.99%99.12%
1399.00%98.99%99.09%
1499.00%98.99%98.98%
1599.00%99.00%99.03%
1699.00%98.99%98.98%

Our totals span 98.91% at the bottom to 99.16% at the top. The source we borrowed the tables from summarizes its own range as 98.91% to 99.16%, so the two agree. Separately, a provider’s page says its game pays back 99%, which fits inside that band.

The Same Number as a House Edge

Turn the return upside down and you get the house edge, the slice of each chip that never comes back. In our tables it runs from 0.84% to 1.09%, which works out to a leak of 0.84 to 1.09 chips per hundred staked. The slice is widest on the 8-row and 10-row medium boards and slimmest on the 11-row high board.

A margin that slim makes a handy yardstick when you set Plinko beside other games, which the comparison guide does. It is also a long-run average over a huge number of drops, so one session hides it.

Redo the Math in a Spreadsheet

You can verify any table in five moves.

  1. Type in the multiplier for each slot, covering both halves of the board.
  2. Work out the paths into each slot, meaning the ways to pick its right turns.
  3. Divide those counts by 2 raised to the row count to get chances.
  4. Multiply chance by multiplier, slot by slot.
  5. Total the products for the return, and subtract that total from 1 for the edge.

We ran the steps on all 27 boards and the totals landed inside the published range. One snag turned up: the reference prints a return column next to its multipliers, and for the low-risk boards of 8 to 11 rows that printed number does not match what the multipliers produce. We therefore recomputed every figure ourselves.

Why One Session Cannot Confirm RTP

Yet a single drop returns only one multiplier, and the possible multipliers sit far apart. Averaging a few hundred drops can leave you well away from the real return, with the error falling in proportion to the square root of how many drops you have.

Board setupDrops needed to cut the error to 1 point of returnUsual error after 1,000 drops, in points
8-row board, low risk3,1001.8
12-row board, medium risk17,0004.1
16-row board, medium risk22,0004.6
16-row board, high risk430,00020.8

After 1,000 drops on the 16-row high board, the usual error is 20.8 points of return, many times the size of the whole edge, and about 430,000 drops are needed to bring it to 1 point. The quiet 8-row low board does better, with 1.8 points after 1,000 drops and 1 point at about 3,100.

So a session of a few hundred drops can neither confirm nor refute a game’s stated return. A hot session proves nothing about the game, and neither does a cold one.

Hit Rate Is Not Return

Players often read the share of drops that break even as a sign of generosity, but it answers a different question. A board can return the stake on most drops and still pay little overall, or do so rarely and pay plenty. On 16 rows the share is 80.36% at low risk and 21.01% at high risk, while the returns are 99.00% and 98.98%.

Think of it this way: hit rate counts how many drops end level or better, whereas return weighs each drop by its size. Only the weighted figure sets what play costs over the long haul.

What the Edge Costs per Hour

Volume is what turns a slim edge into real money. A drop every 3 seconds packs 1,200 drops into an hour, so 1 chip a drop on the 16-row high board should cost about 12.28 chips an hour, and the 9-row medium board about 10.31. Those are averages, and a real hour can land well away from them. The bankroll guide measures how far.

The Plinko hub indexes all eight guides, and the other seven are listed here.

A Small Edge Still Adds Up

A thin edge is still a charge on every drop, and fast games collect it often. The responsible gaming page lists limit tools and the services to call if play stops feeling like a choice.

❓ RTP and House Edge: Frequently Asked Questions

What RTP does Plinko have?

Between 98.91% and 99.16% in the tables we use, depending on the rows and the risk level, a span the source of those tables reports too. A provider’s page puts its own game at 99%. For the game you actually play, trust the stated return on its rules screen. Roughly 99%.

How large is Plinko’s house edge?

Around 1% on our tables, within a band of 0.84% to 1.09%. In chips, that is roughly 1 chip lost for every hundred staked, spread across every drop however the ball falls, and far too small to notice in one sitting. About 1% per drop.

How are Plinko odds calculated?

By counting paths. Each board has 2 to the power R equally likely paths for R rows, and a slot’s odds equal the paths into it divided by that total. On 9 rows, each of the two middle slots takes 126 of 512 paths, which is 24.61%. Paths over total paths.

How long until Plinko’s true RTP shows up?

Much longer than a session lasts. On the 16-row high board, a typical error of 1 point needs about 430,000 drops, and after 1,000 drops it is still 20.8 points. Even the calm 8-row board wants about 3,100. Longer than any session.

Author:

Lucas Portela

Lucas Portela

Owner, BoldGambler • Avanhandava/SP

Oddsmaker, affiliate and content creator in the iGaming industry.