Contents
Plinko Gambling Math: Exact Odds and Real RTP (2026)
A friend messaged me a screenshot last month: forty Plinko drops, thirty-one of them in the middle three buckets, and the caption "this thing is rigged". He was almost exactly on distribution. The expected number was 21.8, he got 31, and that gap is ordinary luck on a sample of forty.
The thing is, he was right to be suspicious, just about the wrong thing. Plinko has a genuine problem that nobody writes about, and it has nothing to do with the ball path.
Here is the exact maths of every bucket, the real answer to how often 1000x lands, and the RTP question that actually costs money.
Quick answer: the Plinko odds that matter
The numbers
| Question | Answer (16 row board) |
|---|---|
| Total possible paths | 65,536 |
| Chance of a specific edge bucket | 1 in 65,536 |
| Chance of either edge bucket | 1 in 32,768, or 0.00305% |
| Drops for a 50% chance of one edge hit | 22,713 |
| Share of drops landing in the centre three buckets | 54.55% |
| Verified RTP range across Plinko titles | 88.2% to 99% |
What this does not tell you
It does not tell you which bucket your next ball lands in, because nothing does. What it tells you is which outcomes are normal, and the distribution is far more lopsided than the board looks.
The board is drawn symmetrically and every bucket is the same width, so the eye reads seventeen roughly equal outcomes. The maths says the middle bucket is 12,870 times more likely than the edge next to it.
Disclosure
We earn a commission when a reader opens an account through some of the links on this site. No operator influenced anything below. The probability figures are computed from the binomial distribution and verified to sum to exactly 1. The RTP figures come from our own game database, each with a provider source and a verification date.
How a Plinko drop is actually decided
Sixteen rows, sixty-five thousand paths
A ball crossing a peg goes left or right. On a 16 row board it makes sixteen of those choices, so there are 2 to the power of 16 possible routes, which is 65,536.
Every route is equally likely. What is not equally likely is where you end up, because thousands of different routes lead to the middle and exactly one route leads to each edge. Reaching the far left bucket requires sixteen consecutive left bounces with no exceptions.
That is the entire game. Everything else is presentation.
Why this is not physics
On a physical Plinko board the peg collisions are chaotic but real. In the online version there is no ball and no pegs. The outcome is drawn from a seed, and the animation is rendered afterwards to match a result that already exists.
This matters for one practical reason: there is nothing to observe. Drop speed, where you release, how the previous ball fell, none of it exists as a physical process that could influence anything.
The formula
The chance of landing in bucket k on an n row board is:
The top part counts how many routes reach bucket k. The bottom is the total number of routes. In plain terms: count the ways to get there, divide by the ways to get anywhere.
If you want to watch this happen with a real ball count, the University of Colorado publishes a free Plinko probability simulator that drops thousands of balls and draws the distribution as it fills in.
The full probability table
Sixteen rows
| Bucket | Paths leading there | Probability | How often |
|---|---|---|---|
| 0 (edge) | 1 | 0.0015% | 1 in 65,536 |
| 1 | 16 | 0.0244% | 1 in 4,096 |
| 2 | 120 | 0.1831% | 1 in 546 |
| 3 | 560 | 0.8545% | 1 in 117 |
| 4 | 1,820 | 2.7771% | 1 in 36 |
| 5 | 4,368 | 6.6650% | 1 in 15 |
| 6 | 8,008 | 12.2192% | 1 in 8 |
| 7 | 11,440 | 17.4561% | 1 in 6 |
| 8 (centre) | 12,870 | 19.6381% | 1 in 5 |
| 9 | 11,440 | 17.4561% | 1 in 6 |
| 10 | 8,008 | 12.2192% | 1 in 8 |
| 11 | 4,368 | 6.6650% | 1 in 15 |
| 12 | 1,820 | 2.7771% | 1 in 36 |
| 13 | 560 | 0.8545% | 1 in 117 |
| 14 | 120 | 0.1831% | 1 in 546 |
| 15 | 16 | 0.0244% | 1 in 4,096 |
| 16 (edge) | 1 | 0.0015% | 1 in 65,536 |
Those probabilities sum to exactly 1.000000, which is the check that the table is right rather than plausible.
Eight and twelve rows
| Rows | Total paths | Edge bucket odds | Centre bucket |
|---|---|---|---|
| 8 | 256 | 1 in 256 | 27.34% |
| 12 | 4,096 | 1 in 4,096 | 22.56% |
| 16 | 65,536 | 1 in 65,536 | 19.64% |
Why more rows makes the top prize rarer
Adding rows multiplies the number of paths by two each time, while the number of routes to the edge stays at exactly one. That is why a 16 row board can afford to advertise a 1000x top multiplier and an 8 row board cannot: the 16 row edge is 256 times harder to reach.
Bigger top number, proportionally smaller chance. The advertised maximum tells you about the row count, not about generosity.
How often does Plinko actually hit 1000x
The exact answer
Two of the seventeen buckets on a 16 row board carry the top multiplier, one on each side. Each is reached by exactly one path out of 65,536, so the chance of hitting either on a given drop is 2 in 65,536.
That is 1 in 32,768, or 0.00305%.
What that looks like in playing time
| Drops | Chance of at least one edge hit |
|---|---|
| 1,000 | 3.0% |
| 5,000 | 14.2% |
| 10,000 | 26.3% |
| 22,713 | 50.0% |
| 50,000 | 78.3% |
Twenty-two thousand drops is the halfway mark. At one drop per second with no breaks that is 6.3 hours, and even then it is a coin flip whether you have seen one.
Every screenshot of a 1000x hit you have ever seen is real and also represents somebody's tens of thousands of drops, or somebody else's luck arriving early. Both stories produce the same screenshot, which is why screenshots are useless as evidence about a game.
The realistic ceiling is one bucket inwards
Chasing the edge is chasing something most players will never see. The bucket next to it is a different proposition entirely.
| Bucket | Paths | Odds per drop | Drops for a 50% chance |
|---|---|---|---|
| Edge | 1 | 1 in 65,536 | 45,426 |
| Second from edge | 16 | 1 in 4,096 | 2,839 |
| Third from edge | 120 | 1 in 546 | 379 |
| Fourth from edge | 560 | 1 in 117 | 81 |
The second bucket is sixteen times more reachable than the edge and still pays a large multiplier on most high-risk boards. The fourth is something you will meet several times in an ordinary session.
If you are playing for the top prize, the honest framing is that you are buying lottery tickets with a 1 in 32,768 hit rate while the game quietly charges its edge on every drop in between.
What "due for a big one" costs
After a long dry run the instinct is that the edge is overdue. Each drop is drawn independently, so a bucket that has not appeared in 30,000 drops is exactly as likely on the next one as it was on the first: 1 in 32,768.
The cost of acting on that instinct is measurable. Ten thousand extra drops at 398 of expected loss, bought in exchange for raising your chance of an edge hit from zero to about 26%.
Why your balls keep landing in the middle
Fifty-four percent inside three buckets
| Distance from centre | Share of all drops |
|---|---|
| Centre bucket only | 19.64% |
| Within one bucket | 54.55% |
| Within two buckets | 78.99% |
| Within three buckets | 92.32% |
| Within four buckets | 97.87% |
More than half of every session lands in three of the seventeen buckets, and those are the buckets that typically pay less than the stake back. That is the actual shape of Plinko, and it is why the game drains a balance smoothly rather than dramatically.
My friend's forty drops with thirty-one in the middle looked wrong to him because the board looks even. Run the numbers and he was inside normal variance. Our session simulator does the same job across thousands of sessions if you want to see what a normal bad run looks like before you meet one.
Plinko RTP is not a single number
Verified RTP by title
These come from our database of 3,764 tracked game entries, sourced from provider pages with a verification date on each record rather than from casino marketing.
| Title | Provider | RTP | Volatility |
|---|---|---|---|
| Plinko 2 | BGaming | 99% | Low |
| Plinko XY | BGaming | 99% | n/a |
| Football Plinko | BGaming | 99% | Low |
| Easter Plinko | BGaming | 99% | n/a |
| Plinko 2 Halloween | BGaming | 99% | Very high |
| Plinko | Hacksaw Gaming | 88.2% to 98.98% | n/a |
| Triple Pot Plinko Hercules | Pragmatic Play | 96.55% | n/a |
The Hacksaw problem
Hacksaw Gaming's Plinko ships in seven certified RTP builds. Not seven games. One game, seven versions of the maths, and the casino picks which one it licenses.
| Build | RTP | Cost per hour at 500 drops, $1 stake |
|---|---|---|
| 1 | 98.98% | $5.10 |
| 2 | 98.28% | $8.60 |
| 3 | 97.27% | $13.65 |
| 4 | 96.02% | $19.90 |
| 5 | 94.30% | $28.50 |
| 6 | 92.03% | $39.85 |
| 7 | 88.20% | $59.00 |
The worst build costs 11.6 times more per hour than the best one. Same artwork, same animation, same buckets, same 65,536 paths. The distribution above is identical across all seven, because the distribution is geometry. What changes is the paytable printed on those buckets.
This is the thing my friend should have been suspicious about, and it is invisible from inside the game.
Why a provider ships seven versions at all
This is not a scandal, it is a business model. A studio sells the same game into markets with different regulatory floors and different operator margins, and a lower RTP build is worth more to a casino because it earns more per player.
Some jurisdictions set a legal minimum return and the local build is cut to sit just above it. Elsewhere the operator simply chooses. Neither choice is announced to the player, and the game client looks the same in every case.
The consequence is uncomfortable but simple: "Plinko has 99% RTP" is a statement about a specific game from a specific studio, not about the genre. Repeat it as a general fact and you may be describing a game that returns 88.2%.
How to check which build you are on
Honestly, usually you cannot, and pretending otherwise would be the same trick the review sites pull. Three things are worth trying.
Open the game's own info or paytable panel, since many builds state the RTP there. Compare the multiplier values against another casino running the same title, because a lower build usually shows visibly smaller numbers on the middle buckets. And treat any casino that will not display an RTP figure anywhere as a casino that has a reason.
Where the figure is published, our RTP database and highest RTP games rankings track it, and BGaming's own Plinko page is an example of a provider that publishes a single flat number instead.
Compute the RTP of any board yourself
The method
Multiply each bucket's probability by its multiplier, then add everything up. That sum is the RTP.
Take an 8 row board with the multipliers 5.6, 2.1, 1.1, 1, 0.5, 1, 1.1, 2.1, 5.6 across its nine buckets. Bucket probabilities on 8 rows are 1, 8, 28, 56, 70, 56, 28, 8, 1 out of 256.
| Bucket | Probability | Multiplier | Contribution |
|---|---|---|---|
| 0 | 0.3906% | 5.6 | 0.0219 |
| 1 | 3.1250% | 2.1 | 0.0656 |
| 2 | 10.9375% | 1.1 | 0.1203 |
| 3 | 21.8750% | 1.0 | 0.2188 |
| 4 | 27.3438% | 0.5 | 0.1367 |
| 5 | 21.8750% | 1.0 | 0.2188 |
| 6 | 10.9375% | 1.1 | 0.1203 |
| 7 | 3.1250% | 2.1 | 0.0656 |
| 8 | 0.3906% | 5.6 | 0.0219 |
| Total | 0.9898 |
That board returns 98.98%. The multipliers above are an illustration rather than any particular operator's table, but the method transfers to whatever numbers are printed on your screen. Our RTP calculator handles the arithmetic if you would rather type the multipliers in than multiply seventeen rows by hand.
Notice what the biggest contribution is. The centre bucket pays 0.5x and still contributes more to the RTP than both 5.6x edges combined, because it happens seventy times as often.
Risk levels: what they change
| Setting | Edge multipliers | Centre multipliers | House edge | What actually changes |
|---|---|---|---|---|
| Low risk | Small | Near 1x | Unchanged | Smooth, slow bleed |
| Medium risk | Moderate | Below 1x | Unchanged | Middle ground |
| High risk | Very large | Well below 1x | Unchanged | Long droughts, rare spikes |
Risk mode is a variance dial. The operator rebalances the paytable so the sum still lands on the same RTP, which means high risk does not buy you value, it buys you a different shape of losing. This is the same trade-off we cover for slot volatility, and our volatility calculator puts a number on how big a bankroll each shape needs.
What Plinko costs to play
Speed is the part players underestimate. Plinko resolves in about a second and supports auto-drop, so turnover per hour is much higher than the stake suggests.
| Drops per hour | Stake | Turnover | Cost at 99% | Cost at 96.02% |
|---|---|---|---|---|
| 250 | $1 | $250 | $2.50 | $9.95 |
| 500 | $1 | $500 | $5.00 | $19.90 |
| 1,000 | $1 | $1,000 | $10.00 | $39.80 |
| 500 | $5 | $2,500 | $25.00 | $99.50 |
Doubling your drop rate costs exactly as much as doubling your stake, and the auto-drop button makes the first one effortless. Run your own numbers through the house edge calculator before deciding what pace is comfortable.
The auto-drop trap
Manual play has a natural ceiling. You get bored, you pause, you look away. Auto-drop removes all three brakes at once and it is presented as convenience rather than as the largest cost multiplier in the game.
A player who drops manually might manage 250 an hour. The same player on auto-drop with a short interval clears 1,000 without noticing, which is a four times increase in expected loss with no change to stake size, risk level or row count. Nothing about the decision felt like betting more.
Set a drop count limit before starting rather than a time limit, since time is the thing auto-drop distorts.
Strategies and what they actually do
| Strategy | The claim | Reality |
|---|---|---|
| Alternating risk levels | Catches hot streaks | Each drop independent, no streak exists |
| Increasing stake after losses | Recovers the deficit | Same edge, larger exposure |
| Targeting a specific bucket | Improves the odds | You cannot aim, the seed decides |
| Reading previous drops | Finds a pattern | Drops carry no memory |
| Playing fewer rows for safety | Reduces the edge | Reduces variance only |
The row count is the one honest lever, and it is a variance lever rather than a value one. If the pattern-hunting instinct is familiar, our page on the gambler's fallacy explains where it comes from.
Verification is worth doing once. The provably fair checker will confirm that a drop was committed before you released it, which settles the rigging question in a way that watching the board never will.
For the same treatment of the two games most often played alongside this one, see crash game odds and Mines cash-out maths. If you are looking specifically at how the game is presented at one operator rather than at the maths, we cover Plinko at 1win separately.
Where to play Plinko
The paytable decides more than anything else in this article, which makes the operator a bigger lever than the row count.
Outside the US, Plinko at 1win is the version we cover. The site takes players from India, Nigeria, Bangladesh, Pakistan, Kenya and dozens of other markets, and its row and risk settings are broken down in a dedicated guide. From the same instant-game shelf it also runs JetX and Lucky Jet. US players are not accepted there, so the licensed option named earlier applies instead.
Three things to confirm before the first drop, on any site:
- Whether the RTP is published for that specific build, not for "Plinko" in general
- Whether the round hash is shown before the ball starts falling
- Whether the edge-bucket multipliers match what the description promised
None of this makes the game profitable. It rules out the versions where you overpay.
The verdict
Plinko's geometry is the most honest thing about it. Sixteen rows, 65,536 paths, one route to each edge, and a distribution that piles half of everything into three buckets. None of that is hidden and none of it is adjustable.
The paytable is where the money is decided, and that is the part that varies by an order of magnitude between builds of the same game. A player worrying about whether the ball path is fair is watching the wrong half of the screen.
Find the published RTP before you play. If nobody will show you one, that is the answer. Readers in the US who want a licensed operator rather than an unregulated crypto site can start at Bovada, though as always the RTP question is worth asking there too.
Frequently Asked Questions
Evgeniy Volkov
Verified ExpertFullstack developer with a background in mathematics. I build the calculators and game-style tools on ToolsGambling with Pixi.js and modern web tech, and every result uses transparent probability formulas you can verify yourself.