Home › Plinko gambling explained
Plinko Gambling Explained: How It Works, Odds & RTP
Plinko looks like the simplest game in the crypto casino: drop a ball, watch it bounce down a pyramid of pegs, and collect whatever multiplier slot it lands in. Underneath, it is one of the most mathematically transparent games ever built — its outcomes follow a textbook binomial distribution, which means every probability on the board can be computed exactly rather than estimated. We ran that maths ourselves for a standard 16-row board. The finding that matters most: every risk level — low, medium and high — returns the same ~99% to players and keeps the same ~1% house edge. Changing the risk setting does not change how much the house wins; it only changes how wildly your balance swings on the way to losing it.
Plinko gambling: key facts (16-row board)
- Plinko's landing pattern is a binomial distribution: on 16 rows the ball lands in one of 17 slots across 65,536 possible paths.
- The centre slot is the single most likely landing spot at 19.64% (16Best analysis).
- The three central slots together catch 54.5% of all drops (16Best analysis).
- Each edge slot — where the biggest multiplier sits — is hit just 1 in 65,536 drops (0.0015%).
- Hitting a 1000× top multiplier (either edge) on high risk is 1 in 32,768 drops, about 0.003% (16Best analysis).
- Computed RTP is ~99% and the house edge ~1.0% on low, medium and high risk (16Best analysis).
- High risk is ~20× more volatile than low risk at the same RTP (standard deviation 6.57 vs 0.33) (16Best analysis).
- Top multipliers by risk on 16 rows: 16× (low), 110× (medium), 1000× (high); expert mode can reach 10,000×.
- At $1 a drop and a typical 500 drops/hour, expected loss is about $5/hour (16Best analysis).
- Provably-fair Plinko proves the drop was not rigged — it does not remove the 1% edge.
How does Plinko gambling work?
A ball drops through rows of pegs, bouncing left or right at each peg with a 50/50 chance, and lands in one of the multiplier slots at the bottom. On a 16-row board the ball makes 16 independent left-or-right decisions, so it can finish in any of 17 slots. The centre slots pay the least — often less than your stake back — while the rare edge slots pay the headline multipliers.
The game is a Stake Original at its most famous, but near-identical versions run at BC.Game, BGaming, Spribe and dozens of other crypto casinos (for context on the operators, see how crypto casinos work). You choose two settings before each drop: the number of rows (typically 8 to 16) and a risk level (low, medium or high). Those two dials decide the entire multiplier layout — and, as we show below, they leave the house edge untouched.
Because each peg is a fair coin flip, the landing position is not random guesswork — it is governed by the same maths that describes flipping 16 coins and counting heads. That is what makes Plinko unusually easy to analyse: the odds are not a casino secret, they are Pascal's Triangle.
What are the odds of landing in each Plinko slot?
The probability of each slot follows the binomial formula P(slot k) = C(16,k) / 216 — the number of distinct peg-paths that reach that slot, divided by the 65,536 total paths. The centre is common, the edges are vanishingly rare, and the shape is a perfect bell curve.
16Best analysis. Binomial distribution P(k)=C(16,k)/2^16. Bars for off-centre positions combine the two symmetric slots. The three central columns alone catch 54.5% of drops; the edge multiplier is reached 3 times in 100,000.
| Slot (from left) | Paths C(16,k) | Probability | Odds |
|---|---|---|---|
| Centre (slot 8) | 12,870 | 19.638% | ~1 in 5 |
| Slots 7 & 9 | 11,440 each | 17.456% each | ~1 in 5.7 |
| Slots 6 & 10 | 8,008 each | 12.219% each | ~1 in 8.2 |
| Slots 5 & 11 | 4,368 each | 6.665% each | ~1 in 15 |
| Slots 4 & 12 | 1,820 each | 2.777% each | ~1 in 36 |
| Slots 3 & 13 | 560 each | 0.854% each | ~1 in 117 |
| Slots 2 & 14 | 120 each | 0.183% each | ~1 in 546 |
| Slots 1 & 15 | 16 each | 0.0244% each | 1 in 4,096 |
| Edge slots 0 & 16 | 1 each | 0.00153% each | 1 in 65,536 |
16Best analysis: the centre slot at 19.64% is more than 12,800× more likely than an edge slot at 0.00153%. Put differently, in a night of 1,000 drops you would expect to land in the exact centre about 196 times and hit a specific edge 0.015 times — i.e. almost certainly never. This is why marketing that shows a ball dropping into the 1000× slot is showing you an event that happens roughly once every 65,536 drops. The bell curve is the whole game.
Why is hitting the top multiplier so rare?
Because the biggest multiplier sits in the two edge slots, and reaching an edge requires the ball to bounce the same direction 16 times in a row. On 16 rows that is 1 in 216 = 1 in 65,536 for one specific edge, or 1 in 32,768 (0.00305%) for either edge since the top multiplier is paid on both.
Hitting the 1000× top multiplier on a 16-row board is a 1 in 32,768 drop — about 0.003%.
To make that concrete: if you dropped one ball every second without stopping, you would expect to wait roughly 9 hours between top-multiplier hits on average — and "on average" hides enormous variance, because rare events cluster and drought unpredictably. The 10,000× expert-mode payout is rarer still, gated behind the same single edge slot on a board built to make that outcome almost unreachable.
What is the RTP and house edge of Plinko?
Plinko's return-to-player is about 99%, giving the house roughly a 1% edge. That figure is not a claim we have to take on trust — it is the expected value of the payout, which we can compute directly by multiplying each slot's probability by its multiplier and summing. We did exactly that for Stake's standard high-risk 16-row layout.
| Distance from centre | Combined probability | Multiplier (high risk) | Contribution to RTP |
|---|---|---|---|
| Edge (±8) | 0.00305% | 1000× | 0.0305 |
| ±7 | 0.0488% | 130× | 0.0635 |
| ±6 | 0.3662% | 26× | 0.0952 |
| ±5 | 1.709% | 9× | 0.1538 |
| ±4 | 5.554% | 4× | 0.2222 |
| ±3 | 13.330% | 2× | 0.2666 |
| ±2 | 24.439% | 0.2× | 0.0489 |
| ±1 | 34.912% | 0.2× | 0.0698 |
| Centre | 19.638% | 0.2× | 0.0393 |
| Expected value (sum) = RTP | 0.990 → 99.0% | ||
Every Plinko risk level returns about 99% to players — a ~1% house edge. Risk changes variance, not the edge.
16Best analysis: summing probability × multiplier across all 17 slots gives an expected value of 0.990 — you get back 99 cents on the dollar on average, a 1.0% house edge. Notice that the single most-common outcomes (the centre and its neighbours, 74% of all drops combined) pay just 0.2× — they are net losers. The entire 99% return is propped up by the far right of the table: the rare 2×, 4×, 9× and beyond. Plinko funds its "you almost broke even" feel with edge slots you will essentially never reach. For the general principle, see what is RTP and house edge.
Do the risk levels change the house edge?
No. Low, medium and high risk all return ~99% with a ~1% house edge on a 16-row board. This surprises most players, who assume "high risk" must mean the casino takes more. It does not. We computed the expected value for all three of Stake's standard 16-row layouts and they converge on the same number.
16Best analysis of Stake standard 16-row layouts. The top multiplier ranges from 16x to 1000x across risk levels, yet all three produce the same ~99% RTP. Risk redistributes the payout, it does not change the average return.
| Risk level (16 rows) | Centre payout | Top multiplier | Computed RTP | House edge |
|---|---|---|---|---|
| Low | 0.5× | 16× | 99.0% | 1.0% |
| Medium | 0.3× | 110× | 99.0% | 1.01% |
| High | 0.2× | 1000× | 99.0% | 1.02% |
16Best analysis: the three risk levels return 0.9900, 0.9899 and 0.9898 respectively — identical to two decimal places. The casino built each multiplier table to hit the same expected value on purpose. What "high risk" actually buys you is a lower centre payout (0.2× instead of 0.5×) financed into a much larger — but much rarer — top prize. You are not paying a bigger edge for the 1000×; you are trading the frequency of small wins for the faint possibility of a huge one. Same house cut, redistributed.
How much more volatile is high risk than low risk?
About 20 times more volatile, measured by standard deviation — even though the average return is identical. Variance, not edge, is the real difference between Plinko's risk levels. We computed the standard deviation of the payout for each 16-row layout.
| Risk level (16 rows) | RTP (mean return) | Variance | Standard deviation | Relative volatility |
|---|---|---|---|---|
| Low | 99.0% | 0.11 | 0.33 | 1× (baseline) |
| Medium | 99.0% | 2.15 | 1.47 | 4.4× |
| High | 99.0% | 43.10 | 6.57 | 19.8× |
High-risk Plinko is ~20× more volatile than low risk (SD 6.57 vs 0.33) — at the same 99% RTP.
16Best analysis: the variance gap is even starker than the standard-deviation gap — high risk's variance of 43.1 is roughly 390× low risk's 0.11. In plain terms: on low risk your balance drifts down slowly and quietly at 1% a spin; on high risk your balance mostly bleeds at 0.2× per drop while you wait for an edge hit that, statistically, most sessions never see. The expected destination is the same — the ride there is 20 times bumpier. Volatility is the product being sold, and it is orthogonal to the house edge. This is the same distinction that separates Crash cash-out strategies from the underlying edge.
How do the number of rows change the odds?
More rows means more slots, longer odds on the edges, and a bigger maximum multiplier — but the ~1% house edge holds at every row count. Each added row doubles the number of possible paths, because it adds one more coin-flip to the ball's journey.
| Rows | Slots | Total paths (2N) | Odds of one edge slot | Typical top multiplier (high) |
|---|---|---|---|---|
| 8 | 9 | 256 | 1 in 256 | ~29× |
| 12 | 13 | 4,096 | 1 in 4,096 | ~170× |
| 14 | 15 | 16,384 | 1 in 16,384 | ~420× |
| 16 | 17 | 65,536 | 1 in 65,536 | 1000× |
16Best analysis: going from 8 to 16 rows makes a specific edge slot 256× rarer (1-in-256 to 1-in-65,536), and the casino compensates by dangling a top multiplier roughly 34× larger (29× to 1000×). The two effects roughly cancel in expected-value terms — which is exactly why the RTP stays at ~99% no matter how many rows you pick. Row count, like risk level, is a variance dial, not an odds-of-winning dial.
How much does Plinko cost per hour?
At a 1% house edge, expect to lose about 1% of everything you wager — roughly $5 an hour at $1 a drop and 500 drops per hour. Plinko is fast: with auto-play and turbo mode a player can fire off hundreds or over a thousand drops per hour, and every drop is a fresh 1% expected loss on the amount staked.
| Stake per drop | Drops per hour | Total wagered/hr | Expected loss/hr (1% edge) |
|---|---|---|---|
| $0.10 | 500 | $50 | $0.50 |
| $1 | 500 | $500 | $5 |
| $1 | 1,000 (turbo) | $1,000 | $10 |
| $10 | 500 | $5,000 | $50 |
| $10 | 1,000 (turbo) | $10,000 | $100 |
16Best analysis: expected loss per hour = stake × drops-per-hour × 0.01. The danger of Plinko is not a high edge — 1% is one of the lowest in any casino — it is speed × volume. A player who would never sit at a 5%-edge slot for $5,000 of action can wager that same $5,000 through Plinko in an hour almost without noticing, because each individual drop feels tiny. Low edge, high throughput: the same economics that define crash gambling and most crypto-native "originals". The turbo button is the real house advantage.
Does provably fair make Plinko beatable?
No. Provably fair proves the drop was not rigged — it does nothing to remove the 1% house edge. This is the single most misunderstood point about crypto Plinko, and it is worth stating plainly.
In a provably-fair Plinko, the casino generates a server seed and publishes its cryptographic hash (SHA-256) before you play, committing to a result it cannot later change. Your client seed is mixed in, the combined value is hashed to a number from 0 to 65,535, and that number maps to a landing slot via the binomial (Pascal's Triangle) weights. After the round, the server seed is revealed and you can confirm it matches the original hash — proving the outcome was fixed at commit time, not manipulated against you. The full mechanism is covered in provably fair gambling explained.
What provably fair does and does not do. It does guarantee the ball's path was a genuine 50/50 at each peg and that the casino did not steer it away from a slot you were about to hit. It does not change the multiplier table, and the multiplier table is where the 1% edge lives. A perfectly fair, fully verifiable 50/50 drop into a board that pays 99 cents on the dollar is still a board that pays 99 cents on the dollar. Verifiability is honesty about the odds, not better odds. When choosing where to play, see how to spot a safe crypto casino.
Why do sources disagree on Plinko's RTP?
Because different Plinko versions ship with different multiplier tables, and a few operators advertise a headline RTP that only applies under specific settings. Not all Plinko is 99%.
- The 99% figure is Stake/BGaming's Original. Stake, BGaming and several clones publish a stated 99% RTP (1% edge) across rows and risk levels — and our own expected-value maths confirms it for the standard 16-row tables.
- Other studios ship lower. Some third-party Plinko games run RTPs around 97% (a 3% edge) or advertise capped "555× max win" boards with different payout distributions. A game called "Plinko" tells you the mechanic, not the edge.
- "Zero-edge" versions exist but are conditional. A handful of operators market 100% RTP Plinko, usually within a capped daily allowance or as a promotional loss-leader — not a permanently beatable game.
- Single-session RTP is not the number you experience. RTP is a long-run average over hundreds of thousands of drops. In any one session, high-risk variance means your realised return can be 0× or 400× — the 99% only emerges across an enormous number of drops.
16Best analysis — the trap in the RTP number. A stated 99% RTP is a lifetime figure. Because high-risk Plinko puts ~74% of its payout weight on 0.2× centre outcomes, the median session on high risk loses money faster than the 1% average suggests, and the average is only rescued by rare edge hits that most players never get. When comparing two Plinko games, always check (a) the stated RTP, (b) whether it applies to all risk levels or just one, and (c) the max multiplier — a higher cap always means a more punishing centre. Two games both labelled "99%" can feel completely different at the bankroll level.
Key takeaways
- Plinko is pure binomial maths. On 16 rows, P(slot k) = C(16,k)/216; the centre is 19.64%, each edge is 1 in 65,536.
- The top multiplier is a mirage. 1000× on high risk is a 1-in-32,768 drop (0.003%).
- Every risk level returns ~99%. Low, medium and high all compute to a ~1% house edge — risk does not change the edge.
- Risk = variance, not edge. High risk is ~20× more volatile than low risk (SD 6.57 vs 0.33) at identical RTP.
- More rows = rarer edges, bigger cap, same edge. Row count is a variance dial too.
- The real cost is speed. At $1 × 500 drops/hr you lose ~$5/hr; turbo doubles that. Low edge, high throughput.
- Provably fair ≠ beatable. It proves the drop was honest; the 1% edge is untouched.
- Not all "Plinko" is 99%. Some clones run 97% or lower — check the specific game.
Frequently asked questions
What is the RTP of Plinko?
About 99% on the standard Stake and BGaming versions, giving a roughly 1% house edge. Our own expected-value calculation on the 16-row tables confirms ~99% for low, medium and high risk alike. Some third-party Plinko games run lower, around 97%, so the exact figure depends on which version you play.
Does high risk in Plinko have a bigger house edge?
No. Low, medium and high risk all return about 99% with a ~1% house edge on a 16-row board. High risk does not cost you a bigger edge — it only makes your results far more volatile, roughly 20 times more volatile by standard deviation, while the average return stays the same.
What are the odds of hitting the 1000x multiplier in Plinko?
On a 16-row high-risk board, the 1000x multiplier sits in both edge slots, so the chance of hitting it is 2 in 65,536, or 1 in 32,768 — about 0.003% per drop. A single specific edge slot is 1 in 65,536.
Where is the ball most likely to land in Plinko?
In the centre. On a 16-row board the centre slot is hit 19.64% of the time, and the three central slots together catch 54.5% of all drops. Those central slots pay the lowest multipliers — often less than your stake — which is why most drops are small losses.
Can you beat Plinko with a strategy?
No strategy changes the house edge. Every drop is an independent event with a fixed ~1% edge, so no pattern, row count or risk level makes the game profitable over time. Risk and row settings only change variance. Provably-fair verification proves the game is honest but does not make it beatable.
Is Plinko provably fair?
The crypto-casino Original versions are. The result is committed via a SHA-256 hash before you play, your client seed is mixed in, and the server seed is revealed afterward so you can verify the outcome was not manipulated. This guarantees a fair 50/50 drop but does not remove the built-in house edge.
How much can you lose playing Plinko per hour?
Expect to lose about 1% of everything you wager. At $1 per drop and 500 drops per hour that is roughly $5 an hour; at $10 per drop it is about $50, and turbo auto-play can double those figures by doubling the number of drops.
Sources
- Stake.com — Plinko Game — Rules, Risk Levels & Provably Fair
- GamblingCalc — Plinko Calculator: Odds, Multipliers & RTP by Rows/Risk (2026)
- GamblingCalc — Plinko Probability Explained: Odds, Distribution & Binomial Math
- GamblingCalc — Stake Plinko Strategy: Do Rows and Risk Levels Change Your Edge?
- Tech-Insider — Crypto Casino Plinko: How It Works, RTP, and Strategy
- The Coin Republic — Why Stake Plinko Is a Trusted Provably Fair Casino Game
- FreeSlots99 — Plinko Odds Explained: Probabilities, RTP & Winning Chances (2026)