A fair game can still
wipe out your bankroll.

01Fair doesn’t mean smooth

On Duel’s Plinko at 16 rows, High risk, 79% of bets pay back 0.20x. Long losing runs are not evidence something is wrong; they are the published design. Nearly all of the payback sits in rare hits at the edges of the board. That is the appeal: this mode exists for the shot at the big multiplier.

Until one of those hits lands for you, a fair game and a rigged one feel identical from inside a session. The gap between what the math guarantees over millions of rounds and what it does to one bankroll is variance.

100 bets. Mostly small losses.

79 SMALL LOSSES20 OTHER WINS1 OUTSIZED HIT

Highlight the 131× hit ›

Tap a bet to inspect its exact payout.

10.220.230.240.250.262×70.280.290.2100.2110.21226×130.2149×150.2160.2170.2180.2190.2200.2212×220.2230.2242×25131×260.2270.2280.2290.2300.2310.2320.2330.2342×350.2360.2370.2380.2392×400.2410.2422×432×440.2450.2460.2472×480.2490.2500.2510.2520.2532×540.2550.2560.2570.2580.2590.2600.2610.2620.2630.2640.2650.2660.2672×680.2690.2702×710.2720.2732×740.2750.2760.2770.2782×790.2800.2814×820.2830.2840.2852×860.2870.2880.2890.2904×910.2920.2930.2942×950.2962×970.2980.2990.21000.2

212×220.2230.2242×25131×260.2270.2280.2290.2300.2310.2320.2330.2342×350.2360.2370.2380.2392×400.2

Bets 21–40 of 100←→

Grey: 0.2× · Blue: wins · Orange: outsized hit

Selected bet

Bet 25

Exact payout

131.21304367×

Net at a $5 stake

+$651.07

ALL 100 BETS+$614

Net result at $5 per bet

WITHOUT THE SINGLE 131× HIT−$37

Net result of the remaining 99 bets

100 consecutive real bets — seed pair #00 from our audit capture, bets 1–100. Hover or select a square for its exact payout. 79 paid 0.2×. A single 131× turned the window positive (+$614 at $5 bets). Remove that one bet and the remaining 99 net −$37.

02100 simulated players, one real seed pair each

Our audit of Duel’s Plinko captured 8,100 live bets through ordinary, anonymous accounts and verified 162 revealed seed pairs: every seed matched its pre-committed hash, every bet reproduced exactly against our independent rebuild. Because each outcome is a deterministic function of the server seed, client seed and nonce, a revealed pair can be rolled forward past the captured bets.

To illustrate how variance works, we selected 100 of those pairs, in capture order. Each pair is one deterministic betting sequence: think of it as one player, replayed forward as though they kept betting. These are not live bets we placed: they are exactly what the live game would have produced on those seeds, and anyone can recompute every line.

100 players. Find their story.

Choose an outcome, then hover or select a player to inspect their session.

All · 100× Wiped out · 78↘ Down · 2↗ Up · 5↑↑ Big wins · 15

Player 00 selected. ↑↑ Finished up big. Ending bankroll $5,188 after 10,000 bets.

00010203040506070809101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960616263646566676869707172737475767778798081828384858687888990919293949596979899

PLAYER00PLAYER01PLAYER02PLAYER03PLAYER04PLAYER05PLAYER06PLAYER07PLAYER08PLAYER09

Players 1–10 of 100←→

SELECTED PLAYER

Player 00

↑↑ Finished up big

Ending bankroll

$5,188

Highest bankroll

$6,168

Bankroll over the run · $5 bets

Hover or tap the chart to inspect the nearest recorded sample. Use arrow keys for adjacent samples, Page Up or Page Down for ten samples, Home or End for the first or last sample. Escape clears the selection.

Hover or tap the chart to inspect the session.

100 players, each starting with $1,000 and betting $5. Select a player to follow their session. The outcome is fixed by the revealed seeds; the game’s rules are the same for everyone.

78 wiped out

Median bust: bet 2,599 · y-axis to $5,000

Tap to zoom →2 holding on, down

Finished $411 and $778 · y-axis to $3,500

Tap to zoom →5 finished up

$1,224–$2,376 · y-axis to $4,000

Tap to zoom →15 finished up big

$2,519–$8,362 · all ten 1009×-hitters live here · y-axis to $9,500

Tap to zoom →

These are the same players and the same starting bankroll. Each chart has its own vertical scale. A winning screenshot shows one path, not the experience of the whole group.

03Same seeds. Only the bet size changed.

Rerun the identical 100 seed pairs — the same bets, in the same order — and change only the stake. The outcomes don’t change. How long a bankroll survives them does:

Turn the bet-size dial

$5 bets$10 bets$20 bets

78 of 100busted within 10,000 bets

bet 2,599median bust point

22survivors

0%

25%

50%

75%

100%

0

2,000

4,000

6,000

8,000

10,000

Bets

$5 · 22

$10 · 13

$20 · 5

0%

25%

50%

75%

100%

0

10,000

Bets

$5 · 22

$10 · 13

$20 · 5

2,000

4,000

6,000

8,000

Bet 10,000$5: 22 remaining$10: 13 remaining$20: 5 remaining

$5 per bet22 of 100 reached 10,000 bets$10 per bet13 of 100 reached 10,000 bets$20 per bet5 of 100 reached 10,000 bets

Stake	Busted by 1,000	By 3,000	By 5,000	By 10,000	Median bust	Survivors
$5 bets	5	44	62	78 of 100	Bet 2,599	22
$10 bets	47	73	80	87 of 100	Bet 937	13
$20 bets	74	86	92	95 of 100	Bet 361	5

Same seeds, same outcomes. Doubling the stake more than halves the median time to bust. A smaller bet buys time to encounter rare wins; it does not change the game’s return or guarantee a profit.

Compare all three stakes⌄

Stake	Busted by 1,000	By 3,000	By 5,000	By 10,000	Median bust	Survivors
$5 bets	5	44	62	78 of 100	Bet 2,599	22
$10 bets	47	73	80	87 of 100	Bet 937	13
$20 bets	74	86	92	95 of 100	Bet 361	5

04What the advertised RTP actually refers to

If the RTP is 99.9%, why is your balance down? Because an advertised RTP is a statement about the game at scale: it is the average the math converges to over millions of rounds. It is not a description of a session, and at 10,000 bets it never had the chance to become one. The number on the label is real; it just lives at a volume only the casino ever plays.

This is the audit’s simulation pass: every one of the game’s 27 configurations, run for 1,000,000 rounds each (27,000,000 in total) on fresh seeds, using the independently verified algorithm. Every line below is one configuration finding its way to the advertised return.

Switch to the all-27 view and note where the orange line sits: it is the configuration on this page (16 rows, High risk), and after a full million rounds it still reads 100.74%; high-variance configurations are the slowest to settle. That is not the RTP being wrong: even after a million rounds this configuration is still fluctuating naturally around its theoretical return, and with more rounds it keeps converging toward 99.9%. The pooled result across all 27 lands at 99.92% against 99.90% theoretical, with zero chi-squared failures. Your 10,000-bet session lives in the first pixel of this chart.

Zoom out to a million rounds

Combined average27 configurations

ADVERTISED RTP99.9%

Reference for the dashed line

ALL 27 CONFIGURATIONS99.92%

27,000,000 rounds pooled

SIMULATED ROUNDS27M

27 configurations × 1M rounds

96%

98%

100%

102%

1k

10k

100k

1M

Simulated rounds per configuration (log scale)

Advertised RTP: 99.9%

Fair-game envelope (±3 SE)

96%

98%

100%

102%

1k

1M

Rounds per configuration · log

Advertised RTP: 99.9%

Fair-game envelope (±3 SE)

10k

100k

The pooled average of all 27 configurations: 27,000,000 rounds settling onto the advertised return, 99.92% observed against 99.90% theoretical. The shaded funnel is the envelope a fair 99.9% game stays inside (±3 standard errors, computed from the published paytables); the audited line never leaves it. Switch views to see how much a single configuration wobbles on the way there.

All 27 configurations⌄

Configuration	Rounds	Theoretical	Observed	Δ pts
8 rows · Low	1,000,000	99.90%	99.89%	-0.01
8 rows · Medium	1,000,000	99.90%	99.68%	-0.22
8 rows · High	1,000,000	99.90%	100.00%	+0.10
9 rows · Low	1,000,000	99.90%	99.90%	-0.00
9 rows · Medium	1,000,000	99.90%	99.99%	+0.09
9 rows · High	1,000,000	99.90%	99.57%	-0.33
10 rows · Low	1,000,000	99.90%	99.90%	+0.00
10 rows · Medium	1,000,000	99.90%	99.82%	-0.08
10 rows · High	1,000,000	99.90%	99.46%	-0.44
11 rows · Low	1,000,000	99.90%	99.88%	-0.02
11 rows · Medium	1,000,000	99.90%	99.92%	+0.02
11 rows · High	1,000,000	99.90%	100.09%	+0.19
12 rows · Low	1,000,000	99.90%	99.92%	+0.02
12 rows · Medium	1,000,000	99.90%	99.90%	-0.00
12 rows · High	1,000,000	99.90%	99.85%	-0.05
13 rows · Low	1,000,000	99.90%	99.91%	+0.01
13 rows · Medium	1,000,000	99.90%	99.75%	-0.15
13 rows · High	1,000,000	99.90%	100.36%	+0.46
14 rows · Low	1,000,000	99.90%	99.87%	-0.03
14 rows · Medium	1,000,000	99.90%	99.95%	+0.05
14 rows · High	1,000,000	99.90%	99.91%	+0.01
15 rows · Low	1,000,000	99.90%	99.92%	+0.02
15 rows · Medium	1,000,000	99.90%	99.92%	+0.02
15 rows · High	1,000,000	99.90%	100.00%	+0.10
16 rows · Low	1,000,000	99.90%	99.88%	-0.02
16 rows · Medium	1,000,000	99.90%	99.79%	-0.11
16 rows · High — this page’s game	1,000,000	99.90%	100.74%	+0.84

05The audit verdictWhat this means for players

✓ VERIFIED162 / 162

Seed commitments matched

Every seed matched its committed hash. Outcomes could not be altered or swapped after the bet.

✓ VERIFIED8,100 / 8,100

Live bets reproduced exactly

Recomputed from seeds alone, with zero mismatches across every live bet in the capture.

✓ VERIFIED27 million

Simulated rounds checked

The distribution matches fair binomial math, with no chi-squared failures across the simulated rounds.

✓ VERIFIED162 epochs

Seed rotation tests passed

No seed cherry-picking: rotation-timing tests passed on all 162 real seed epochs.

Michael SmargiassiAuthor · ProvablyFair.org