196 million spins on floating dragon: a hold-and-spin classic that busted 74% of $0.50/$100 sessions
exp · 064 · 2026-06-27 · simulation-based
Run it yourself in the live simulator. All figures are simulation-based observations, not predictions. See our methodology.
what we measured
| parameter | value |
|---|---|
| provider / engine | Pragmatic Play · hold & spin, money-symbol respins |
| configs simulated | 96.71% (provider default) · 94.62% (floor of the published ladder) |
| published rtp ladder | 96.71% / 95.67% / 94.62%, a verified ~2.1pp spread the casino selects |
| volatility | high |
| hit frequency | 26.41% (~1 in 3.8 spins) |
| stakes | $0.20 / $0.50 / $1.00 per spin |
| bankrolls | $50 / $100 / $200 |
| sessions | 10,000 per stake/bankroll cell, 90,000 per version |
| spin cap | 2,000 spins per session |
| max win | 5,000x stake (a feature event) |
model inputs worth flagging: the RTP ladder and the 5,000x max are sourced; the hit frequency (26.41%) is single-source. Pragmatic does not publish a bonus trigger frequency for this game, so we modelled it (about one respin round every ~120 spins) as a documented assumption; the bonus payout distribution and base-versus-feature split are model assumptions calibrated to the published RTP and the 5,000x cap. the cross-version comparison holds all inputs constant, so it is robust; absolute bonus figures are estimates.
how long bankrolls survived
the stake sets the clock. at the $100 bankroll the default-version median ran 693 spins at $0.50. the base game ticks along with small wins, but the money-symbol respins that pay are concentrated in the feature, and most sessions run dry before a big one lands.
bust rates
bust rates within the 2,000-spin cap, 96.71% default version, 95% confidence intervals:
| $50 bankroll | $100 | $200 | |
|---|---|---|---|
| $0.20/spin | 67.5% ±0.9 | 35.2% ±0.9 | 1.6% ±0.2 |
| $0.50/spin | 87.4% ±0.7 | 74.3% ±0.9 | 46.9% ±1.0 |
| $1.00/spin | 94.2% ±0.5 | 88.2% ±0.6 | 74.8% ±0.9 |
plain reading: hold the bankroll at $100 and move the stake from $0.20 to $0.50, and the bust rate doubles from 35.2% to 74.3%. the only safe cell is $0.20 against $200 (1.6%).
the bonus wait, and what it pays
on our modelled trigger rate (about 1 in 120 spins, an assumption), the respin feature carries about 64% of the total return. the payout side (model-based estimates): the average round paid 75x stake but the median was 41x, and 57% paid under 50x. the 5,000x ceiling is the rare tail; a typical round pays around half the average.
what a finished session looks like
the final-bankroll distribution at $0.50/$100 has almost no middle. seven sessions in ten ended with under $0.50 of the original $100, busted, in effect, with the median finish near $0.47. then it leaps: the 90th percentile kept $339. the hold-and-spin format lands in the same die-or-detonate shape as the rest of our library.
the rtp version lottery
floating dragon is published at 96.71% and also at 95.67% and 94.62%, a ~2.1-point spread the casino selects. we ran the full grid at the default and the floor; same model, same inputs, only the RTP changed:
| cell | 96.71% default | 94.62% floor | delta |
|---|---|---|---|
| $0.20/$100 bust | 35.2% ±0.9 | 38.5% ±1.0 | +3.3pp |
| $0.50/$100 bust | 74.3% ±0.9 | 76.4% ±0.8 | +2.1pp |
| $0.50/$200 bust | 46.9% ±1.0 | 50.7% ±1.0 | +3.8pp |
| $1.00/$200 bust | 74.8% ±0.9 | 76.9% ±0.8 | +2.1pp |
the floor version raised the bust rate in every cell. the effect is moderate because the spread is only two points, but it is free margin for whichever operator runs it. the version is the one variable a player can actually pick, so play it where the published RTP is highest. our casino hub ranks operators by exactly that.
methodology note
we simulate models calibrated to published math, RTP, hit frequency, volatility profile, bonus behaviour, not the provider's game engine. results are sample-based observations from 195,948,592 simulated spins (180,000 sessions across two RTP versions), with 95% confidence intervals shown. hit frequency (26.41%) is single-source; the bonus trigger frequency is not published and was modelled (~1 in 120) as a documented assumption; the bonus payout distribution and base-versus-feature split are modelled. the base-game ceiling (300x) is a reasoned estimate; the 5,000x max is a feature event. casino-by-casino RTP figures are the published ladder, not statements about any operator's current configuration. model validation: floating-dragon v1, analytic calibration exact at 96.71% and 94.62%, 10M-spin checks within tolerance. slots are negative-expectation games; nothing here predicts outcomes or improves odds. corrections policy: methodology.html.
Where the max win actually comes from
64% of this game's RTP is locked inside the bonus you rarely trigger; the base game on its own returns just 34%.
A normal spin in our simulation never returned more than ~300x (€150). The 5,000x top win is a feature event, it only came out of the bonus. (base-game ceiling: model estimate)
Play the Floating Dragon demo, or stress-test it
Looking for the Floating Dragon demo or free play? A demo shows you a handful of spins. Our free simulator runs Floating Dragon across thousands of sessions and shows what actually happens to a bankroll over time: the bust rate, how long the money lasts, and the wait for the bonus. It is the demo with the math switched on.
stress-test Floating Dragon free
FAQ
Is there a Floating Dragon demo or free play?
Yes. You can play Floating Dragon in demo mode at most casinos, and you can stress-test it free in our simulator, which runs thousands of sessions and reports the bust rate and session length, the demo with the math switched on.