100 Bonus Buys in a Row: 37.3% of Sessions End in Profit, the Median One Loses 8.7 Buys' Worth
A stream is a long row of bonus buys with the best moments cut into clips. We ran 20,000 whole sessions of 100 buys to see what the uncut version looks like.

The simulation covered 20,000 sessions, each stringing together 100 bonus purchases that on paper return 96.5% of their cost.
When the median outcome is examined, the session loses the price of 8.7 buys, a far steeper hit than the average loss of 3.5 buys calculated from the 96.5% RTP.
- Only 37.3% of 100-buy sessions finish with a profit.
- The median session loses the price of 8.7 buys.
- 79.7% of individual buys pay back less than they cost.
- A buy pays ten times its price once in 96 purchases, and 64.9% of sessions contain at least one such hit.
- The top 5% of sessions win the price of 49.2 buys or more.
One session, start to finish
The simulation strings together 100 bonus purchases and tracks every outcome. Each purchase costs a fixed amount and is priced so that, on average, it returns 96.5% of its cost. The run uses the bonus round of a high-volatility model calibrated to a 96.5% RTP, with an any-win frequency of 25.4%, a bonus round appearing on average once per 200 spins, 36% of the return coming from the bonus round and a top payout of 4,965x the stake. These figures come from our simulation model, not from any specific casino game.
Across 20,000 simulated sessions of 100 bonus buys each, 79.7% of individual buys paid back less than they cost, even though the long-run return of a buy is 96.5%. A buy that paid ten times its price or more occurred once in 96 buys, and 64.9% of the 100-buy sessions contained at least one such hit. The best single buy of the median session paid 14.9x its price; in the top 5% of sessions the best buy paid 50.9x its price or more.
When the session is taken as a whole, 37.3% of sessions ended in profit. The median session lost the price of 8.7 buys, while the average session loses the price of 3.5 buys, showing that most sessions do worse than the average because a few very large payouts pull the average up.
The unluckiest 5% of sessions lost the price of 38.6 buys or more, and the luckiest 5% won the price of 49.2 buys or more. In a typical session the single best buy is one out of the hundred, and 80% of the remaining buys lost money.
Most buys lose
In the simulation, 79.7% of buys paid back less than they cost, even though the model’s long-run return sits at 96.5%.
A win that multiplies the buy price by ten or more is a rarity. The model recorded a 10-times-price hit once in 96 buys, which translates to just over one such event per hundred purchases. Because each of those hits carries a 1,000-fold stake multiplier, they dominate the tail of the distribution.
Despite their scarcity, the large hits appear in most sessions.
The size of the best win within a session varies dramatically. In the median session, the top buy paid 14.9 times its price, whereas in the top 5% of sessions the best buy paid 50.9 times its price or more. Those outliers lift the overall average, but they do not offset the steady stream of sub-cost outcomes that define most sessions.
The practical upshot is that a player who watches only the headline win will miss the fact that 79.7% of buys lose and that the average session still ends below break-even. To gauge true performance, track each buy’s multiplier and compare the observed frequency of 10-times-price hits to the benchmark of one hit per 96 buys. A buy paid 10 times its price or more once in 96 buys.
The average session and the typical one
A 100-buy run that follows the 96.5% return rate loses the price of 3.5 buys on average. Translated into cash, that is a loss of $70 when each buy costs $20, $700 when each buy costs $200, and $7,000 when each buy costs $2,000. The arithmetic comes straight from the model’s 96.5% RTP and the 100-buy length.
The median session, by contrast, loses the price of 8.7 buys. The data state that most sessions do worse than the average, and a few very large hits pull the average up.
A $20 buy leads to an expected $70 deficit per session; a $200 buy leads to an expected $700 deficit; a $2,000 buy leads to an expected $7,000 deficit.
What a highlight clip leaves out
The clip freezes on the single buy that produced the highest payout in a 100-buy run. That buy represents one out of 100, and the accompanying commentary notes that roughly 80% of the remaining buys lost money.
The highlighted win is therefore an outlier rather than a norm.
The session’s overall result further undercuts the illusion. While the average session loses the price of 3.5 buys, the median session loses the price of 8.7 buys, showing that most sessions fall below the arithmetic mean.
A solitary 10x payout cannot offset the dozens of losing buys.
To see the full picture, add the cost of all 100 buys and compare it to the payout shown in the clip.
The median 100-buy session loses the price of 8.7 buys, so most runs end down.
The data behind this piece
Every figure in the text comes from the runs below. They are computer simulations and exact calculations, not real-money play, and they can be reproduced: the model parameters are stated and the random seed is fixed.
100 bonus buys in a row
How we ran it: 20,000 simulated sessions of 100 consecutive bonus buys on the bonus round of the high-volatility model calibrated to 96.5% RTP: any-win frequency 25.4%, bonus round on average once per 200 spins, 36% of the return comes from the bonus round, top payout 4,965x the stake. Each buy is priced so that it returns 96.5% of its cost on average; results are measured in multiples of the price of one buy. Random seed: 13608
| Sessions in profit | Median session | Unluckiest 5% | Luckiest 5% | Buys that paid less than they cost | Sessions with a 10x-price hit |
|---|---|---|---|---|---|
| 37.3% | -8.7 buys | -38.6 buys | 49.2 buys | 79.7% | 64.9% |
Questions players ask
What is the chance my 100-buy session will be profitable?
The simulation of 20,000 runs shows that only 37.3% of 100-buy sessions finish with a profit.
How much can I expect to lose on average?
By arithmetic, 100 buys at a 96.5% return lose the price of 3.5 buys. That translates to a $70 loss when each buy costs $20, $700 at $200 per buy, or $7,000 at $2,000 per buy. The figure is the expected average loss per session.
How often will I see a big win like 10x the buy price?
A single buy pays ten times its price or more roughly once every 96 purchases. Moreover, 64.9% of the simulated 100-buy sessions contain at least one such ten-fold hit.
What does the median loss mean for my bankroll?
The median outcome loses the price of 8.7 buys, which is more than double the average loss of 3.5 buys.