> For the complete documentation index, see [llms.txt](https://torch-1.gitbook.io/litepaper/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://torch-1.gitbook.io/litepaper/system-mechanics/bonus-pool.md).

# Bonus pool

Bonus Pool defines how much of the reserve is available for distribution to winners at any moment. It is calculated using an Exponential Moving Average (EMA) of recent losing stakes, but never exceeds the reserve surplus:

$$
\text{BonusPool}*t =
\begin{cases}
\min(\text{EMA}*t,\ R\_t - R*{\text{target}}), & \text{if } R\_t > R*{\text{target}} \\
0, & \text{otherwise}
\end{cases}
$$

Where:

* `Rt`: Current Reserve Liquidity
* `Rtarget`: Reserve's Target Level
* `EMAt`: Exponential moving average of recent losing stakes

The EMA is updated each timestep using:

$$
\text{EMA}*t = \alpha \cdot L\_t + (1 - \alpha) \cdot \text{EMA}*{t-1}
$$

Where:

* `Lt`: Losing stake inflow at time t
* `α`: Smoothing factor, initially set to 0.2

The payout pool increases when recent losses are high, allowing the system to reward winners more generously during periods of heavy market activity. However, no bonuses are paid if the reserve is at or below its target level.