> ## Documentation Index
> Fetch the complete documentation index at: https://docs.sporttoken.app/llms.txt
> Use this file to discover all available pages before exploring further.

# Fees and risk

# Fees and Risk Controls

We use a two-part fee model plus strict exposure caps to keep the vault profitable over time.

## 1) Baseline platform fee

* 0.3% on all transactions across the platform.

## 2) Risk-based fee (linear with exposure)

* When a bet increases vault risk on a side, we charge a fee proportional to the added risk.
* We target Kelly-style optimal sizing: fee ≈ added risk as a % of vault.

### Example: first bet on a side

* Vault size: \$100,000.
* Market: Packers -150 / Lions +150, zero existing exposure.
* Alice bets $1,500 on Packers. Vault risk: ~$1,000 (1% of vault).
* Fee charged ≈ 1% → Alice effectively bets $1,485 to win ~$990.
* The fee curve moves from 0% to \~2% after this bet; Alice pays the average over that move.

### Example: subsequent bet on same side

* After Alice, vault risk on Packers is \~2% on Lions side.
* Bob wants another \$1,500 on Packers. That would push risk to \~4%.
* Average fee over that move ≈ 3% ((2% + 4%) / 2) → applied to Bob’s stake.

## 3) Hard exposure cap (per event)

* We never willingly risk more than \~1.5% of the vault on a single event from new bets.
* If a new bet would exceed that cap, we partially fill only up to the cap (e.g., Bob can only place \~\$750 in the example).
* Positions can exceed 1.5% only if odds move in our favor after acceptance (P\&L drift).

## Fee recycling

* Any market-specific fee above the 0.3% base is made available to users who take the opposing side from us (flattening our risk).

> Proofs and deeper math: formal Kelly-style derivations will be published separately.