> For the complete documentation index, see [llms.txt](https://docs.uniocean.io/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.uniocean.io/rewards/overview.md).

# Reward Overview

Understand Uniocean rewards and the UCPRA algorithm.

![Rewards](https://2863428870-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2F80Mih2wy0x6vMgmpFKdI%2Fuploads%2Fgit-blob-6786b1346e39d43ecfdecac5705fe97f77b5fef6%2Fdoc-card-04-rewards.png?alt=media)

Uniocean uses the **Uniocean Community Price Reward Algorithm (UCPRA)** to distribute rewards after a market settles. UCPRA is designed to be fair, transparent, deterministic, and configurable at the event level.

{% hint style="warning" %}
UCPRA supports dynamic event configurations, including entry counts, entry fees, winner coverage, Ranker counts, and payout-curve parameters. Always use the configuration published for the specific event; the examples on this page are illustrations, not a universal payout promise.
{% endhint %}

## Reward Pool

An event fee of **3%–10%** may be deducted from total entry fees before the reward pool is formed. The remaining amount — the reward pool — is distributed to rewarded participants according to that event's published reward configuration. The applicable event fee is shown on the market page.

![Reward Pool Split](https://2863428870-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2F80Mih2wy0x6vMgmpFKdI%2Fuploads%2Fgit-blob-cda6cc9f52961537ed0197d43aa42202df98e905%2F09-reward-pool-split.png?alt=media)

## Who Is Rewarded?

UCPRA does **not** use one universal reward shape for every event. Each event publishes its own reward configuration, which determines:

* what share of participants receive rewards,
* how many top forecasts qualify as Rankers,
* and how rewards are distributed across winning positions.

The standard configuration used across the platform is:

| Configuration     | Typical Use                        | Typical Winner Coverage | Typical Bucket Shape                    |
| ----------------- | ---------------------------------- | ----------------------: | --------------------------------------- |
| Regular / Public  | Standard public events             |    **51%** of forecasts | 2% Ranker tier and a 1% bucket gradient |
| Mass Distribution | Long-form or broad-coverage events |    **70%** of forecasts | 1% gradient with a 1% Ranker tier       |

Participants outside the rewarded range receive no payout.

A market may be cancelled when participation is at or below the event's published threshold, currently **14 forecasts or fewer**, or under another applicable platform cancellation process.

## Reward Groups

| Group          | Who                                                              | Reward Type                                            |
| -------------- | ---------------------------------------------------------------- | ------------------------------------------------------ |
| Rankers        | Best forecasters (count depends on event size and configuration) | Premium top rewards                                    |
| Bucket Winners | Remaining rewarded forecasts below the Rankers                   | Reduced rewards across the remaining winning positions |

### Standard public behavior

* Standard public events use a **51%** win rate.
* The exact values used for any live event should always be taken from that event's published reward configuration.

### Custom Configured Events

Some events use custom reward settings for special formats, campaigns, or longer-duration distributions.

One supported profile rewards **70%** of forecasts. Both standard profiles use a **1% gradient**; the exact Ranker and payout parameters should be read from the event's published configuration.

## Flow

![Reward Flow](https://2863428870-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2F80Mih2wy0x6vMgmpFKdI%2Fuploads%2Fgit-blob-fc267186841102b8a69f534ab9da920f042d60e0%2F24-reward-flow.png?alt=media)

## Algorithm Properties

| Property      | Description                                           |
| ------------- | ----------------------------------------------------- |
| Deterministic | Same inputs always produce exactly the same rewards   |
| Complete      | Every token in the reward pool is distributed         |
| Fair          | Higher accuracy always means higher reward            |
| Verifiable    | Anyone can recompute results using public data        |
| Transparent   | Full settlement data published in Event Audit Reports |

## Reward Calculation After Settlement

The reward calculation happens as part of the automated settlement pipeline:

1. All forecasts are ranked by accuracy
2. The event's reward configuration determines how many participants are rewarded
3. Rankers are assigned premium shares from the reward pool
4. Bucket Winners are assigned rewards from the event's configured payout band
5. Results are validated by independent validator nodes
6. Merkle tree is generated for on-chain verification
7. Rewards are auto-claimed to rewarded wallets

***

## Related Pages

* [Reward Distribution](/rewards/reward-distribution.md)
* [Settlement Process](/settlement/settlement-process.md)