> For the complete documentation index, see [llms.txt](https://docs.uma.xyz/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.uma.xyz/protocol-overview/how-does-uma-resolve-prediction-markets.md).

# How does UMA resolve prediction markets?

The flexibility of UMA makes it a natural fit for resolving prediction markets which serve a wide range of markets with different resolution requirements

The major prediction markets that UMA serves integrate UMA in the following way:

#### Market Lifecycle

An oracle request including the market rules is made at the time of market creation, the request proceeds through the oracle lifecycle as described in [How does UMA work?](/protocol-overview/how-does-umas-oracle-work.md), and the market resolves as per that request unless one of the below cases occurs:

* if the first request for a given market is disputed, a 2nd request is made with the same rules. The settlement of the prediction market then ignores the 1st request and only resolves as per the 2nd request. If the 2nd request is disputed, the market will resolve as per the 2nd request. This acts as a form of escalation and prevents minor one-time proposer or disputer errors from slowing down the market resolution process with dispute resolution.
* if a market's 2nd or greater oracle request resolves as TOO EARLY (aka P4, learn more [here](https://blog.uma.xyz/articles/what-is-p4)), a subsequent oracle request is created and the market stays open and resolves as per the latest request. &#x20;

#### Rule Clarifications and Updates

The market rules specify how the integration can provide updates or clarifications to the rules in between market creation and settlement

#### Proposer Whitelist

Requests use the [ManagedOptimisticOracleV2](/developers/managedoptimisticoraclev2.md) to specify a proposer whitelist of addresses that are able to propose the request. Learn more at [Default Proposer Whitelist](/using-uma/default-proposer-whitelist.md).

#### Challenge Period Extensions

Requests that use the [ManagedOptimisticOracleV2](/developers/managedoptimisticoraclev2.md) specify a minimum challenge period. Risk Labs operates an automated proposal review system that uses LLMs and market data to flag proposals. Flagged proposals have an extended challenge period that allows additional time for all UMA partipicants to review and potentially dispute the proposal. The extended challenge period ends after the Risk Labs team has reviewed the proposal. This feature allows challenge periods to be right sized to the proposal. For example, an unflagged sports game proposals can safely resolve after a short minimum challenge period (e.g. 15 minutes) and complex or subjective geopolitics proposal can be extended past the default 2 hour minimum challenge period.


---

# Agent Instructions
This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com.

## Querying This Documentation
If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question.

Perform an HTTP GET request on the current page URL with the `ask` query parameter, and the optional `goal` query parameter:

```
GET https://docs.uma.xyz/protocol-overview/how-does-uma-resolve-prediction-markets.md?ask=<question>&goal=<endgoal>
```

`ask` is the immediate question: it should be specific, self-contained, and written in natural language.
`goal` is optional and describes the broader end goal you are ultimately trying to accomplish on behalf of the user. GitBook uses it to tailor the answer towards what is most useful for that goal.

The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.
