How long does a Polymarket market take to resolve?

Polymarket's documentation gives about two hours after a proposal if nobody disputes it. If the proposal is disputed, it describes a debate period of 24 to 48 hours and a UMA vote of about 48 hours, and puts a disputed resolution at 4 to 6 days in total.

What happens to my position if a Polymarket market resolves 50/50?

According to the documentation, if the outcome is unknown or neither outcome applies, the market resolves 50/50 and each token redeems for $0.50. If you hold the other side of the same event on Kalshi, that leg settles on Kalshi's own rules, so the two legs can stop offsetting each other.

What is resolution mismatch in arbitrage?

It is when two contracts that look like the same event are settled by different rules: a different source, cutoff time, treatment of overtime or postponement, or a different oracle. The hedge only holds if both contracts resolve the same way on every outcome.

Polymarket UMA Disputes & Resolution Risk

A cross-platform arbitrage is only a hedge if both contracts settle the same way. The price gap you can see on a screen says nothing about that. This article covers how Polymarket resolves markets with the UMA optimistic oracle, what a dispute does to your timeline, how Kalshi settlement differs, and a checklist for catching a resolution mismatch before you pair two markets. The Polymarket details come from its resolution documentation and the Kalshi ones from its settlement documentation; both can change, so read the current pages before relying on a number.

How a Polymarket market resolves

Polymarket's documentation says prediction markets use the UMA Optimistic Oracle, and that Chainlink TWAP resolves the separate up-down markets. For a UMA market the steps are:

Rules first. Each market has pre-defined rules naming the resolution source, the end date and how edge cases are handled. The docs say to read the rules, because the title only describes the question and the rules define how it resolves.
Proposal. Anyone can propose an outcome by posting a bond, typically $750 in pUSD according to the docs.
Challenge period. After a proposal there is a two-hour window in which anyone can dispute it. With no dispute, the market resolves.
Dispute. A dispute starts a new proposal round. If the second proposal is also disputed, the question escalates to UMA's Data Verification Mechanism, a vote by UMA token holders.

The documentation lists three possible flows: no dispute (about two hours), one dispute (a second proposal is accepted), and two disputes (resolved by DVM vote). For a disputed market it gives a debate period of 24 to 48 hours, a vote of about 48 hours, and 4 to 6 days in total.

The outcomes a vote can produce

Besides the two obvious results, the documentation describes "Too Early", where the event has not concluded and the proposal is rejected, and "Unknown/50-50", which it calls rare: neither outcome applies, and each token redeems for $0.50. It also says Polymarket may publish an "Additional context" clarification onchain after trading starts, which cannot change the fundamental intent of the question but which voters should consider.

How Kalshi settles

Kalshi's documentation says that on a Yes outcome Yes holders receive $1 per contract, and on a No outcome No holders do. It says markets typically settle shortly after expiration, but timing can vary with the market type, data source availability and manual review. That page does not describe the public proposal-and-dispute steps that UMA has, so check how Kalshi determines the outcome for the specific series you trade, using the rules printed on the market.

What a dispute does to a two-leg position

Take an illustrative pair, not a live trade: YES on Polymarket at $0.45 and NO on Kalshi at $0.50, so $0.95 for a $1.00 payout if both settle normally. Three things can break the picture:

Timing. One leg can settle and free its capital while the other waits through a dispute of several days. Your capital is tied up longer than your return calculation assumed.
A 50/50 result. If Polymarket pays $0.50 per token and Kalshi's No leg pays either $1.00 or $0, the position pays $1.50 or $0.50 against a $0.95 cost, and it is no longer a hedge. This is arithmetic from the two documented rules, not a prediction that it will happen.
Different answers. If the two platforms read the same real-world event differently, both legs can lose or both can win. That is the case the whole strategy is designed to exclude.

A checklist before you pair two markets

Same resolution source? An official body, a named website, a specific data release. If the sources differ, note which one is more likely to change.
Same cutoff? The exact date, time and time zone, and whether an event after the cutoff still counts.
Same thresholds? "Above" against "at or above", rounding, which version of a published number counts, preliminary or revised.
Same edge cases? Postponement, cancellation, overtime or extra time, a tie, an abandoned game, a withdrawn candidate. See sports arbitrage on prediction markets for the sports versions.
Complementary sides? Confirm you hold opposite outcomes of the same event and not two versions of the same bet.
What happens if it is void or unknown? Read the clause on both platforms and work out the payout of each leg in that case.
How long can settlement take? Allow for the dispute timeline on the Polymarket side and manual review on the Kalshi side when you size the position.
Any clarification posted? Check for "Additional context" on the Polymarket market before you rely on the original text.

Why gaps that persist are suspicious

A large gap that stays open for hours is more often two markets that are not equivalent than a mispricing that nobody has taken yet. Treat it as a reason to read both rule sets again, not as a bargain. We discuss this in is prediction market arbitrage really risk-free.

Where Arbitrage Agent fits

Arbitrage Agent uses AI matching with every pair validated, including a check of the resolution rules on both sides, and then shows the result in a live feed with a step-by-step guide for both legs. You place the orders yourself, and the final check of the rules is still yours. See the plans or start with the Starter plan.

Key terms: Resolution Risk · Two-Leg Execution · Surebet

Polymarket UMA Disputes and Resolution Mismatch: What Happens to Both Legs (2026)