Polymarket resolution risk: how markets settle and what to verify

Polymarket resolution risk is the chance that a market's settlement path is unclear, delayed, disputed, or narrower than the market headline suggests. It is not about whether YES or NO is correct. It is about whether the market will resolve the way readers expect.

Orrery treats resolution risk as a first-class signal because many bad interpretations come from confusing price with settlement. A 99 cent YES share may look resolved, but unless upstream flags or UMA state confirm settlement, it is still a market state, not final truth.

Prediction market headlines compress reality into a readable question. Resolution rules define what counts. The difference can be small but decisive: official statement versus media report, calendar date versus event date, closing price versus intraday high, first 90 minutes versus extra time.

The right workflow starts with the rule. Read the source, the cutoff time, the authority, and any excluded cases. If the resolution text is missing or says the outcome will be determined by a vague consensus, the market needs a warning label.

• Look for the named source: exchange, government agency, league, court, company filing, or official account.
• Look for the timestamp: end date, settlement date, time zone, and cutoff.
• Look for exclusions: extra time, revised data, preliminary reports, cancellations, or ambiguous definitions.

Many Polymarket markets settle through UMA's optimistic oracle. A proposer asserts an outcome, then a challenge window gives others time to dispute it. If no dispute happens, the assertion can settle. If disputed, voting and finalization take longer.

For users, the practical rule is simple: proposed, challenged, or disputed states are not final. Orrery surfaces those states as resolution risk even when the price looks certain.

Rail-pinned markets are markets trading very close to 0 or 100. They often behave like resolved markets in the interface, but they are still open unless the source confirms closure.

This distinction matters for alerts, briefs, and agent APIs. A rail-pinned market may still have residual dispute risk, settlement delay, or a narrow rule that prevents the obvious outcome from resolving as expected.

• Rail-pinned YES: price is near 100 but not necessarily resolved YES.
• Rail-pinned NO: price is near 0 but not necessarily resolved NO.
• Resolved: upstream status confirms closure or a final outcome.

Orrery normalizes source flags, price, expiry, liquidity, UMA status, and market metadata into one derived status. The goal is consistency: the scanner, market page, watchlist, alerts, brief, MCP tools, and x402 API should all describe the same market the same way.

The important invariant is that future-expiry markets cannot become resolved solely because price hits a rail. Likewise, past-expiry markets without final confirmation should be called expired-unresolved, not live.

Before trusting a large move near settlement, run a short loop: read the exact market text, identify the source, check UMA status, compare price with derived status, and inspect sibling markets. If any step is unclear, the right action is verify source, not infer direction.

• Open the market page and read Resolution & Risk first.
• Check whether the source is official, statistical, exchange-based, media-based, or ambiguous.
• Check whether UMA is pending, disputed, challenged, settled, or absent.
• Compare event siblings for contradictory pricing.
• Set an alert if the outcome depends on a future official update.