The question is not the contract. The criteria are.

A market’s headline is a question. “Will the agency publish the report before the end of the month.” It reads as though the world settles it.

It does not. Underneath every event contract sits a block of rules naming a resolution source, a deadline, and a list of edge cases. That text is the contract. The headline is a summary of it, and summaries lose things.

Most of the avoidable losses in event contract trading come from reading the headline, forming a view about the world, and never checking whether the rules underneath measure the thing you formed a view about.

Three fields decide almost everything

The source. Rules usually name a specific authority — an official release, a named site, a particular agency’s own announcement. If the source is one outlet and the story is broken by a different one, the market has not resolved. You can be right about the event and hold a contract that has not moved, for days, because the designated source has not spoken.

The deadline. Almost every set of rules carries a cutoff, and the cutoff is frequently more specific than the headline. A month named in the title can mean a calendar month, a date, or a time of day in a stated zone. What happens if the cutoff passes with nothing announced is also specified — normally the contract resolves No rather than staying open. A question phrased as “will X happen” is often, in practice, “will X be confirmed by this source before this timestamp”, and those are different propositions with different probabilities.

The edge cases. This is where the drafting effort goes and where the careless reading costs most. Postponements, partial outcomes, withdrawals, revised figures, ties, and events that technically occur but are reported differently than expected — the rules say what each becomes. If the rules are silent on a case you can imagine, that silence is itself information about how much resolution risk you are taking on.

Read those three before you form a price. Not after.

What happens when nobody agrees

Your Call evaluations trade Polymarket, so it is worth knowing how resolution runs there specifically.

Resolution is not decided by the venue in the way a broker decides a settlement price. It runs through UMA’s optimistic oracle. Anyone may propose an outcome by posting a bond. A challenge window opens — two hours, per the documentation at the time of writing — and if nobody disputes the proposal within it, the proposal stands and the market settles. Winning shares pay one dollar. Losing shares pay nothing.

If someone disputes, the process escalates: a second proposal round, and past that, a vote by UMA token holders, with a debate period in between for evidence. An escalated dispute takes days rather than hours.

Two outcomes of that process are worth holding in mind because they do not exist in ordinary markets. A dispute can return Too Early, meaning the event is judged not yet determined. And it can return the equivalent of unknown, in which the market settles 50/50 and every share on both sides redeems at fifty cents.

That second one deserves a moment. It means a contract you bought at 30 because you were confident, and which the world then settled in your favour, can still pay you fifty cents if the adjudication is judged ambiguous. Being right about the event is necessary. It is not sufficient.

Prices are about adjudication too

This is not a theoretical worry, and there is now published work on it. In Economics Letters (volume 268, 2026), Guangzhe Wang studied 233 disputed markets settled through UMA’s token vote, under the title “Do prediction markets price events or adjudication?”

The framing is the useful part: in an oracle-settled market, your cash flow is fixed not by the event but by the outcome the resolution mechanism certifies, so the price can come apart from the event itself. Wang’s data covers a window where the real-world result was already known but the payout was still pending, which isolates adjudication risk from event risk. The finding is that prices in that window carried information about which way the vote would go — the claim that a neutral audit called correct traded at a visible discount in the markets the oracle went on to overturn.

Wang also reports that overturns happened in something on the order of 3 to 12 percent of that disputed sample, depending on how the cases were coded. Note the denominator: that is a share of disputed markets, which are themselves a small minority of all markets. It is not a claim that one in ten contracts is decided wrongly. But it does say plainly that the adjudication layer is a real source of variance and that other traders are already pricing it.

What to do with this inside an evaluation

Three practical consequences, given how a Your Call evaluation is scored.

Resolution ambiguity is a position-selection filter, not a risk to manage after entry. Once you hold the contract you have no lever over how the rules are written. The whole of your control sits before the order.

Slow resolution costs you more here than elsewhere. Every event contributes at most 0.5% of account size toward a 10% Equity Growth Target, so the binding resource inside a 30-day window is the number of events that finish, not the size of any one of them. A contract stuck in an escalated dispute for a week is occupying capital and returning nothing. Markets must resolve within 60 days to be eligible at all; inside an evaluation, far sooner is better.

Prefer questions a machine could settle. A closing level against a named index, an official figure from a named agency on a scheduled release date, a result with an undisputed scoreboard. The narrower the discretion in the resolution text, the smaller the gap between being right and being paid.

None of this makes resolution risk disappear. It is a permanent feature of contracts that settle against the world rather than against a price. The point is that it is legible in advance, in a block of text most people scroll past, and reading it is free.

Evaluations are conducted on simulated accounts. Not financial advice.