Resolution: How a Market Decides Who Was Right
Resolution is where more money is lost than in forecasting. What a resolution source is, how disputes work, and the rule wording that catches people out.
The forecast is the interesting part. Resolution is the part that pays you. It is also, reliably, where people lose money they thought they had won.
What resolution is
Resolution is the process that converts an event in the world into a settlement price of one dollar or zero. It has three components, all fixed before trading opens.
The condition. The precise thing that must be true. Not the title of the market — the sentence in the rules.
The source. What decides whether the condition is met. An official statistical release, a named authority, a certified result, a specified publication, or an oracle process.
The timing. When the source is consulted, and what happens if it has not published by then.
Why the title is not the rule
Market titles are written to be readable. Rules are written to be unambiguous. When they disagree, the rule wins, every time.
A market titled "Will the Fed cut rates in March?" resolves on whatever the rules say — most likely the target range announced at a specific scheduled meeting, not an intermeeting move, not a different committee's action, not a cut announced in March but effective later. Each of those distinctions has decided a real market somewhere.
The discipline is simple: before you form a view on the probability, read the rule and restate it in your own words. If you cannot restate it, you do not know what you are pricing.
Sources, and their failure modes
Official statistics. Clean, scheduled and usually unambiguous, with one catch: revisions. A rule that specifies "as first published" and a rule that specifies "as of a later date" can settle differently on the same data. Check which one you have.
Certified results. Reliable but slow. A result can be known on the night and certified weeks later, and if the rule keys on certification your capital is tied up until then.
Named publications. Introduces the risk that the publication changes its methodology, stops publishing, or publishes something the rule did not anticipate. Good rules include a fallback. Read whether yours does.
Oracles. On-chain venues typically resolve via an optimistic process: someone proposes an outcome, a window opens for challenges, and a dispute escalates to a wider vote. It works well for clear questions and is genuinely stressed by ambiguous ones. How Polymarket resolves a market covers the mechanism, and when a resolution is disputed covers what happens when it goes wrong.
The wording traps
A short list of the phrases worth slowing down for.
"Officially announced" — an announcement by whom, and what counts as official? Reporting by a credible outlet is not always an announcement.
"By December 31" — in which time zone, and is the 31st inclusive?
"Exceeds 3%" — is 3.0% exceeding 3%? Usually not. Is 3.04% reported as 3.0% exceeding it? Depends whether the rule keys on the published figure or the underlying one.
"Consecutive" — consecutive reports, or consecutive months, and what happens to a month with no report?
"Or any successor" — good drafting, and its absence is a real risk on long-horizon markets.
Early resolution and the unexpected ending
Some markets resolve early when the outcome becomes impossible or certain. Others run to their stated date regardless. A market that can resolve early is worth more than an identical one that cannot, because your capital comes back sooner — which is the entire subject of capital lockup.
Some markets are voided, refunded at the traded price or cancelled. The circumstances are written into the rules, and they usually involve the source failing rather than the event being unclear.
Cross-venue resolution differences
This is the quiet reason two venues can price the same event differently and both be right. If Kalshi settles on the first published figure and a Polymarket contract settles on a different source or a different date, they are not the same contract, and the difference between their prices is not an edge — it is a description of the difference between the rules.
This site grades every cross-venue match on how equivalent the two contracts actually are, and downgrades pairs whose resolution terms differ. Equivalence grades explains the grading, and two markets, one event, different rules gives worked examples of where it bites.
The habit
Read the rule before the price. Restate it. Check the source and the date. Then look at the number. It takes ninety seconds and it is the highest-return ninety seconds available in this entire field.
Next
A short history of prediction markets puts the modern venues in context, or skip ahead to your first week watching markets for a practical routine.
Educational content only. Nothing here is financial advice, a solicitation, or an offer to trade. Prediction markets can move against you and a position can lose its entire value. Where this guide describes a difference between venues it describes a theoretical cross-market edge in observed prices, which is not a profit: fees, spreads, capital lockup, transfer time and differing settlement rules all come out of it first.
Keep reading
Prediction Market Basics · 3 min readWho Is on the Other Side of Your TradeSomebody took the opposite view, or was paid to be indifferent. Knowing which one you are dealing with changes how you read a price and a spread.
Prediction Market Basics · 4 min readA Short History of Prediction MarketsFrom nineteenth-century election betting on Wall Street to regulated event exchanges and on-chain venues, with the research that made the idea respectable.
Prediction Market Basics · 4 min readHow to Read a Prediction Market PagePrice, volume, open interest, the book and the history. What each number on a market page is telling you, and which ones deserve your attention first.
Cross-Market Gaps · 3 min readWhy Two Venues Price the Same Event DifferentlyA price gap between two prediction markets is usually not an error. Here are the structural reasons the same event trades at different numbers in different places.