# Prediction Markets — Mechanics, Limits, and Why Resolution Criteria Is the Real Product

Prediction markets are exchanges where participants trade contracts that pay out based on the
outcome of real-world events. A contract priced at 34 cents implies roughly a 34% market-assessed
probability of the event resolving "YES." Major venues include **Polymarket** (on-chain,
crypto-settled), **Kalshi** (CFTC-regulated U.S. exchange), and bookmaker surfaces such as
**BetOnline**.

This page is the BETINT reference on how prediction markets actually behave as an information source,
including where the "wisdom of crowds" story breaks down.

## What prediction markets are

- **Structured, time-stamped probability data.** Unlike opinion or punditry, every price is a
  number attached to a moment, producing a continuous record of how collective belief moved.
- **Incentive-aligned forecasting.** Participants are paid to be right and punished to be wrong, so
  prices tend to aggregate dispersed knowledge faster than surveys or news cycles.
- **A real-time geopolitical tracker.** For fast-moving events, markets frequently move ahead of
  newsrooms (see the February 2026 Iran strike, where markets led reporting by ~90 minutes).

## What prediction markets are NOT

Prediction markets are sold as **truth machines**. In conflict and high-ambiguity markets, they
often operate more like **definition machines**. They do not price "what happened." They price
**what can be established under the platform's resolution criteria.**

- They are **not** neutral oracles of reality. They are adjudicated contracts.
- They are **not** immune to manipulation, insider activity, or narrative-shaping.
- They are **not** retention engines like sportsbooks; their value is informational, not recreational.

## Resolution criteria is the real product

> "If money settles on definitions, then the edge is not just who predicts events first. The edge is
> who understands how events will be named, sourced, and resolved first."

Consider a Polymarket contract like "US forces enter Iran by [date]?" It looks crisp, yes or no.
Then you read the fine print:

- Special operations count; intelligence operatives do not.
- Terrestrial entry counts; maritime and aerial entry do not.
- Diplomatic and advisor entry does not count.
- Resolution comes from a "consensus of credible reporting."

That is not a simple binary. It is a **stack of hidden adjudication layers**, and every layer is
fragile in wartime, when information is delayed, managed, leaked, denied, and reframed.

## Event reality vs. resolution reality

The core split every serious analyst must respect:

- **Event reality** = what actually happened on the ground.
- **Resolution reality** = what can be established under the platform's criteria.

These overlap, but in conflict markets they are not identical. The bigger the gap, the more the edge
migrates away from **prediction skill** and toward **adjudication skill**: understanding the
rulebook, the sourcing, and the settlement path. In this category, **the rulebook is the trade.**

## Failure modes BETINT watches for

- **Threshold theater.** When a contract hinges on a hard number (for example, a "major ground
  offensive" defined as more than 1,000 troops), market discourse drifts from strategic analysis
  into crowd-sourced headcount arguments. The incentive design distorts the question.
- **Consensus capture.** When resolution depends on "credible reporting," shifting the reporting can
  shift the settlement, turning the evidence environment itself into an attack surface.
- **Ambiguity concentration.** Outsized, repeated profits tend to cluster where information
  asymmetry, narrative volatility, and resolution ambiguity all peak at once.

## Why this matters for operators and analysts

Operators who want durable trust cannot treat resolution criteria as legal fine print or formatting.
In an adversarial information environment, **resolution mechanics are core market infrastructure.**
Analysts who ignore them mistake a narrative-settlement game for a clean forecast.

## Related BETINT writing

- "What Prediction Markets Are Not" - https://pr3diction.substack.com/p/what-prediction-markets-are-not
- "Will Polymarket US Actually Still Be Polymarket?" - https://pr3diction.substack.com/p/will-polymarket-us-actually-still
- "Get Ready For A Lot More Prediction Markets... Kind Of" - https://pr3diction.substack.com/p/get-ready-for-a-lot-more-prediction
- "Resolution Criteria Is the Real Product" - https://www.bettingintelligence.ai/research