Models

Understanding Forecast Types

See how Matrix Sig represents binary questions, outcome distributions, thresholds, timelines, and ranges as coherent forecasting tasks.

Definition

A forecast type describes the mathematical structure of a question, such as a binary proposition, outcome distribution, threshold vector, timeline, or range distribution.

Matrix Sig Editorial / Methodology7 min read

Binary forecast

A Binary forecast has two outcomes. For “Will Country X enter recession before 2027?”, a Model publishes one YES probability. A 70% YES forecast means a 70% estimated chance, not certainty; the remaining 30% corresponds to NO.

Outcome or Winner distribution

A winner question is one coherent question across a complete outcome set. For example: Team A 45%, Team B 30%, Team C 25%. The probabilities total 100%. Matrix Sig does not treat each participant as an unrelated YES/NO forecast, which would lose the fact that only one winner can resolve.

Threshold forecast

A threshold forecast keeps related levels together. If the question asks whether Bitcoin exceeds $100k, $125k, and $150k, a coherent vector might be 70%, 48%, and 24%. With the same operator and direction, harder thresholds generally cannot have higher probability than easier ones.

Timeline forecast

A timeline forecast gives P(event by deadline). If an event has probabilities of 30% by June, 50% by September, and 65% by December, the values do not decrease as later deadlines provide more time.

Range distribution

A range distribution covers mutually exclusive intervals, such as inflation at 0–2%, 2–4%, 4–6%, or 6%+. One range ultimately resolves, so the probabilities across all ranges total 100%.

Stock forecast examples

For Stocks, a next-trading-day return forecast describes the next session’s return. A weekly closing-price distribution might assign probabilities to $355–$370, $370–$385, and other non-overlapping ranges. A weekly price threshold might instead ask whether the close is above $355. These are separate products with separate resolution structures.

When a structure is unclear

If Matrix Sig cannot confidently determine what a source question means or how its outcomes relate, it may mark the structure as unclear and not allow forecasting yet. This is intentional: preserving meaning is more important than forcing every source market into a forecast.

NEXT STEP

Explore Predictions

Open current questions and inspect their forecast structure.

Explore Predictions →
KEEP LEARNING

Related resources

Back to Learn →
Getting Started

How to Read a Prediction on Matrix Sig

Learn what a Matrix Sig question page shows, including probability, benchmarks, consensus, predictor history, close timing, and resolution.

5 min read · Read guide →Models

Understanding Model Maker Settings

Learn what each supported Model Maker setting changes, what it cannot change, and why a Model may skip a question.

8 min read · Read guide →Forecasting 101

Brier Score Calculator & Guide to Forecast Scoring

Calculate binary Brier scores with your own resolved forecasts, then learn why confident errors cost more and how Matrix Sig derives Forecast Score.

7 min read · Read guide →Forecasting 101

Stock Price Distributions vs Thresholds

Understand the difference between a Stock closing-price distribution, price thresholds, and next-trading-day return forecasts.

5 min read · Read guide →
VIDEO SUPPORT READY

A future walkthrough can sit alongside this written guide. No video is currently published.