Chinese Football Pre-Match Prediction Analyst

Problem

Pre-match football forecasts often lean on isolated experience, historical results, and a single odds source, confusing “likely to win” with “enough probability.” This skill turns the analysis into a checkable workflow: collect lineups, injuries, schedule fatigue, weather, and referee signals, then layer tactical matchups, quantitative factors, model math, and market validation to reduce ad-hoc judgment.

How It Works

• Six-layer stack: moves from intelligence collection, tactical matchup, and feature extraction to model simulation, market signals, and expert adjustment.
• Quantitative models: uses home/away separated data, EWMA weighting, adjusted Poisson via Dixon-Coles for score matrices, and Elo for strength calibration.
• Dual-market validation: extracts implied probabilities from European odds and Asian handicap lines/odds levels, flagging Euro-Asian anomalies and model/market gaps.
• Output constraints: requires win/draw/loss probabilities, total-goals bands, top-five scores, risk notes, and disclaimers for later review and calibration.

Boundaries

Best suited for leagues or cups with reasonably complete public data and a need for structured probabilities and review logs. It does not predict certain outcomes or recommend betting; when intelligence is incomplete, lineups are unknown, or odds conflict, confidence drops and outputs should be treated as probabilistic references, not conclusions.