World Cup 2026 AI Quant: baseline prediction and evaluation workflow

Research applicability

Experimental predictive modelling and evaluation in sports research.

Getting started

Read the linked primary sources and the stated adoption requirements. Reuse suggestions are editorial and untested; no setup recipe or trial was executed.

Maintenance

Sources last checked on 2026-10-19. Future availability, maintenance, provider terms and compatibility were not verified.

License

The README states MIT; the licence file and data rights were not audited.

Workflow & human oversight

Intended L2 for analyst-orchestrated predictive research: validate data, choose baseline features, train statistical/ML models, evaluate splits, calibration and uncertainty, and report limitations. Non-agentic modelling does not establish L3/L4, prediction advantage or evaluated tournament outcomes.

Workflow level is not an evidence score; higher is not better evidence.

Approach

The README prescribes dataset validation, interpretable ratings/form/context features, baseline training and evaluation with split/metric/calibration records. The short methodology note asks analysts to document targets, feature windows, uncertainty and model assumptions.

Limitations

README and short methodology note only. Data, notebooks, source modules, tests and external resource portals were not inspected or executed. Structure, licensing and CI statements are documentation claims. No complex-model result, calibrated forecast, betting or financial recommendation is established. Moving HEAD documents are not commit-pinned or an atomic version. Promotional links are not endorsed.

Reproducibility

Source text was checked only within this scope: Full README and a short methodology note. No software, dataset, experiment or result was independently reproduced.

Adoption requirements

Untested suggestion: choose a legitimate dataset, record targets, time splits and baseline metrics, and check leakage and uncertainty before an experiment.