About

The Foresportia method:
AI, stats and transparency

Foresportia does not sell "sure picks". It publishes readable, explainable and verifiable probabilities to make football analysis clearer and more accountable.

Probabilitiesnot certainties Historyverifiable Methoddocumented
AI method Past results Research paper AI football blog
DataTeams, competitions, context and events
↓
AI modelsStatistical modelling and machine learning
↓
Probabilities1X2, exact score, BTTS, Over/Under...
↓
VerificationReal results, history and calibration

Why Foresportia exists

The problem

Most football prediction content relies on opinions or promises of profit. On many services, it is difficult to access past predictions and objectively evaluate how reliable the method really is.

Our answer

Foresportia takes a different approach: publish verifiable probabilities based on statistical models, preserve public historical results and make model performance observable over time.

The goal is not to predict the future with certainty, but to provide a quantitative reading of football and better understand the uncertainty inherent to the game.

The project is freely accessible and primarily supported through advertising. Users can support Foresportia and remove ads through the user account page.

Who is behind Foresportia?

My name is Quentin Barbedienne, creator of Foresportia. I have worked for more than 10 years on Data & AI projects across marketing, industry, R&D, advertising, retail, banking and insurance.

I hold a PhD in fundamental physics, with a strong focus on modelling and data analysis. That research background shaped the way I approach uncertainty, calibration and measurable evidence.

I also spent 2 years working with a football club on computer vision and performance analysis. That field experience complements the scientific side of the project.

Foresportia comes from this dual background: statistical rigor and practical football context. The goal is clear probabilistic analysis, not unrealistic promises.

PhD in fundamental physics 10+ years Data & AI 2 years applied football analytics Quantitative modelling
LinkedIn Google Scholar

Our approach in 4 steps

1ObserveCollect and monitor team, competition and contextual data.
→
2ModelStatistical models and AI estimate probability distributions.
→
3PublishReadable probabilities designed for practical interpretation.
→
4VerifyCompare forecasts with real outcomes and historical performance.

Foresportia combines statistical signals and probabilistic models to estimate key markets (1X2, exact score, etc.). The May 2026 paper Beyond Top Probability documents the entropy-based confidence signal used to read forecast concentration.

  • Team dynamics and league-context modelling.
  • Scenario simulation to turn signals into probabilities.
  • Historical tracking to verify model reliability over time.
  • Transparency-first publication and educational framing.

What Foresportia is not

Not a tipsterNo guaranteed wins and no "sure bet" marketing.
Not a betting serviceProbabilities remain analysis tools.
Not magicFootball remains a high-variance sport.

For the full framework: read the methodology page.

You do not have to take our word for it

Read the methodology, inspect past results and explore the statistics: the goal is to keep the process observable and verifiable.

Past results Read the method Statistics