Share.

    7 Kommentare

    1. Worried-Animal-4044 on

      [OC] Made with Python + Pillow. [Live scorecard, source code and methodology](https://mli3w.github.io/world-vs-model/?utm_source=reddit&utm_medium=social&utm_campaign=kickoff#outcome).

      **Data:** real World Football Elo ratings (eloratings.net) + live, de-vigged Polymarket prices for the market baseline. **Method:** 20k Monte Carlo runs through the actual FIFA 2026 bracket (Round of 32 → Final), with Dixon-Coles goals correction and rating uncertainty.

      The story is the **spread**: even Spain — the favourite — only wins about **1 in 6** simulations. No single outcome is likely; the distribution itself is the forecast.

      Every call is frozen pre-tournament and scored Brier-style vs the betting market as results land. We’ll know by July whether a transparent model can beat a liquid crowd. **Research/education only — not betting.**

    2. Account_Eliminator on

      Great visualisation.

      Wow how Germany have fallen off over the years.

    3. Germany less likely to win it all then Ecuador, Mexico or Colombia? I know we struggle a lot lately, but that bad?

    4. How is Argentina more likely to go out in round of 16 rather than 32? First is Spain or Uruguay, the following match is the likes of Egypt, usa, turkey, etc.

    Leave A Reply