World Cup 2026 Host Nations: A Retained-Simulation Audit
Publisher: Football Sim Lab · Published · Corrected
Correction, 18 October 2026: The earlier June-labelled estimates could not be attributed to a retained input snapshot. They have been withdrawn. This article now reports counts from named stored output files, not verified forecasts or actual tournament results.
Appearances are not qualification guarantees
This table counts each team once per run if it appears in any quarter-final with a match ID from M97 to M100. It does not use the erroneous round labels in some old output records.
| Retained file | USA quarter-finals / 1,000 | Mexico / 1,000 | Canada / 1,000 |
|---|---|---|---|
| simulation-results.json | 0 | 0 | 0 |
| simulation-results-v2.json | 303 | 265 | 133 |
| simulation-results-v3.json | 293 | 196 | 87 |
All three hosts appear in the round of 32 in every output of the first file, yet none appears in its quarter-finals. This is a reason to inspect locked inputs, not proof that a country was guaranteed to qualify or could not progress in real football.
The earlier article presented June-labelled numbers, including guaranteed qualification and a 43% USA quarter-final figure, without an input record establishing their origin. Those claims have been withdrawn. The table reports what can be counted from the retained files.
Host advantage in the current game
The current tournament engine adds 50 rating points when host advantage is enabled. This is a game-model choice, not an empirical estimate of every home-field effect. It does not continuously account for travel, attendance, weather or confirmed lineups.
Changing opponents, the draw or a fixed score can change a host's path. To isolate host advantage, keep the other inputs constant. An undocumented historical batch cannot establish the accuracy of the current engine.
Evidence and limits
The research audit and downloadable data contain source filenames, SHA-256 checksums, run counts and aggregates. The repository script scripts/generate-research-summary.mjs recomputes the counts. All three batches contain 1,000 outputs.
Original seeds, complete rating inputs, locked results and model commits were not retained alongside these files. Filenames cannot certify a generation date. These outputs cannot establish forecast accuracy or a calibrated real-world probability. Some stored round labels are wrong, so the audit uses match IDs to identify stages. Batches with unknown assumptions must not be pooled.
Read Methodology for the current game engine and How to play for a controlled example. Official tournament information is available from FIFA. These are independent game outputs, not FIFA predictions.