How the ratings work

Squads come from Wikidata: each league season's participants, each player's club memberships with dates, positions, birth dates, and — where recorded — appearances and goals per club stint. Text facts only: no images, crests, or kits. Licensed CC BY-SA from the Wikimedia community.

1 · Fame curve.A player's base rating comes from how widely documented they are (the number of Wikipedia language editions covering them) on a logarithmic curve: 50 + 5.5·log₂(sitelinks), with a bonus beyond 64 editions. Fame tracks footballing weight surprisingly well across eras.

2 · League adjustment. Each league-era carries a small offset so a mid-table star in a stronger league outrates one in a weaker league.

3 · Age at that season. Ratings peak between 24 and 30 and taper on both sides, so a 19-year-old future legend is not yet his peak self, and a 35-year-old icon is past it.

4 · Stint performance.Where Wikidata records appearances and goals for a player's spell at a club, fringe stints are penalized (few appearances per year) and prolific ones rewarded (goals per appearance, weighted by position). This is why a legend's forgotten half-season at the wrong club rates honestly low.

5 · Anchors. About 150 all-time greats carry hand-set peak ratings (Messi 97, and so on) that bound their curve from above and below — the model fills in every season around those anchors.

An XI becomes two numbers: attack (forwards weigh most, then midfield, then defence) and defence (goalkeeper and back line weigh most). Chemistry — the share of player pairs that link by club, nation, or era — multiplies both by up to 6%.

Matches are simulated with a Dixon-Coles double-Poisson model, the standard in football analytics: expected goals scale exponentially with the attack-vs-defence gap, home advantage is worth about a third more goals, and a correction term keeps low-scoring draws realistic. Every match is drawn from a seeded random stream, which is why any season can be replayed identically from its seed — the "verify fairness" button does exactly that.

Ratings are opinions computed from public signals, not measurements. Year-precision transfer dates can bleed a player into a neighboring season at their old or new club. Side-specific positions (left vs right) are missing for many players, so the game offers the whole line where the data is silent. Wikidata coverage thins for early-90s squads of smaller clubs. Corrections are welcome — see about & contact.

The aggregate squad ratings are published as an open dataset on the open data page (CC BY-SA, attribution: 38nil.app).