How 38-0 works

38-0 asks a deliberately fictional question: if you drafted an all-time football XI, how might one 38-match league season unfold? It combines a curated player pool, formation roles, a chemistry model, opponent strengths, and random match events to produce one coherent campaign.

It is a game, not a scouting system or prediction service. The model cannot recreate the tactical, medical, economic, and historical context of real football. Its purpose is to make draft choices legible and replayable while leaving enough randomness for the perfect 38-0 season to remain uncertain.

The selectable competitions are the Premier League, La Liga, Serie A, Bundesliga, and Ligue 1. You can draft within one competition or use All leagues to mix club-season squads across the five pools. League selection also determines the source of opponents used to build the completed team's schedule.

Each player record has a club, season, league, era, primary position, optional alternate positions, an overall rating, and gameplay attributes. The pool is curated and may not include every player or every season. Ratings are editorial game inputs, not official values licensed from a league, club, FIFA, UEFA, EA Sports, or another publisher.

A formation contains 11 ordered slots. Each slot expects a football position such as GK, CB, CDM, CM, CAM, LW, RW, or ST. A player's primary position and explicitly recorded alternate positions count as exact fits. A compatibility map allows plausible coverage between related roles when an exact option is unavailable.

The three engine weights are 100% for an exact role, 70% for a compatible role, and 30% for an incompatible assignment. The draft normally prevents an incompatible player from being placed, but the third tier remains in the engine as a defensive fallback. These values describe role fit, not a new overall player rating.

Position fit supplies the largest controllable part of team chemistry. The model also considers era diversity, gives a small contribution for players marked as legends, and rewards groups of at least three players from the same club. Mixing many eras can reduce chemistry, while natural roles and club groups can raise it.

Chemistry is bounded before it scales the outfield team units. That is why the interface shows both a slot's fit percentage and a separate team chemistry bar: the first explains one assignment, while the second summarizes relationships across the XI.

Players are grouped by their recorded primary roles. Attack strength emphasizes shooting, pace, and dribbling for attacking players. Midfield emphasizes passing, dribbling, and physical ability. Defence emphasizes defending, physical ability, and pace. Goalkeeper strength uses the selected goalkeeper's overall rating.

The overall team number combines the four units with the average player rating, then applies chemistry to the outfield contribution. These summaries are designed for internal comparison inside 38-0; they are not intended to match another football game's ratings scale.

The season contains 19 opponents, played home and away for 38 matches. A single-league run draws from that competition's opponent pool. All-leagues mode draws from the combined pool. Fixtures are shuffled, and opponent unit strengths include limited variation around each club's stored strength.

The game does not simulate a full independent table in which every opponent also plays every other opponent. It simulates the user's 38 fixtures and estimates a final position from the resulting points total. That trade-off keeps the game fast while still producing a familiar season summary.

Each match compares one side's attack with the other side's defence and accounts for home advantage. The strength gaps are converted into expected-goal rates. The simulator then uses Poisson sampling to draw a possible goals-for and goals-against scoreline from those rates.

Sampling is the source of match uncertainty. A stronger side is more likely to win but is not guaranteed to do so. Repeating the same XI starts a new stochastic season and can therefore return a different record.

Scorers and assists are distributed across the XI using player quality, position, and relevant attacking or creative attributes. Goalkeepers receive clean-sheet credit when the team concedes no goals. These events are generated after the team scoreline and make the final campaign easier to read and share.

Wins earn three points and draws earn one. The result screen reports wins, draws, losses, goals, points, awards, and an estimated league finish. The position estimate is a points-based interpretation, not a reconstructed real league table.

• The model does not simulate injuries, substitutions, transfers, suspensions, fatigue, or changing form.
• It does not reproduce a manager's tactics or how roles changed across football eras.
• Historical players did not face the mixed squads and opponent pools created by the game.
• Ratings and club strengths are curated and can be corrected as the data improves.
• Results are entertainment and should not be used for betting, recruitment, or real-match prediction.