Formal Specification
Papal Historic Football — Formal Specification
Vatican City · ball-team · Confidence: low
← Back to game pageImplementation Suitability
Suitability: adaptation-requiredFidelity: discretizedGap-fills: 1
# Calcio storico pontificio — AI-optimized specification
## General idea
A high-contact, hybrid football-grappling team field game. Two teams battle on a sand pitch using full-body tackling, 1v1 pin downs, and multi-directional ball progression to score into the opponent's end-zone net (*caccia*), while penalizing wide shots with half-goal concessions (*mezza caccia*).
## Formal model
- **Abstraction taken:** The continuous 80x40m sand pitch is discretized into a 2D zonal grid `[-4, +4] x [-2, +2]`. The physical grappling contests between players are modeled via 1v1 stochastic contest resolvers (pinning/holding states).
- **Player/agent count:** 2 teams of 27 agents (or reduced to 7v7 squad form: 1 GK, 2 Defenders, 2 Midfielders, 2 Forwards).
- **Turn model:** Synchronous simulation ticks (`dt = 1.0s`) over a fixed 50-minute match duration (3000 ticks).
## Formal rules
R1. Pitch setup: Team A defends goal line `x = -4`; Team B defends goal line `x = +4`. Score begins at `0.0 : 0.0`.
R2. Positional state & actions per tick:
R2.1. Unpinned agents select: `move(dx, dy)`, `pass(target_agent)`, `carry_ball()`, `shoot_goal()`, or `grapple_opponent(target_agent)`.
R2.2. If an agent executes `grapple_opponent(B)`: A stochastic check resolves whether agent B enters a `PINNED` state, immobilizing both agents for `k` ticks.
R3. Ball movement:
R3.1. Ball may be carried by an active agent or passed along unobstructed line-of-sight grid vectors.
R3.2. Interception checks are evaluated for any opposing defender in the pass vector path.
R4. Scoring rules:
R4.1. An agent in `x = +4` (or `x = -4`) executing `shoot_goal()` rolls against goalkeeper defense:
R4.2. Successful shot -> `score += 1.0` (*caccia*). Teams swap field ends; ball resets to center `(0, 0)`.
R4.3. Missed/wide shot -> `opponent_score += 0.5` (*mezza caccia*). Opponent gains possession at goal line.
R5. Match termination: Match ends when tick count reaches `T_max = 3000` (50 minutes). The team with the highest net score wins.
R6. [gap-fill] Tie-break rule: If tied at `T_max`, sudden-death overtime of 500 ticks is played; first score of any value wins immediately.
## Edge cases
- Tie at the end condition: Resolved via sudden-death overtime (R6).
- A player with no legal move: Pinned agents pass their turn; unpinned agents always have legal movement/tactical options.
- Repetition / cycles: Bounded strictly by the 3000-tick match timer (R5).
- Simultaneous claims on the same resource or position: Ball loose-ball recovery resolved via proximity distance and agent strength parameter.
- Player elimination mid-round: Disqualified/injured agents are removed from the field without replacement, altering numerical balance per historical rules.
- Resource exhaustion: Not applicable — ball state resets after goals.
- Termination guarantee: Guaranteed by the strict 3000-tick duration and sudden-death cap.
## State representation
- Persistent state: `scores` `(float, float)`, `match_ticks` (int 0..3000), `field_orientation` (normal/swapped).
- Per-tick ephemeral state: `agent_states` (array of structs: `pos (x,y)`, `role`, `status: FREE / PINNED`, `has_ball: bool`), `ball_pos` `(x,y)`.
- Observable information: Fully observable by all agents.
## Action space
- `move(dx, dy)`: Step to adjacent grid cell.
- `pass(agent_id)`: Deliver aerial or ground pass.
- `grapple(opponent_id)`: Attempt 1v1 pin.
- `shoot()`: Attempt shot on end-zone net.
## Playability notes
- Single-player-vs-AI: Needs adaptation — best played as a tactical squad manager (controlling overall team formations and ball carriers) vs. AI.
- Group games (n>4): Reduced 7v7 agent squad preserves core tactical mechanics without excess compute overhead.
- Physical-skill games: Martial grappling and ball handling abstracted into discrete probability tables.
- What is lost: The raw brutality of bare-knuckle Renaissance hand-to-hand combat in deep sand.
## Implementation hints
Two-layer architecture: a positional discrete grid layer coupled to a 1v1 combat resolver. First milestone: Verify the *mezza caccia* (0.5 point penalty) scoring feedback loop before adding 27-agent pathfinding.
Download as JSON
Download the full game corpus entry for this game as structured JSON.
Download JSONCite this page
Papal Historic Football (Vatican City). (n.d.). Games of Countries. https://gamesofcountries.com/games/va-vatican-city/calcio-fiorentino-pontificio
@misc{gamesofcountries-va-vatican-city-calcio-fiorentino-pontificio,
title = {Papal Historic Football (Vatican City)},
year = {n.d.},
publisher = {Games of Countries},
url = {https://gamesofcountries.com/games/va-vatican-city/calcio-fiorentino-pontificio}
}