Skip to main content
Formal Specification

Pitching to the Line (Czech Coin / Button Pitching) — Formal Specification

Czechia · dexterity-throw · Confidence: high

← Back to game page

Implementation Suitability

Suitability: engine-readyFidelity: probabilistic-proxyGap-fills: 2
# Hra v čáru — AI-optimized specification ## General idea A two-stage pitching and coin-flipping wagering game where agents pitch tokens toward a boundary line to establish priority order. The closest non-overshooting agent gains the first opportunity to toss all pooled tokens into the air, claiming every coin landing heads-up before passing remaining tails to the next closest agent. ## Formal model - **Abstraction taken:** Token pitching along a 1D line of length Dtarget=4.0 m is modeled via chosen throw force F∈[0,1] yielding distance x=5.0⋅F+ϵ, with ϵ∼𝒩(0,σ2). Flipping tokens is modeled as independent Bernoulli trials B(0.5) (fair coin toss). - **Player/agent count:** N agents (2≤N≤8, standard N=4). - **Turn model:** Two phases per round: (1) Sequential pitch; (2) Priority-ordered coin flipping until pool exhaustion. ## Formal rules R1. Setup: Round begins with each of the N agents contributing 1 token to the pool P (|P|=N). Target line is at x=4.0 m. R2. Pitching phase: Each agent i chooses throwing parameter Fi∈[0,1]. Landing coordinate xi=5.0⋅Fi+ϵi, where ϵi∼𝒩(0,0.152). R3. Ranking determination: - If xi>4.0: Token is disqualified (overshot); rank assigned as +∞ (or sorted by largest overshoot). - If xi≤4.0: Proximity metric is Δi=4.0−xi≥0. - Priority queue Q sorts agents by ascending Δi. Ties in Δi broken by coin flip or agent ID. R4. Flipping phase: - Active agent is A=Q.pop(). - Agent A flips all currently remaining tokens in pool P. - Each token lands `HEADS` with p=0.5, `TAILS` with p=0.5. - All tokens landing `HEADS` are transferred to agent A's bank. - Remaining `TAILS` tokens stay in pool P. - If |P|>0 and Q is not empty: Next agent in Q takes turn flipping remaining pool. - [gap-fill] If Q is exhausted while |P|>0, priority wraps back to the 1st-ranked agent in Q. R5. Round end: Round concludes when |P|=0. R6. Match win condition: After M rounds (default M=5), the agent with the highest total accumulated tokens in their bank wins the match. R7. [gap-fill] Sudden death tiebreaker: If scores are tied after M rounds, tied agents participate in a single sudden-death pitch; closest to the line wins. ## Edge cases - Tie at the end condition: Resolved via R7 sudden-death pitch. - A player with no legal move: Not applicable; pitching and flipping actions are strictly automatic given parameters. - Repetition / cycles: Bounded because each flip pass has P(heads)=0.5>0 per token, guaranteeing geometric decay of pool |P|→0. - Simultaneous claims on the same resource: Priority queue Q strictly orders all token claims. - Player elimination mid-round: An agent who runs out of bank tokens is eliminated from subsequent rounds; current round completes normally. - Resource exhaustion: Monotonic depletion of round pool to 0. - Termination guarantee: Strict termination guaranteed by geometric convergence of fair coin tosses and finite round count M. ## State representation - Persistent state: `agent_banks: list[int]`, `current_round: int`, `total_rounds: int`. - Ephemeral round state: `pitches: dict[int, float]`, `priority_queue: list[int]`, `remaining_pool: int`. - Observable information: Fully observable (all pitches, token states, and scores are public). ## Action space - `pitch_token(force: float)` with force∈[0.0,1.0]. - `flip_pool()`: Triggers Bernoulli flip of remaining pool tokens. ## Playability notes - Single-player-vs-AI: Natural; 1 human vs. 1–3 AI bots tuning pitch force against noise variance. - Group games (n>4): Scales smoothly up to 8 agents. - Physical-skill games: Underhand coin tossing and slide friction are captured by the Gaussian error model. - What is lost: The physical tactile art of flicking and spinning coins off the thumb. ## Implementation hints Low complexity. The two-phase structure (pitching sort followed by geometric flip loop) is straightforward to code and test. First milestone: implement round logic for 4 agents with deterministic force inputs and randomized coin flips.

Download as JSON

Download the full game corpus entry for this game as structured JSON.

Download JSON
Cite this page

Pitching to the Line (Czech Coin / Button Pitching) (Czechia). (n.d.). Games of Countries. https://gamesofcountries.com/games/cz-czechia/hra-v-caru

@misc{gamesofcountries-cz-czechia-hra-v-caru,
  title = {Pitching to the Line (Czech Coin / Button Pitching) (Czechia)},
  year = {n.d.},
  publisher = {Games of Countries},
  url = {https://gamesofcountries.com/games/cz-czechia/hra-v-caru}
}