Skip to main content
Formal Specification

Japanese Battledore and Shuttlecock — Formal Specification

Japan · dexterity-throw · Confidence: low

← Back to game page

Implementation Suitability

Suitability: adaptation-requiredFidelity: probabilistic-proxyGap-fills: 3
# Hanetsuki — AI-optimized specification ## General idea A traditional two-player racket-and-shuttlecock rally duel played with rectangular wooden paddles (*hagoita*) and a feather-tipped soapberry seed (*hane*). Players alternate striking the shuttlecock to keep it continuously aloft without a net; missing a shot incurs a fault point and the traditional penalty of having ink brushed across the face. ## Formal model - **Abstraction taken:** Continuous aerodynamic flight, wind drift, and physical wrist-and-paddle impact dynamics are abstracted into a discrete stroke-by-stroke rally exchange. Strategic choice consists of shot selection (lob, drive, angled placement) with stochastic execution probability `P(success) = f(difficulty, incoming_pressure, skill)`. - **Player/agent count:** 2 players (Player 0 and Player 1). - **Turn model:** Sequential alternating strokes within rallies, across a match of multiple rallies. ## Formal rules R1. Match setup: Both players start with 0 fault points (`ink_marks = 0`). Target match loss threshold is 5 fault points. R2. Rally initiation: The designated server serves the shuttlecock across the 4-meter playing gap, initializing the rally with baseline parameters `(speed = 1.0, pressure = 0.0)`. R3. On their turn, the receiving player chooses one stroke type from `{High_Clear, Fast_Drive, Angled_Slice}`. R4. Stroke resolution: - Each stroke has an inherent execution difficulty `D` and generates outgoing pressure `P_out`. - Success probability: `P(success) = clamp(skill_active - D - pressure_incoming, 0.05, 0.95)`. - If successful, the shuttlecock crosses to the opponent with `pressure_incoming = P_out`, and the stroke count increments by 1. R5. Fault condition: - If the stroke fails, the shuttlecock touches the ground (*Ochita*). - The player who failed to return the shuttlecock receives 1 fault point (`ink_marks += 1`), and the rally terminates. R6. Next rally: The player who won the previous rally serves the next rally. R7. Match end: The match terminates immediately when either player accumulates 5 fault points. The player with fewer fault points wins the match. R8. [gap-fill] Rally stroke cap: If a single rally reaches 50 unbroken strokes without a fault, the rally terminates automatically and is awarded to the player who executed more high-difficulty strokes during that exchange. R9. [gap-fill] Stroke taxonomy specifications: - `High_Clear`: Difficulty = 0.10, Output Pressure = 0.10. - `Fast_Drive`: Difficulty = 0.25, Output Pressure = 0.35. - `Angled_Slice`: Difficulty = 0.40, Output Pressure = 0.50. R10. [gap-fill] Match tiebreaker: If both players reach 5 fault points simultaneously under variant penalty scoring, play continues to a 2-point margin advantage (win by 2). ## Edge cases - **Tie at the end condition:** Prevented by sequential fault assignment; if variant enables simultaneous faults, resolved by R10 (win-by-2 advantage). - **A player with no legal move:** Not applicable; all 3 stroke types are always available to the receiving player. - **Repetition / cycles:** Rallies can cycle between conservative high clears; bounded by R8 (50-stroke cap). - **Simultaneous claims on the same resource or position:** Not applicable; turns alternate strictly with each stroke. - **Player elimination mid-round:** Not applicable; strictly a 2-player match. - **Resource exhaustion:** Only 1 shuttlecock is active per rally; match ends upon 5 fault points. - **Termination guarantee:** Guaranteed by non-zero failure probability on every stroke (`P(failure) >= 0.05`), the 50-stroke rally cap (R8), and the 5-fault match limit (R7). ## State representation - **Persistent state:** `fault_scores[2]` (ints in `[0..5]`), `current_server` (0 or 1), `rally_number` (int). - **Per-rally state:** `active_player` (0 or 1), `stroke_count` (int), `incoming_pressure` (float), `shuttle_in_air` (bool). - **Observable information:** Fully observable perfect-information game. ## Action space - `serve()`: Initiate rally. - `strike(stroke_type: HIGH_CLEAR | FAST_DRIVE | ANGLED_SLICE)`: Execute chosen return shot. ## Playability notes - **Single-player-vs-AI:** Highly natural; AI uses standard risk-reward evaluation balancing safe baseline returns against aggressive drives that stress opponent skill. - **Group games (n>4):** Reduces cleanly to 2 players on court; extra players form a King-of-the-Court ladder where winners stay on. - **Physical-skill games:** Manual eye-hand coordination is parameterized into player skill and stroke difficulty values. - **What is lost in the adaptation:** The physical feedback of wooden paddle resonance, wind reading, and the celebratory visual spectacle of painted ink faces. ## Implementation hints Discrete turn-based Markov chain. Each stroke is a simple state transition updating rally metrics. First milestone: implement 1v1 rally loop with fixed skill parameters, then add stroke difficulty trade-offs.

Download as JSON

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

Download JSON
Cite this page

Japanese Battledore and Shuttlecock (Japan). (n.d.). Games of Countries. https://gamesofcountries.com/games/jp-japan/hanetsuki

@misc{gamesofcountries-jp-japan-hanetsuki,
  title = {Japanese Battledore and Shuttlecock (Japan)},
  year = {n.d.},
  publisher = {Games of Countries},
  url = {https://gamesofcountries.com/games/jp-japan/hanetsuki}
}