Skip to main content
Formal Specification

Stick and Peg — Formal Specification

Nepal · stick-team · Confidence: low

← Back to game page

Implementation Suitability

Suitability: adaptation-requiredFidelity: discretizedGap-fills: 3
# Dandi Biyo — AI-optimized specification ## General idea Dandi Biyo is a team-based tip-cat and striking game where batters launch and juggle a wooden peg (*biyo*) with a stick (*dandi*), scoring points based on distance and mid-air strikes (*tyak*), while fielders attempt to catch airborne projectiles or throw them to hit the home base. The AI model formalizes the physical trajectory dynamics into discrete risk-reward decisions. ## Formal model - **Abstraction taken:** Physical aerodynamics of flipping and hitting the *biyo* are modeled via a parameterized stochastic outcome generator based on batter skill, chosen aggression level (A∈{SAFE,BALANCED,AGGRESSIVE}), and fielder positioning grid. - **Player/agent count:** 2 teams of 7 agents each (Team A batting, Team B fielding). - **Turn model:** Sequential batter rotations per inning; 2 innings per team. ## Formal rules R1. Match Setup: Team A bats in Inning 1; Team B fields. 7 batters in fixed batting order. R2. Phase 1 (The Flip / *Ghat*): Active batter executes a flip action. - [gap-fill] With probability Pcatch(A), a fielder catches the *biyo* in flight → Batter OUT immediately. - Otherwise, *biyo* lands at distance D1∼LogNormal(μ(A),σ). R3. Field Return: Fielders execute a throw-at-base action. - [gap-fill] With probability Phit_base(D1), the return hits the *khopi* or resting *dandi* → Batter OUT immediately. - If miss, batter proceeds to Phase 2. R4. Phase 2 (Mid-Air Strike / *Tyak*): Batter chooses target *tyak* multiplier k∈{1,2,3}. - Batter attempts k aerial taps. Success probability Ptyak(k)=pbasek. - If all k taps succeed, *biyo* is launched to final distance D2∼LogNormal(μ2,σ2), and points scored =k×⌊D2/Ldandi⌋. - If tap fails 3 times, batter is OUT with 0 points scored for that turn. R5. Batter Dismissal and Rotation: When a batter is OUT, turn advances to the next batter in the lineup. When all 7 batters are dismissed, the inning concludes. R6. Role Inversion: Teams swap batting and fielding roles for Inning 2. R7. Win Condition: The team with higher aggregate points at the end of 2 full innings wins. ## Edge cases - Tie at the end condition: If aggregate scores are tied after 2 innings, [gap-fill] resolved by a single-batter sudden death *tyak* playoff. - A player with no legal move: Not applicable — striking actions are always available. - Repetition / cycles: Bounded by finite innings (7 dismissals per inning). - Simultaneous claims on the same resource or position: Turns are strictly sequential. - Player elimination mid-round: Dismissed batters are skipped until the next team inning. - Resource exhaustion: Innings terminate strictly upon 7 dismissals. - Termination guarantee: Guaranteed by strictly bounded batter counts and probabilistic dismissal per attempt. ## State representation - `current_inning`: Integer ∈[1,2]. - `batting_team`: Team ID ∈{0,1}. - `team_scores`: Array [S0,S1] of integers. - `current_batter_index`: Integer ∈[0,6]. - `outs_in_inning`: Integer ∈[0,7]. ## Action space - `launch_ghat(aggression_level)`: Select risk profile ∈{SAFE,BALANCED,AGGRESSIVE}. - `attempt_tyak(target_multiplier)`: Select intended aerial tap count k∈{1,2,3}. ## Playability notes - Single-player-vs-AI: Natural as a 1v1 tactical batting-profile duel against an AI fielding/batting agent. - Group games (n>4): Reduced to 2 captain agents controlling team rosters and risk policies. - Physical-skill games: Physical batting finesse is abstracted into expected value curves over risk-reward choices. - What is lost in the adaptation: Hand-eye reaction timing and physical ballistics are replaced by probabilistic sampling. ## Implementation hints Branching factor is small (3 risk levels × 3 multiplier targets). Easy to implement as a discrete Markov Decision Process. First milestone: build the single-batter 2-phase pipeline with placeholder Gaussian distance samplers.

Download as JSON

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

Download JSON
Cite this page

Stick and Peg (Nepal). (n.d.). Games of Countries. https://gamesofcountries.com/games/np-nepal/dandi-biyo

@misc{gamesofcountries-np-nepal-dandi-biyo,
  title = {Stick and Peg (Nepal)},
  year = {n.d.},
  publisher = {Games of Countries},
  url = {https://gamesofcountries.com/games/np-nepal/dandi-biyo}
}