Formal Specification
Persian Tip-Cat Two-Stick Game — Formal Specification
Iran · stick-team · Confidence: low
← Back to game pageImplementation Suitability
Suitability: adaptation-requiredFidelity: discretizedGap-fills: 3
# Alak Dolak — AI-optimized specification
## General idea
A traditional tip-cat bat-and-trap game featuring asymmetric batting and fielding roles. The core gameplay balances aerial launch trajectory, fielder positioning for interceptions or target throwing, and multi-stage aerial strikes where risk of missing a mid-air tap is weighed against cumulative distance scoring measured in discrete stick-lengths.
## Formal model
- **Abstraction taken:** The outfield is modeled as a 1D distance axis `x >= 0` with discrete spatial sectors. Physical batting contact, mid-air taps, and fielder throws at the base stick are abstracted into parameterized stochastic resolvers dependent on agent attributes (`batting_skill`, `fielding_skill`, `aim_skill`).
- **Player/agent count:** 2 teams of `K` agents each (`2 <= K <= 6`). Team 0 = Batting Team, Team 1 = Fielding Team.
- **Turn model:** Sequential phase-based model per batter, transitioning through Launch -> Field Return -> Aerial Strike -> Measure.
## Formal rules
R1. Base Setup:
- Base is located at `x = 0`.
- Alak length is defined as 1 unit (`1 AL = 0.70 m`).
- Current batter index `b ∈ [0..K-1]`.
R2. Launch Phase:
- Batter executes `launch_dolak()`.
- [gap-fill] Landing distance `x_land = random_uniform(10, 40) * batting_skill(b)`.
R3. Fielding Catch Check:
- Fielders are positioned at distributed field sectors.
- Probability of catching in air: `P(catch) = sum(fielding_skill(f) * proximity_factor(f, x_land))`. [gap-fill]
- If caught: Batter is eliminated immediately (`batter_status[b] = OUT`), advance to next batter.
R4. Base Throw Check:
- If not caught, the fielder at `x_land` executes `throw_at_base()`.
- Probability of hitting base stick from distance `x_land`: `P(hit_base) = max(0.05, aim_skill(f) * (1.0 - x_land / 60.0))`. [gap-fill]
- If hit: Batter is eliminated (`batter_status[b] = OUT`), advance to next batter.
R5. Aerial Strike Phase:
- If base throw misses, batter stands at `x_land` and may execute up to 3 aerial strikes: `strike_midair(strike_number)`.
- For each attempt `i ∈ {1, 2, 3}`:
- `P(contact) = batting_skill(b) * (0.85 - 0.10 * (i - 1))`.
- If hit succeeds: `x_land += random_uniform(5, 25) * batting_skill(b)`.
- If hit misses: Aerial strike phase terminates immediately.
R6. Scoring Phase:
- Team score increases by `floor(x_land / 1.0)` points.
- Batter status becomes `COMPLETED`, next batter in batting order steps up.
R7. Inning Inversion & Match End:
- When all `K` batters of a team are `OUT` or `COMPLETED`, roles invert (Team 0 becomes Fielders, Team 1 becomes Batters).
- Match runs for 2 full innings (each team bats twice). Team with higher total score wins.
## Edge cases
- **Tie at the end condition:** If scores are equal after regulation innings, 1 sudden-death batter per team is played until tie is broken.
- **A player with no legal move:** Fielders cannot move during the aerial strike phase; their legal action set activates only during ball flight.
- **Repetition / cycles:** Bounded strictly by the finite number of batters per inning (`K`) and finite innings (2).
- **Simultaneous claims on the same position:** Resolved by assigning catching priority to the fielder closest to `x_land`.
- **Player elimination mid-round:** Eliminated batters cannot bat again in the current inning; batting order advances monotonically.
- **Resource exhaustion:** Not applicable; sticks are persistent.
- **Termination guarantee:** Guaranteed by finite innings and finite batters per team.
## State representation
- **Persistent state:**
- `scores: Tuple[int, int]`.
- `current_inning: int` (1..2).
- `active_batting_team: int` (0 or 1).
- `active_batter_idx: int` (0..K-1).
- `batter_status: Array[K] of enum {PENDING, OUT, COMPLETED}`.
- **Ephemeral phase state:**
- `phase: enum {LAUNCH, CATCH_CHECK, THROW_CHECK, AERIAL_STRIKE, SCORING}`.
- `dolak_position: float` (distance `x` along field axis).
- `strike_counter: int` (0..3).
- **Information set:** Fully observable common knowledge.
## Action space
- `launch_dolak()`: Precondition: `phase == LAUNCH`, actor is active batter.
- `attempt_catch(fielder_id)`: Precondition: `phase == CATCH_CHECK`, actor is fielder in range.
- `throw_at_base(fielder_id)`: Precondition: `phase == THROW_CHECK`, actor is fielder holding dolak.
- `strike_midair()`: Precondition: `phase == AERIAL_STRIKE`, actor is active batter, `strike_counter < 3`.
- `pass_strike()`: Precondition: `phase == AERIAL_STRIKE`, actor opts to cash in current distance without risking a miss.
## Playability notes
- **Single-player-vs-AI:** Highly playable. Player controls the batter aiming launches and timing multi-strike combos, while AI fielders optimize coverage positioning and throw accuracy.
- **Group games (n>4):** Scales cleanly from 2v2 to 6v6 by adding fielders to coverage sectors.
- **Physical-skill model:** Real-world hand-eye coordination and stick batting are mapped to distance-dependent stochastic contact functions.
- **What is lost:** The physical tactile feel of striking a spinning wooden peg and judging wind resistance on the fly.
## Implementation hints
Model the field as a 1D scalar line with fielder coverage intervals. Build the state machine to cleanly transition through R2 -> R3 -> R4 -> R5 -> R6, and ensure the batter has the strategic option to `pass_strike()` to lock in score.
Download as JSON
Download the full game corpus entry for this game as structured JSON.
Download JSONCite this page
Persian Tip-Cat Two-Stick Game (Iran). (n.d.). Games of Countries. https://gamesofcountries.com/games/ir-iran/alak-dolak
@misc{gamesofcountries-ir-iran-alak-dolak,
title = {Persian Tip-Cat Two-Stick Game (Iran)},
year = {n.d.},
publisher = {Games of Countries},
url = {https://gamesofcountries.com/games/ir-iran/alak-dolak}
}