Skip to main content
Formal Specification

Belt Wrestling / Tatar Wrestling — Formal Specification

Türkiye · wrestling-combat · Confidence: low

← Back to game page

Implementation Suitability

Suitability: adaptation-requiredFidelity: probabilistic-proxyGap-fills: 3
# Kuşak Güreşi — AI-optimized specification ## General idea A traditional belt wrestling game modeled as a simultaneous-action combat game with stance, balance, and grip endurance. Wrestlers must maintain their double-handed belt grip while executing sweeps, lifts, or defensive blocks to deplete the opponent's balance and secure a clean throwing fall. ## Formal model - **Abstraction taken:** The complex continuous physical grappling, hip leverage, and leg sweeps are abstracted into a 2D state space containing balance, stamina, and grip strength. [gap-fill] Grappling maneuvers are resolved using simultaneous-action rock-paper-scissors priority trees. - **Player/agent count:** 2 agents (Wrestler 0, Wrestler 1). - **Turn model:** Simultaneous turns (ticks). On each tick, both agents declare their stance/maneuver, which are then resolved. ## Formal rules R1. Wrestler State: Each wrestler `p` has: - `balance[p]: float` (initialized to 100.0). - `stamina[p]: float` (initialized to 100.0). - `grip_strength[p]: float` (initialized to 100.0). - `points[p]: integer` (initialized to 0). R2. Simultaneous Actions: On each tick, both wrestlers choose one action from `action_set ∈ {BLOCK, LIFT, SWEEP, PIVOT}`. R3. Resolution Matrix: - **LIFT beats SWEEP:** - If Wrestler A chooses `LIFT` and Wrestler B chooses `SWEEP`: - A attempts to lift and throw B. - Success probability: `P(Lift_Success) = skill(A) * (stamina(A) / stamina(B)) * 0.70`. - If success: B is thrown. We check fall quality: - If `random() < 0.60`: **Temiz Düşüş (Clean Fall)**: `points[A] += 1`. Reset round. - Otherwise: **Yan Düşüş (Side Fall)**: B's balance is reduced: `balance[B] -= 40.0`. - If failure: A's stamina decreases: `stamina[A] -= 15.0`. - **SWEEP beats PIVOT:** - If A chooses `SWEEP` and B chooses `PIVOT`: - A attempts to trip B's foot. - `P(Sweep_Success) = skill(A) * (balance(A) / balance(B)) * 0.65`. - If success: B's balance decreases: `balance[B] -= 30.0`. - If failure: A's balance decreases: `balance[A] -= 15.0`. - **PIVOT beats BLOCK:** - If A chooses `PIVOT` and B chooses `BLOCK`: - A attempts to turn and throw B over their hip. - `P(Pivot_Success) = skill(A) * 0.60`. - If success: B's balance decreases: `balance[B] -= 25.0`. - **BLOCK beats LIFT/SWEEP:** - If A chooses `BLOCK` and B chooses `LIFT` or `SWEEP`: - B's offensive action is nullified. B's stamina decreases: `stamina[B] -= 10.0`. R4. Fall Trigger: - [gap-fill] If at any tick an agent's balance drops to `<= 0.0`: - That agent is thrown onto their back. - The throwing agent scores 1 point: `points[thrower] += 1`. - Reset round: both agents' stamina, balance, and grip strength reset to 100.0. R5. Grip Break: - On each tick, grip strength is updated: `grip_strength[p] -= random(0.0, 5.0)` depending on fatigue. - If either wrestler's `grip_strength` reaches `<= 10.0`: - [gap-fill] The referee pauses the match. Grip strength is reset to 100.0 for both players, but stamina is reduced by 10.0 for the player whose grip failed. R6. End condition: - First wrestler to score 2 points wins the match. R7. Win condition: The wrestler with 2 points wins. ## Edge cases - **Tie at the end condition:** Not applicable (first to 2 points wins). - **A player with no legal move:** Not applicable. - **Repetition / cycles:** Bounded, because stamina and grip are depleted over time, leading to lower defense and eventual successful throws. - **Simultaneous claims on the same resource:** Handled by R3's simultaneous action matrix. If both choose the same action, they nullify each other and both lose 5.0 stamina. - **Player elimination mid-round:** Not applicable. - **Resource exhaustion:** Not applicable. - **Termination guarantee:** Guaranteed by R3's non-zero success chance and the monotonic decay of stamina and grip, ensuring eventual round completion. ## State representation - Persistent state: - `points: Array[2] of integer`. - `balance: Array[2] of float`. - `stamina: Array[2] of float`. - `grip_strength: Array[2] of float`. - Information set: Public, fully observable. ## Action space - `declare_maneuver(action)`: where `action ∈ {BLOCK, LIFT, SWEEP, PIVOT}`. Valid for both wrestlers on each tick. ## Playability notes - Single-player-vs-AI: Natural. Plays like a fighting or tactical card game where the player reads the AI's past stance patterns to select counter-maneuvers. - Group games: Not applicable (strictly 1v1). - Physical-skill games: Physical wrestling holds, balance, and leverage are fully abstracted into stamina, balance, and RPS action priorities. - What is lost: The raw, muscle-straining physical contact of belt-grappling. ## Implementation hints Medium complexity. Requires implementing a simultaneous-turn state machine and resolving the rock-paper-scissors priority matrix (R3). First milestone: build the turn-update loop with static stance inputs and verify the balance-depletion and point-scoring triggers.

Download as JSON

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

Download JSON
Cite this page

Belt Wrestling / Tatar Wrestling (Türkiye). (n.d.). Games of Countries. https://gamesofcountries.com/games/tr-turkiye/kusak-guresi

@misc{gamesofcountries-tr-turkiye-kusak-guresi,
  title = {Belt Wrestling / Tatar Wrestling (Türkiye)},
  year = {n.d.},
  publisher = {Games of Countries},
  url = {https://gamesofcountries.com/games/tr-turkiye/kusak-guresi}
}