Skip to main content
Formal Specification

Hatay Felt-Cloak Wrestling — Formal Specification

Türkiye · wrestling-combat · Confidence: low

← Back to game page

Implementation Suitability

Suitability: adaptation-requiredFidelity: probabilistic-proxyGap-fills: 3
# Aba Güreşi — AI-optimized specification ## General idea A traditional felt-cloak wrestling game modeled as a simultaneous-action combat game with stance, balance, and grip endurance. Unlike belt wrestling, players can grip any part of the opponent's cloak (*aba*) or execute direct leg picks, adding a high-risk takedown mechanic that trades balance for immense throw power. ## Formal model - **Abstraction taken:** The complex continuous physical grappling, hip leverage, and leg picks are abstracted into a 2D state space containing balance, stamina, and cloak grip stability. [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). - `points[p]: integer` (initialized to 0). R2. Simultaneous Actions: On each tick, both wrestlers choose one action from `action_set ∈ {BLOCK, CLOAK_THROW, LEG_PICK, PIVOT}`. R3. Resolution Matrix: - **LEG_PICK beats CLOAK_THROW:** - If Wrestler A chooses `LEG_PICK` and Wrestler B chooses `CLOAK_THROW`: - A dives to grab B's leg. - Success probability: `P(Pick_Success) = skill(A) * (stamina(A) / stamina(B)) * 0.75`. - If success: B is thrown. We check fall quality: - If `random() < 0.50`: **Açık Tuş (Clean Victory)**: `points[A] += 3`. Reset round. - Otherwise: **Yan Düşüş (Side Fall)**: B is thrown to hands/knees: `points[A] += 1`. - If failure: A is highly vulnerable, their balance decreases: `balance[A] -= 30.0`. - **CLOAK_THROW beats PIVOT:** - If A chooses `CLOAK_THROW` and B chooses `PIVOT`: - A grips B's collar and pulls to throw. - `P(Throw_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 using a belt lift. - `P(Pivot_Success) = skill(A) * 0.60`. - If success: B's balance decreases: `balance[B] -= 25.0`. - **BLOCK beats LEG_PICK/CLOAK_THROW:** - If A chooses `BLOCK` and B chooses `LEG_PICK` or `CLOAK_THROW`: - B's offensive action is nullified. B's stamina decreases: `stamina[B] -= 15.0`. 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 3 points (Açık Tuş): `points[thrower] += 3`. - Reset round: both agents' stamina and balance reset to 100.0. R5. Match Progression: - [gap-fill] The match lasts exactly 50 ticks (representing the 5-minute time limit). - On each tick, stamina decays slightly: `stamina[p] -= random(0.5, 2.0)`. R6. End condition: - The match ends immediately if a wrestler scores 3 points (Açık Tuş), or when the 50-tick limit is reached. R7. Win condition: The wrestler with the highest score wins. Ties enter a 10-tick golden-score extension. ## Edge cases - **Tie at the end condition:** Handled by a 10-tick golden score extension (R7). - **A player with no legal move:** Not applicable. - **Repetition / cycles:** Bounded by the 50-tick limit (R5). - **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 10.0 stamina. - **Player elimination mid-round:** Not applicable. - **Resource exhaustion:** Not applicable. - **Termination guarantee:** Guaranteed by the strictly decreasing tick counter (R5). ## State representation - Persistent state: - `points: Array[2] of integer`. - `balance: Array[2] of float`. - `stamina: Array[2] of float`. - `tick_counter: integer`. - Information set: Public, fully observable. ## Action space - `declare_maneuver(action)`: where `action ∈ {BLOCK, CLOAK_THROW, LEG_PICK, PIVOT}`. Valid for both wrestlers on each tick. ## Playability notes - Single-player-vs-AI: Highly suitable. Plays like a fighting game where the player must balance the high-risk `LEG_PICK` (which scores points but risks a massive balance penalty if blocked) against safer moves like `CLOAK_THROW`. - Group games: Not applicable (strictly 1v1). - Physical-skill games: Physical cloak-grappling, balance, and sweeps are fully abstracted into stamina, balance, and RPS action priorities. - What is lost: The physical sensation of pulling thick wool felt and picking a peer's ankle in the dirt. ## 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 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

Hatay Felt-Cloak Wrestling (Türkiye). (n.d.). Games of Countries. https://gamesofcountries.com/games/tr-turkiye/aba-guresi

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