Skip to main content
Formal Specification

Tipcat (Polish Wood-Tip Batting) — Formal Specification

Poland · dexterity-throw · Confidence: medium

← Back to game page

Implementation Suitability

Suitability: adaptation-requiredFidelity: probabilistic-proxyGap-fills: 2
# Klipa — AI-optimized specification ## General idea A traditional two-agent projectile batting game combining a two-stage kinetic strike (popping a tapered wooden peg into the air and driving it downfield) with distance-measuring point accumulation and fielding counter-throws. The central strategic balance lies in deciding whether to risk additional advancement hits or lock in distance points before fielders return the peg to base. ## Formal model - **Abstraction taken:** The continuous 3D aerodynamics of the tapered peg (*klipa*) and bat swing are abstracted into two sequential probabilistic skill rolls: (1) `P_launch` (pop up height/stability), and (2) `P_drive(power, angle)` yielding a 1D downfield displacement `D_drive` in meters. The field is modeled as a 1D polar sector from the origin base (radius 0). - **Player/agent count:** 2 agents (Agent 1 = Striker, Agent 2 = Fielder) taking alternating innings. - **Turn model:** Sequential round structure comprising: (1) Striker launch and drive, (2) Fielder airborne intercept check, (3) Fielder return-throw check, (4) Striker distance measurement / advancement. ## Formal rules R1. **Setup:** Base circle is at coordinate x=0 (radius r=1 m). Target score is Starget=100 points. Both players start with 0 score. R2. **Strike sequence:** The active Striker attempts a launch-and-drive. - Strike roll succeeds with probability Phit=clamp(0.3+0.5⋅striker_skill,0.1,0.95). - If missed, attempt count increases by 1. Three consecutive misses forfeits the turn with 0 points scored. - On success, the peg travels to distance D1=sample_distance(power,striker_skill)∈[10,80] meters. R3. **Fielding intercept check:** - Fielder has catch probability Pcatch=f(D1,fielder_skill). - If caught, turn ends immediately, Striker scores 0 points, and roles swap. R4. **Return throw check:** If uncaught, the Fielder executes `return_throw(accuracy)` from D1 toward x=0. - The thrown landing distance Dreturn is sampled around 0 with variance inversely proportional to `fielder\_skill`. - If |Dreturn|≤1.0 m (lands inside base circle), the Striker is retired with 0 points for the round, and roles swap. R5. **Advancement strikes & scoring:** If the return throw lands outside the base (|Dreturn|>1.0 m): - The Striker may execute up to 3 optional advancement strikes from Dreturn, each adding stochastic displacement ΔD≥0, or choose `claim_points()`. - [gap-fill] An advancement strike carries a 20% risk of total mis-hit (whiff), which ends the turn immediately with current points retained. - Final round points earned = ⌊Dfinal/Lbat⌋, where Lbat=0.75 m (1 point per bat-length). R6. **Role alternation & Victory:** After points are recorded, the players swap roles (Striker ↔ Fielder). The first player to reach or exceed Starget=100 points at the end of an equal number of innings wins. ## Edge cases - **Tie at the end condition:** If both players reach ≥100 points at the end of an equal number of innings, [gap-fill] additional innings (1 batting turn each) are played until the tie is broken. - **A player with no legal move:** Not applicable; Striker always has `strike()` and Fielder always has `return_throw()`. - **Repetition / cycles:** Bounded because points strictly accumulate (≥0) each inning, and Starget is finite. - **Simultaneous claims on the same resource or position:** Not applicable; turns and roles are strictly sequential. - **Player elimination mid-round; turn order repair:** Not applicable in 2-player format. In team mode, eliminated batters pass turn to the next agent in roster. - **Resource exhaustion:** Not applicable; wooden pegs and bats are reusable. - **Termination guarantee:** The Markov chain of innings has absorbing states when cumulative score reaches 100 with probability 1 as innings N→∞. ## State representation - **Persistent state:** `scores` (Map[AgentId, Int]), `active_striker_id`, `active_fielder_id`, `inning_count` (Int). - **Per-round ephemeral state:** `striker_miss_count` (0..3), `peg_position` (Float), `advancement_hits_used` (0..3), `is_in_base_circle` (Bool). - **Observable information:** Fully observable to both agents. ## Action space - `strike(power: Float[0..1])`: Striker action to launch and hit peg. - `return_throw(aim_bias: Float)`: Fielder action to target the base circle. - `advancement_strike()`: Optional post-return hit to increase distance. - `claim_points()`: Lock in current distance and finalize score. ## Playability notes - **Single-player-vs-AI:** Natural — human as Striker against AI Fielder or vice versa. - **Group games (n>4):** Reduces to 2 teams alternating full batting rosters. - **Physical-skill games:** Modeled via two primary skill parameters per agent: `striking_accuracy` and `fielding_throw_accuracy`. - **What is lost:** The micro-timing of double-contact bat-peg physics is replaced with a composite bivariate distribution. ## Implementation hints Extremely lightweight state space (3 integers, 1 float). Branching factor per decision point is 1–3 choices. Recommended first milestone: implement R1–R4 with simple uniform random distance distributions, then tune skill-dependent distance distributions and R5 advancement risk.

Download as JSON

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

Download JSON
Cite this page

Tipcat (Polish Wood-Tip Batting) (Poland). (n.d.). Games of Countries. https://gamesofcountries.com/games/pl-poland/klipa

@misc{gamesofcountries-pl-poland-klipa,
  title = {Tipcat (Polish Wood-Tip Batting) (Poland)},
  year = {n.d.},
  publisher = {Games of Countries},
  url = {https://gamesofcountries.com/games/pl-poland/klipa}
}