game-theory

• HeroClix
• The Nash Bargaining Problem
• The Multiplayer Coalition Problem
• Bilateral Trade Valuation
• MOO1 Optimal Strategy
• MOO1 Opening Optimizer
• Diplomacy: 7 AI Models
• D&D Spell Damage Model
• Newcomb’s Paradox
• The Fake Science of Economics (Yarvin x McCormack)
• Master of Magic — Economic Analysis
• MoM Tier System and MIRR
• MIRR — 4X Strategy as Capital Allocation — the cross-game 4X capital-allocation thesis (MoO + MoM)
• Capability Without Leverage — a paid capability is worthless without downstream leverage; cross-game (Monopoly/MoM/Catan)
• Randomness as Termination (N≥3) — randomness is the kill switch on the gang-up equilibrium; calibration = design quality
• The Dominance-Frontier Lens — map cost vs effect, frontier curve, asymmetric counters; cross-domain (games + markets)
• NA1 — Game-Design Crucible — index tying the Nobunaga RE repo to the theses it sparked
• The Hollow Opponent — NA1’s AI = argmin-weakest + threshold; perceived depth lived in ignorance, migrated to the board once decoded
• The Dead-Verb Test — a verb is viable only if some state makes it the uniquely best response; the empty-home-state anti-pattern (Pact’s cost/value anti-correlate + it’s redundant with the dominant policy)
• Civilization Revolution — The City-Builder That Plays as a Rush — MIRR hurdle-rate dominance (wide > tall); a thin tech spine with backfill that breaks the DAG; N=2 map-RNG as randomness in the wrong place (instant death, not agency); the viable-strategy frontier collapses
• CoM Anti-Colossus Counter Analysis
• CoM Spell Counter-Graph
• Catan — 47,000 Games of Empirical Findings — empirical confirmation of bilateral trade valuation; Monopoly card as timing-conditional state-shock
• Catan 50-Game Validation — small-N empirical run; board-conditional theory refined
• Game Theory as Normative, Not Descriptive — meta-level critique: famous solution concepts work only by importing extra-game assumptions; norms guide behavior more than self-interest
• Force Doctrine — The Theory — force selects each round, structure selects who holds the force across rounds (selection over ideologies); fit → durable might, not might → right; the historicism guard
• BattleTech Simulator — Monte Carlo combat simulator to derive empirical BattleValue for BattleTech mechs and find bargains in the official BV2 system.
• BattleValue — A Universal Combat Comparison Metric — BV = sqrt(Attack × HP).
• CaptainMeme vs. 6 Cicero — The Human/AI Split in Press Diplomacy — An expert human diplomacy player plays full-press blitz Diplomacy against six Cicero AIs and narrates the game.
• Efficient Frontier Trade Theory — Using EPT frontier curves to solve the 3+ player trade valuation problem — reducing the permutation space through structural truths about development velocity and game horizon.
• Gunboat Diplomacy and Diplodocus — The Planner Without the Mouth — Meta AI’s Diplodocus plays Gunboat Diplomacy — the variant with no text communication — and beat every human in Meta’s 2022 Speedboat Tournament.
• Subgraph Trade Engine — Implementation Spec — Combining the build frontier with subgraph-driven trade search to create a complete Monopoly decision engine.
• The Coconut Island and the Manufactured Option Space — Two related debate moves used to force “exploitation under capitalism” or “structural conflict” conclusions.
• Battleship — 30 Billion Boards — best-response (guessing, exploitable prior → deep) vs minimax (placement, adaptive adversary → high-entropy equilibrium → flat); skill gradient tracks opponent-model exploitability
• Yahtzee — 259 Trillion → 405 Million — EV-max ≠ win-max: a state abstraction sound for expected points is provably insufficient for “am I ahead”; variance-seeking when behind pays only as the opponent’s score hardens, and reachability (support) binds before variance tuning
• CGP Grey — “Too Good for Politicians to Allow” (STV) — voting as mechanism design; Arrow + Gibbard–Satterthwaite as the impossibility floor; STV inherits RCV’s non-monotonicity + center-squeeze; system choice is irreducibly normative
• Sitch vs. Ryan Mullally — Birthright Citizenship & Natural Law — finite-horizon PD / default-defect as the formal Ring of Gyges; TFT loses pairwise but wins the tournament (individual vs population); God-as-infinite-iterator; EGT as the population-level answer to the brute
• D&D Monster Tournament — Exact Markov Chains — fixed policy ⇒ MDP collapses to a plain Markov chain (real tactical choice would make it a two-player stochastic game); hunts non-transitive cycles that would mean “same CR” is not a total order
• Aggregation vs. Sorting — median voter (1-D) vs McKelvey–Schofield chaos (≥2-D); increasing returns to coordination unifies Duverger’s law with the monopolization of legal orders (Nozick’s dominant protection agency in economic dress)
• Combinatorial vs Generative Design Space
• Risk — The Attrition Constant — a Risk battle solved exactly as an absorbing Markov chain; the 3-vs-2 dice cap fixes the engagement frontage regardless of stack size, so attrition is Lanchester-LINEAR and concentration of force buys nothing
• Hangman — Solving Both Sides — the minimax chooser is a hard concentration (-ING 100% at lengths 7–9), not max-entropy: the indifference set of an equilibrium is bounded by the substrate’s structure — only ~15 of 26 letters can be made competitive because Z can’t do S’s job
• Civilization Revolution — Project Hub