Capability Without Leverage

A capability you paid to acquire is worth zero — or negative — unless something downstream lets it actually shift the outcome. Face efficiency (EPT-per-dollar, BattleValue, trade-efficiency) is the trap; the right metric is impact conditional on use. Because you paid the acquisition cost, an unleveraged capability is worse than none — you’re down the cost and the opportunity. A cross-game invariant: Monopoly, Master of Magic, and Catan all independently coin it.

Links: BattleValue (BV→0 in a bad matchup is this, in combat units), Bilateral Trade Valuation (impact-share is its third-party / position layer), Frontier Trade Theory, Gaming

A thesis page (portable, source-independent). The worked numbers live in the specimens it cites; this page states the claim.

The thesis

Every acquisition buys a capability — a monopoly, a unit, a port, a tech. Capability is not the asset. Leverage is: the downstream mechanism by which the capability changes the result. A capability with no leverage is a sunk cost wearing the costume of progress — it looks like you climbed the efficiency curve, because face metrics reward it, while the leaderboard doesn’t move.

The diagnostic question is never “is this efficient?” but “if I hold this, what changes for whom?” If the honest answer is “nothing decisive,” the efficient-looking buy is a trap.

Two failure modes

A — Intrinsically low-impact. The capability works exactly as designed; it just doesn’t move the outcome.

Monopoly — the Brown trap. Browns are cheap and EPT-per-dollar-efficient, but generate small absolute rent, apply no positional pressure, drain no reserves, block no high-traffic paths. “A player who spends 3 random properties + cash to acquire Browns… looks like they moved up the EPT curve — but they spent their available trade capital on a monopoly that doesn’t translate into wins.” The fix is a metric: game-impact-share, not EPT-per-dollar. “Browns are EPT-efficient and impact-inefficient.”

B — Conditionally gated. The capability’s payoff depends on an enabling condition the opponent (or your own board) can negate. Remove the condition → value collapses to zero.

MoM — the hard counter to zero. “A unit might have BV 50 against normal units but BV 0 against anything with Weapon Immunity.” Effective value isn’t a scalar; it’s a matchup matrix, and the wrong cell is a cliff.
MoM — the Drake/invisibility play (sharpest, because the gate is removed live). Invisible Great Drakes whose entire edge was “the Colossus can’t shoot what it can’t see” — until Illusion Immunity (which doubles as see-invisible) negates the gate and “it was naked stat-on-stat.” The capability was real; the leverage was contingent, and the contingency failed.
Catan — the unleveraged 2:1 port. A port’s face value is “trade efficiency”; its capability value is “trade efficiency conditional on producing enough of the resource to use it.” Empirically: leveraged ports ≈ 25% baseline win rate, unleveraged ports 12.5% — half baseline. “Holders with no matching production are paying for an unused capability.” A 2:1 port replaces a productive vertex; if the matching resource is weak, you sacrificed leverage for the capability — net negative.

Why it’s worse than nothing (the cost asymmetry)

A free unused capability is merely neutral. A paid one is negative: you spent acquisition cost + the opportunity cost of the higher-impact thing you could have bought instead. That’s why “capability-with-no-leverage is worse than no capability” — the comparison isn’t to zero, it’s to the leveraged buy you forwent. Failure mode B is especially punishing because the capability appears high-value right up until the gate is contested.

The dual — leverage that can’t be negated

Invert the lens and you get a design/strategy target: the capability whose enabling condition the opponent cannot remove. Being the Weapon-Immunity unit (rather than the one it zeroes) is pure leverage; a monopoly on a high-traffic path is leverage that doesn’t depend on a contingency. The strongest acquisitions aren’t the most efficient — they’re the ones whose impact is unconditional. This is the same axis as MoM’s “the broken units are the ones immune to the most counters.”

The metric correction (one pattern, three names)

Game	Face metric (the trap)	Capability/leverage metric (the truth)
Monopoly	EPT-per-dollar	game-impact-share
Master of Magic	BattleValue	effective BV in the actual matchup
Catan	trade efficiency	trade efficiency conditional on matching production

In every case the face metric prices the capability in isolation; the correct metric prices it conditional on the leverage that lets it pay back.

Evidence ledger (specimens)

Monopoly — Frontier Trade Theory (§”The Brown trap”). Contributes: failure mode A in its cleanest form, the EPT-per-dollar→impact-share metric correction, and the “18× more EPT-efficient yet impact-inefficient” raw number.
Master of Magic — Economic Analysis (§”The Hard Counter Problem”). Contributes: failure mode B as a cliff (BV 50 → BV 0 vs Weapon Immunity); the matchup-matrix-not-scalar correction; the dual (immunity = un-negatable leverage).
MoM — Anti-Colossus study. Contributes: the gate-removed-live case (Drakes’ invisibility negated by Illusion Immunity → “naked stat-on-stat”) — the most vivid demonstration that the leverage, not the capability, was the asset.
Catan — 50-Game Validation (§port-leverage). Contributes: the empirical number (unleveraged port = half baseline win rate) and the face-value-vs-capability-value framing the whole thesis borrows.

The information-design half of feedback_capability_value_and_information_design (hidden-trackable = soft flaw, hidden-random = legitimate mechanic) is a separate lens that shares the memory note — a distinct promotion, not this page.
The dominance-frontier analysis lens (map cost/effect edges, find the frontier, flag asymmetric counters) is the methodology that finds these traps; candidate sibling hub.