The question
Why do systems with substantial resources, intelligence, equipment, authority, or formal power still fail to produce results that other people can depend on?
This program begins by separating capacity possessed from effect reliably produced. It examines the conversion between them: the routines, permissions, routes, information, maintenance, legitimacy, coordination, and time required to turn a nominal capability into an outcome.
Working grammar
The recurring structure is:
Reliable effect = capability × consistency × direction
This is not asserted as a universal equation. It is a disciplined prompt.
- Capability: What can the system produce under specified conditions?
- Consistency: How much can another actor plan around that capability?
- Direction: Is reliability strengthening, holding, or decaying?
- Constraint hierarchy: Which limitation binds first, and which constraints become active after it moves?
- Conversion cost: What must be spent—money, time, legitimacy, inventory, attention, or optionality—to produce the effect?
Research rules
- Define the unit of analysis before scoring it.
- Distinguish observed facts from causal inference.
- Name the conversion mechanism rather than treating possession as power.
- Identify the binding constraint and the next constraint likely to appear.
- State the time horizon.
- Compare at least one rival explanation.
- Name evidence that would change the judgment.
Current applications
- Constraint Realism applies the method to strategic action.
- Geopolitical Reliability Index explores whether strategic capacity is dependable enough for others to plan around.
- Throughput Reliability Index explores whether reported manufacturing performance is stable enough for production planning.
- Capacity to Reliable Effect tests the analogy between plants and states and identifies where it breaks.
Boundary
The program will fail if “constraint” becomes a word for everything, if multiplying three terms substitutes for causal explanation, or if the framework protects itself from contradictory evidence. Its value depends on making analysis more discriminating—not merely more unified.
Revision history
- v0.1 — August 7, 2026: Established the public program, common vocabulary, boundaries, and first four linked artifacts.