Speed Per Precise Result (SPPR): A Deterministic Performance Metric for Causal Decision Systems

Description (Abstract) Performance evaluation of intelligent systems is currently dominated by metrics that measure computational volume, latency, or statistical accuracy. While appropriate for probabilistic and data-driven systems, these metrics fail to characterize systems whose primary function is deterministic causal decision-making under constraints. This publication formally introduces Speed Per Precise Result (SPPR), a performance metric defined as the elapsed time required to produce a correct, deterministic decision under fixed premises. SPPR explicitly ignores raw throughput, speculative parallelism, and probabilistic approximation, focusing instead on how efficiently computation is converted into valid causal resolution. This document is intentionally published as a metric-definition and prior-art disclosure, not as an empirical study or implementation report. It defines the SPPR metric precisely, delimits its scope, and establishes authorship and priority through public, timestamped disclosure. The publication does not propose benchmarks, datasets, or experimental results.