In investment analysis, predictive power refers to the capacity of an indicator or analytical framework to provide meaningful advance information about the future direction or magnitude of market prices. An indicator with genuine predictive power does more than correlate with the future after the fact — it enables investment decisions that produce favorable outcomes more reliably than random chance.

See the primary definition at Predictive Power for a complete treatment of this concept, including the distinction between statistical and economic significance, in-sample versus out-of-sample testing, and the decay of predictive value after publication.

The practical standard for predictive power in systematic investing is demanding: a signal must generate consistent, statistically significant predictions out-of-sample, across multiple markets and time periods, with economic significance sufficient to produce returns that exceed costs and risks after implementation. Signals that meet this standard are rare and valuable — and they form the foundation of rigorous evidence-based investment processes.