The Orthogonality Frontier: Reliability-Adaptive Causal Decisions with Many Treatments

Large treatment libraries make dense prediction scalable while causal verification remains scarce. Even with a preliminary predictor and a doubly robust score in hand, a decision maker must determine how strongly causal discrepancies should revise the final normalized target. To formalize this choice, we introduce reliability-adaptive target orthogonalization (RATO), which represents the correction as an operator on the $K-1$ treatment-contrast quotient and distinguishes two criteria for choosing it: first-order orthogonality and local target risk. Full correction, $A=I$, uniquely removes first-order preliminary-predictor sensitivity within the target family. In contrast, minimum local target risk selects $A^{\star }=B(B+S{)}^{-1}$ under a zero cross-second-moment condition, where $B$ and $S$ capture both variance and persistent local bias. Under these local-risk conditions, whenever $A^{\star }$ is not a scalar multiple of $I$, the directional oracle has strictly lower local risk than every scalar correction. Because the required total error moments are not generically identified from routine deployment data, independent validation of the scalar family provides the default deployable route. Controlled experiments examine the frontier against exact nonlinear target error. In a retrospective participant-randomized study with 72 actions per context, increasing correction-and-tuning budget with preliminary fitting and test data fixed moves mean selected correction from $0.59$ at $5\%$ to $0.94$ at $80\%$. A second, weak-signal study shows stronger correction without higher held-out value. The frontier allocates available causal information; it does not create treatment signal.