The purpose of our study was to explore relationships between transcranial Doppler (TCD) indices of cerebrovascular reactivity and those derived from intracranial pressure (ICP). Goals included: A) confirming previously described co-variance patterns of TCD/ICP indices, and B) describing thresholds for systolic flow index (Sx; correlation between systolic flow velocity [FVs] and cerebral perfusion pressure [CPP]) associated with outcome. In a retrospective cohort of traumatic brain injury (TBI) patients: with TCD and ICP monitoring, we calculated various continuous indices of cerebrovascular reactivity: A) ICP (pressure reactivity index [PRx]: correlation between ICP and mean arterial pressure [MAP]; PAx: correlation between pulse amplitude of ICP [AMP] and MAP; RAC: correlation between AMP and CPP) and B) TCD (mean flow index [Mx]: correlation between mean flow velocity [FVm] and CPP; Mx_a: correlation between FVm and MAP; Sx: correlation between FVs and CPP; Sx_a: correlation between FVs and MAP; Dx: correlation between diastolic flow velocity [FVd] and CPP; Dx_a: correlation between FVd and MAP). We assessed the relationships via various statistical techniques, including: principal component analysis, agglomerative hierarchal clustering, and k-means cluster analysis (KMCA). We performed sequential χ2 testing to define thresholds associated with outcome for Sx/Sx_a. Outcome was assessed at 6 months via dichotomized Glasgow Outcome Score (GOS): A) Favorable (GOS 4 or 5) versus Unfavorable (GOS 3 or less), B) Alive versus Dead. We analyzed 410 recordings in 347 patients. All analyses confirmed our previously described co-variance of Sx/Sx_a with ICP-derived indices. Sx displayed thresholds of -0.15 for unfavorable outcome (p < 0.0001) and -0.20 for mortality (p < 0.0001). Sx_a displayed thresholds of +0.05 (p = 0.019) and -0.10 (p = 0.0001) for alive/dead and favorable/unfavorable outcomes. TCD systolic indices are most closely associated with ICP indices. Sx and Sx_a likely provide better approximation of ICP indices, compared with Mx/Mx_a/Dx/Dx_a. Sx provides superior outcome prediction, versus Mx, with defined thresholds.