Bodies of rock that are detached (recovered) from subducting oceanic plates, and exhumed to Earth’s surface, become invaluable records of the mechanical and chemical processing of rock along subduction interfaces. Exposures of interface rocks with high-pressure (HP) mineral assemblages provide insights into the nature of rock recovery, yet various interpretations concerning thermal gradients, recovery rates, and recovery depths arise when directly comparing the rock record with numerical simulations of subduction. Constraining recovery rates and depths from the rock record presents a major challenge because small sample sizes of HP rocks makes statistical inference weak. As an alternative approach, this study implements numerical simulations of oceanic-continental convergence and applies a classification algorithm to identify rock recovery. Over one million markers are classified from 64 simulations representing a large range of subduction zones. We find recovery P’s (depths) correlate strongly with convergence velocity and moderately with oceanic plate age, while PT gradients correlate strongly with oceanic plate age and upper-plate thickness. Recovery rates strongly correlate with upper-plate thickness, yet show no correlation with other boundary conditions. Likewise, PT distributions of recovered markers vary among numerical experiments and generally show poor overlap with the rock record. A significant gap in predicted marker recovery is found near 2 GPa and 550 ˚C, coinciding with the highest density of exhumed HP rocks. Implications for such a gap in marker recovery include numerical modeling uncertainties, petrologic uncertainties, selective sampling of exhumed HP rocks, or natural geodynamic factors not accounted for in numerical experiments.