The present work is a parametric study of the pressure pattern in a two-dimensional recess of a hybrid journal bearing (HJB). It is known that theoretical models of HJB are largely dependent on the recess pressure pattern especially for severe working conditions (high rotation speeds, shallow pockets, etc.). The difficulty is that the recess flow is dominated by the interaction of viscous and inertia forces and cannot be analyzed using a thin film model. The present analysis is based on the numerical resolution of the two-dimensional Navier-Stokes equations where only one recess is modeled (with the film lands and the supply region), the fluid being regarded as incompressible and isothermal. Both the laminar and the turbulent flow regimes are considered. The study is governed by two parameters, one related to the HJB operating conditions and the other related to the recess geometric characteristics. The first parameter is the ratio of the runner versus the supply Reynolds number, Rer/Res∈{0,1/4,1/2,1,4,8}. The supply Reynolds number is fixed at 100 for the laminar regime and at 5000 for the turbulent one. The second parameter is the ratio of the recess depth versus the film thickness. Six values of this ratio are considered, ranging from 4 (shallow recess) to 152 (deep recess). Detailed pressure patterns on the runner wall are presented in a systematic manner giving a clear insight of the flow effects intervening in the recess and of their mutual interaction. Some effects are explained by analyzing the recirculation zones inside the recess. It is also shown that for certain parameters turbulent flows have qualitatively similar effects as laminar ones but they can also have specific trends. In order to sustain this remark, the pressure variation at the recess downstream end is analyzed in the paper. Finally, the present results and specially the turbulent ones are intended to contribute to the understanding of viscous and inertia effects interactions in a recess flow and to represent a database in view of HJB theoretical modeling.