We investigate the connectivity properties between different ocean stations in an idealized open ocean model of a western boundary current system separating two ocean gyres. We applied a Lagrangian framework to compute trajectories from various dynamical setups: a high-resolution (1/9°) 3D velocity field reproducing a large range of the ocean fine-scale (i.e. mesoscale plus part of the submesoscale) dynamics, or a filtered velocity field on a coarse-resolution (1°) grid, and one limited to the surface 2D velocities. As ocean connectivity has been assessed in the published literature using different definitions, in this work we compare different metrics such as the average values of transit time and arrival depth between specified sample stations as well as the probability density functions (PDFs) of transit times and betweenness for the different dynamical setups. Our results indicate that almost none of the PDFs show Gaussian behaviour. When the fine-scale dynamics are taken into account, the numerical particles move and connect pairs of stations faster (between 100 days to 300 days) than when it is absent. This is particularly true, along and near the jets separating the two gyres. Moreover, the connectivity is facilitated when 3D instead of 2D velocities are considered. Finally, our results suggest that western boundary currents are characterized by high betweenness centrality values, which confirms its key role in controlling the transfer of particles in the double-gyre configuration.