Left column: R_A/θ_E vs. R_A for the two first methods discussed in this work. Top panel: R_A/θ_E,I vs. R_A. Green circles depict the secure group candidates (Foëx et al. 2013), green diamonds those with 2.0′′ ≤ R_A ≤ 8.0′′. Green filled squares are groups with multi-component features or very elongated light contours (see Sect. 5). The blue dashed line depicts the mean R_A/θ_E,I, the red continuous line shows the one-to-one relation, blue squares are groups with giant arcs (see Sect. 4.1), and magenta triangles represent those groups with strong lensing models (see Sect. 4.3). The error bars were omitted for clarity. Middle panel: R_A/θ_E,II vs. R_A using the catalog constructed using the V_rms as proxy for the velocity dispersion measured in the observed lenses. Bottom panel: R_A/θ_E,II vs. R_A using the catalog built to match the shape of the observational redshift distribution of the lenses. Right column: θ_E vs. R_A. Black triangles show all the groups (regular or irregular) that satisfy 2.0′′ ≤ R_A ≤ 8.0′′. The black dashed line shows the fit to the data, with the 1σ-error depicted as a gray shaded region. As before, the red continuous line shows the one-to-one relation.