The quality of supernova data will dramatically increase in the next few years by new experiments that will add high-redshift supernova to the currently known ones. In order to use this new data to discriminate between different dark energy models, the statefinder diagnostic was suggested [1] and investigated by Alam et al.[3] in the light of the proposed SuperNova Acceleration Probe (SNAP) satellite. By making use of the same procedure presented by these authors, we compare their analyzes with ours, which shows a more realistic supernovae redshift distribution and do not assume that the intercept is known. We also analyzed the behavior of the statefinder pair {r,s} and the alternative pair {s,q} in the presence of offset errors.
