Adding this here after getting comment in 2l2q AN from EGM POG contact.
EGamma does not recommend to use the same thresholds for impact parameter cuts for barrel and endcap. The reason is that the barrel and endcap regions have different geometries and material budgets, the barrel is also closer to the beamline. The spread in dz is usually larger in the endcaps. This is the general reasoning for the recommendation of different cuts. However, impact parameter cuts are ultimately analysis-driven. What matters is that the impact parameter criteria do not significantly distort spectra of discriminating variables, which is related to the ID MVA SFs (those are measured with prompt electrons). As you use already approved SFs that are frequently used in the HZZ group, I do not object to keeping the selection as is. However, I would advise to consider this point again for future iterations of HZZ analyses. You can also consult https://twiki.cern.ch/twiki/bin/view/CMS/CutBasedElectronIdentificationRun2#Offline_selection_criteria_for_V on this matter
Adding this here after getting comment in 2l2q AN from EGM POG contact.
EGamma does not recommend to use the same thresholds for impact parameter cuts for barrel and endcap. The reason is that the barrel and endcap regions have different geometries and material budgets, the barrel is also closer to the beamline. The spread in dz is usually larger in the endcaps. This is the general reasoning for the recommendation of different cuts. However, impact parameter cuts are ultimately analysis-driven. What matters is that the impact parameter criteria do not significantly distort spectra of discriminating variables, which is related to the ID MVA SFs (those are measured with prompt electrons). As you use already approved SFs that are frequently used in the HZZ group, I do not object to keeping the selection as is. However, I would advise to consider this point again for future iterations of HZZ analyses. You can also consult https://twiki.cern.ch/twiki/bin/view/CMS/CutBasedElectronIdentificationRun2#Offline_selection_criteria_for_V on this matter