Selection is the same that for usual SF.
implementation is the same that for usual SF, but now (Data eff) → (FullSim eff) and (MC eff) → (FastSim eff).
Question: We have big difference between efficiencies in low pt bins. It is due to we have trigger matching for tag lepton for FullSim MC. (we had something similar with MC (without triggers) vs Data (with triggers) SF before)
My question is: is it correct to have trigger matching for FullSim MC (since we have this in our analysis), or it is better to have both FastSim and FullSim without trigger matching?
In the tails SF are in consistent with official https://twiki.cern.ch/twiki/bin/view/CMS/SUSLeptonSF#FullSim_FastSim_TTBar_MC_compari