As in a previous email thread, from myself

what I can think has been changed in terms of TOF reconstruction happened when we were struggling with the TOF SP efficiency:
basically as in https://micewww.pp.rl.ac.uk/projects/computing-software/wiki/Computing-Software-Handover-Document#TOF

(there is much more from Viktor https://micewww.pp.rl.ac.uk/projects/maus/wiki/Some_TOF_Space_Point_Performance_Plots)

was suggested to move the TOF_makeSpacePointCut (and TOF_findTriggerPixelCut) in ConfigurationDefaults.py
from 0.5 ns to 3ns (and this is in MAUS since 3.2.0), to accept as SP with a larger slab dt.

This might have been correct for the pion beam but not for the TOF calibration for the muon runs.

A quick reconstruction of run 7666 moving back the dt cut to 0.5ns shows a better agreement

JN's presentation at CM56: https://indico.cern.ch/event/882351/contributions/3717536/attachments/2003256/3344783/CM56.pdf