The idea was to use the same cuts that we have been using on the shower and neutron arm (B.ts.ps.e>400 to reject pions, time of flight, and coplanarity). Now that we have a y position in the neutron arm I was able to use that as well. There is a nice correlation between the scattering angle of the electron and neutron arm y (Figure 1). I was able to calculate a rough scattering angle from the shower y (Figure 2) which I can then use to select elastic events (very, very roughly) using the total shower energy. Tightly cutting on the neutron arm y and shower y (in the same manner we do the coplanarity cut) as well as tightening all the previous cuts, I determined that these select out a subset of elastic events fairly well.
I then plotted out the number of tracks we found from these events and
found that the tracking efficiency is roughly around 80% (figure 3).
Essentially, this tells me that the missing elastic events we have
doesn't appear to be from the tracking code.
Figure 1
Figure 2
Figure 3