This is a follow-up from yesterday's post:

https://drupal.star.bnl.gov/STAR/blog/dmawxc/update-05152019-run-9-pp-checking-effect-using-detector-level-trigger-tracking-efficienc

I wanted to see what happens when I use the particle-level trigger's phi to calculate the detector-level tracks' delta-phi since there is no guarantee (at the moment) that the detector-level trigger is reconstructed from the particle-level trigger. Here are the results:

The dip at pT > 9 GeV/c is still there (which is to be expected), but it still has the weird peak structure at low pT... For reference, here's a comparison of the resultant tracking efficiency between the two methods of calculating delta-phi:

The case with delta-phi calculated wrt. the particle-level trigger is systematically higher, but still has the same shape... :/