To get an idea of the real statistics and problems, it is useful to bin the data in terms of the triggers. Following Pibero's recommendations, I have implemented the following requirements:

• vertex rank > 0
• |z_vertex| < 90 cm
• R_T < 0.98

I have then separated the histograms into the following bins:

• JP0 (320600) || JP1 (320601) || JP2*L2JetHigh (320602) || AJP (320603)
• BHT1 (320501) || BHT2 (320503) || BHT0*VPDMB (320504)
• BBCMB (320103) || VPDMB (320001)

 Figure 1

Figure 2

Figure 3

I note that the plots in Figs. 1 and 2 are identical. I'm not sure why that is. It could be a bug in my code. It could be that I picked the lucky run. Obviously, I need to figure that out. There are notable differences when comparing to Fig. 3. So, this gives me some confidence that I at least changed the trigger cut for the third bin.