Motivation:

The <pT> is larger in very low multiplicity event, which is different than other publication and PYTHIA simulation.  

Suspect the vertex finding algorithm favors the high-pT track which may have a large effect in low-multiplicity event reconstruction. 

Here is the test to see whethter it is the case

Test:

1. MB data <pT> vs multiplicity pT range dependence  (tracks required TOF matching)

If the vertex finding only used the track with pT> 0.2 GeV or not, it may give different results for including tracks with pT<0.2 GeV/c or not.

The different can not explain the rise in low multiplicity  

2. PYTHIA simulation:
Only pion+, pion- are used. The tracking efficiency using embedding data is applied to the PYTHIA (it affects both
and multiplicity).
X-axis is the number of pion passing the cuts and in |η|<1

VPD requirement is simulated by requiring at least one pion has |η|>4.  

Vertex finding is stimulated by requiring at least one pion has pT > 0.2 GeV/c. However, all pion with pT > 0.15 GeV/c for the event passing the cuts will be used in the calculations.